College Bowl Games

Game Analysis view matchup stats

South Alabama (-6) 32   Western Michigan  26 

Note: South Alabama QB Gio Lopez has been downgraded to OUT.

South Alabama quarterback Gio Lopez was injured late in the season finale. He’s considered a game-time decision, which is why I gave a predicted score with and without Lopez. Backup Bishop Davenport has thrown 55 passes and got a start against Ohio, but he was 1.0 yards per pass play worse than Lopez in that start (adjusted for Ohio’s pass defense). Lopez also adds value as a runner (540 yards on 69 runs) and that part of his game will likely be hampered if he does play. If he doesn’t run at all that would be a difference of 2.4 points and I’ll make the adjustment based on running half as often as normal, which would be an adjustment of 0.3 yards per rushing play for the team and 1.2 points. Davenport has shown no signs of being a running quarterback of the caliber of Lopez and overall he’s about 4 points worse.

South Alabama’s run game will be hurt by the absence of star frosh RB Fluff Bothwell, who has decided to transfer after running for 832 yards at 7.5 yards per rush. Kentrel Bullock is a decent back (701 yards at 5.2 ypr) and he should produce good numbers against a horrible Western Michigan defensive front that allowed 5.8 yards per rushing play to FBS teams that would combine to average just 4.7 yprp against an average defense). However, the absence of Bothwell is worth 2.2 points.

Western Michigan’s defense was horrible this season, allowing 6.5 yards per play to FBS teams that would combine to average only 5.1 yppl against an average defense (adjusted for facing the worst of the quarterbacks in games against Wisconsin, Marshall and Central Michigan). The Broncos will be without their best pass rusher, as DE Corey Walker has announced his intention to transfer. Walker had 5.5 of the team’s 20 sacks and also batted down 5 passes (3rd on the team in passes defended). I value Walker at 1.3 points.

Western Michigan rates at 0.4 yards per play worse than average offensively and veteran quarterback Hayden Wolff has had a low interception rate throughout his career. Wolff threw no interceptions in 7 of 12 games this season while completing 67% of his passes. He should have a solid game against a South Alabama defense that’s allowed 66% completions in FBS games.

Broncos’ TE Blake Bosma, who caught 37 of 43 passes thrown to him for 9.4 yards per target (8.4 YPT vs FBS teams), has hit the transfer portal but the other tight ends have combined for 12 catches on 14 targets at 9.9 YPT (all vs FBS), so I can’t assume that Bosma will be missed. I can’t find an update on receiving yards leader Kenneth Womack, who missed the season finale. Womack wasn’t significantly better than the other wide receivers in compensated yards per target and the pass game rating was the same in the 3 FBS games that he missed as in the games he played.

South Alabama’s defense was good by Sun Belt Conference standards and rated at 0.3 yppl worse than an average FBS defense. The absence of the Jaguars’ top-rated CB Jordan Scruggs (transfer portal) weakens the pass defense, and I value Scruggs at 1.0 points.

I see some value on the over if Lopez plays – especially if he’s able to run at his usual effectiveness, but with his status in doubt I have no leans on this game.

Game Analysis view matchup stats

Strong Opinion – Under (60.5) – Memphis (-5.5)  30   West Virginia  24

I was a bit surprised by the money flowing in on Memphis soon after the bowl lines were released but after finalizing the math I come up with the Tigers by 5.7 points. Where I see some value is on the Under.

The Memphis offense is not as good as the 35.2 points per game that they’ve averaged, as the 6.2 yards per play that the Tigers have averaged has come against teams that would allow 6.3 yppl to an average team. West Virginia’s defense is also not as bad at it seems, as the 6.6 yppl that the Mountaineers allowed to FBS teams was to offenses that would combine to average 6.5 yppl against an average defensive team. I rate that unit at 0.3 yppl worse than average without top LB Josiah Trotter, who has hit the transfer portal, but star DE TJ Jackson II has indicated that he’s excited to play and improve his draft stock. Jackson had 13.5 tackles for loss and will be a challenge for the Memphis offensive line. I project a modest 402 yards at 6.2 yppl for the Tigers in this game.

The West Virginia offense was a disappointment this season and not having WR Traylon Ray in recent weeks has been an issue. Ray averaged 9.1 yards per target this season but his absence has led to more targets for Justin Robinson, who has averaged 9.8 YPT. Garrett Greene should be fired up to end his career with a win and he’ll have to carry the load against a Memphis defense that is very stingy to opposing running backs. The Tigers couldn’t defend the option when they faced Navy in week 4 (536 yards at 10.1 yards per play allowed) but they’ve yielded just 3.8 yards per rushing play in their other 11 games (to teams that would average 4.8 yprp against an average defense), so running backs Jahiem White and CJ Donaldson will have their work cut out for them. Greene is a good runner, and he’ll have to be the one to move the chains with his legs and with his passing, which has been just average on a compensated yards per play basis and worse than average when you include his high interception rate (11 picks on 275 pass attempts). Memphis has been 0.5 yards per pass play worse than average against the pass so Greene should post decent numbers. I project 369 total yards at 5.8 yppl for the Mounties.

Both teams run their offense at a slower than average pace and the projected yards in this game, after adjusting for what looks like low winds, is barely above the national per game average. The national average for scoring is 53.2 points and I project 53.9 total points (and favor Memphis by 5.7 points). The total on this game is what a compensated points model would project (58.0) but my model projects fewer points than that because it’s unlikely that these teams will turn yards into points at the rate they did during the regular season.

West Virginia, offense and defense combined, averaged 5.47 points per redzone opportunity (5.40 on offense and 5.56 on defense) while Memphis averaged a combined 5.37 PPRZ (5.41 off, 5.30 def). Both teams should be around 5.0 PPRZ based on their statistical profiles, which would lower each team’s total points by about 3 points each (about 6.5 points total). A total of 58 points is correct if these teams continue to score and allow about 5.4 points per redzone opportunity, but that’s not likely, and now the total has gone up a few points.

The Under is a Strong Opinion at 59.5 points or more.

Game Analysis view matchup stats

James Madison (-7)  30   Western Kentucky  21

Western Kentucky players have traditionally played in the bowl game under coach Helton after announcing their intention to enter the transfer portal. Quarterback Caden Veltkamp has already stated that he’ll be playing in the bowl despite entering the transfer portal and I’ll assume the WR Easton Messer is out – although he may play –  which is a bit of blow given his 9.1 yards per target.  Only two of the many defensive starters that are entering the portal are not listed on the bowl depth chart. In that past the players on the Western Kentucky bowl depth chart have played in the bowl game and I’m going to assume that’s the case for most given that a few transfers, including the backup quarterback Finley are not on the depth chart, which is an indication that it has been updated. It’s certainly possible that some of the guys on the depth chart have since decided to not play but I’m going to assume that they will.

James Madison quarterback Alonza Barnett will not play due to a leg injury and I’ll assume that starting LT Jesse Ramil will not play given that he’s already committed to Mississippi State. Billy Atkins will start at quarterback and he was not good in his only other career starter – averaging just 2.9 yards on 42 pass plays with 4 interceptions against Marshall back in 2022. I’ll assume he’s gotten better in the past two years but he’s only thrown 4 passes since that start (completed 1 and had 1 intercepted) and will likely be a significant downgrade from Barnett in the pass game. Barnett was also valuable as a runner (689 yards on 101 runs) and Atkins has just 18 yards on 6 runs in his lone start.

James Madison’s offense was 0.3 yards per play worse than average this season with Barnett and I rate that unit at 0.9 yppl worse than average with Atkins at the helm, although that is subject to a lot of variance given that Atkins hasn’t played any meaningful snaps in over 2 years. Western Kentucky’s defense as 0.7 yppl worse than average for the season and I rate that unit at 1.0 yppl for this game assuming that the defenders on the bowl depth chart will play, as has been the case in the past with Western Kentucky under coach Helton.

Western Kentucky’s offense has 0.1 yppl worse than average with Veltkamp in the game and I rate the Hilltoppers’ attack at -0.3 yppl for this game with C Wibberly likely out (he’s committed to Kentucky) and WR Messer probably out. The JMU defense is 0.3 yppl better than average and they appear to be intact for this game.

I still see a bit of value on James Madison even after severely downgrading their offense. However, if Atkins is close to as bad as he was in his other career start in 2022 then Western Kentucky could win this game. Lots of variance here with still some uncertainty regarding the Hilltoppers’ transfers and how Atkins will perform.

Game Analysis view matchup stats

Nevada-Las Vegas (-2)  25   California  20

Cal quarterback Fernando Mendoza entered the transfer portal and is not with the team. That would have been fine if former North Texas and ULM star QB Chandler Rogers were healthy enough to play but Rogers was injured against SMU and former Ohio University backup CJ Harris and inexperienced freshman EJ Caminong are getting most of the reps in practice. There is no definitive word on Rogers’ status but I’m going to assume he won’t play. Harris averaged just 5.2 yards on 150 meaningful pass plays at Ohio and tallied 70 yards on 12 pass plays against SMU in relief of Rogers, but he is a good runner (6.6 yards per rushing play). Mendoza averaged 7.2 yards on his 47 runs this season but Harris would add more to the run game because he runs the ball more often. The wildcard is Caminong, who has more talent than Harris or Rogers but has no collegiate experience. Head coach Justin Wilcox has stated that he’s not afraid of having his freshman play in this game and there is a chance that he could start and be better than Harris or Rogers would be.

A certain blow to the Cal offense is the absence of top WR Nyziah Hunter, who is transferring and won’t play in this game. Hunter led the team in receiving yards and his 9.8 yards per target is much better than the other wide receivers, who combined for 6.7 YPT. I do expect to see more balls thrown to star TE Jack Endries, who caught 49 of 54 balls targeted at him for an average of 10.4 YPT. The running backs are also good receivers, as Ott and Thomas combined for 33 catches and 7.8 YPT, which is very good for running backs. I project Hunter to be worth 0.3 yards per pass play, which would amount to about a full point given that the Bears are likely to throw less than they did in the regular season – particularly if Harris starts at quarterback.

Cal only averaged 5.6 yards per play this season while facing teams that would allow 5.8 yppl to an average offense and UNLV’s defense, with all starters expected to play in this game (CB Grimes is in the transfer portal but has stated that he will play), was 0.5 yppl better than average this season (5.3 ypp allowed to teams that would combine to average 5.8 yppl against an average defense). Cal’s offense is certainly worse without Hunter, and they should be worse at quarterback without Mendoza.

UNLV’s offense was 0.6 yppl better than average in their last 10 games with Hajj-Malik Williams at quarterback (6.6 yppl against teams that would allow 6.0 yppl) but the Rebels are at a slight disadvantage against a Cal stop unit that’s been 0.8 yppl better than average this season while defending the run and the pass equally well. There was a concern about NFL caliber WR Ricky White opting out to prepare for the NFL combine but there’s been no word about that being the case, so I’ll assume he’ll play in an effort to boost his draft status.

My math would have favored UNLV by 3.7 points with Mendoza at quarterback for Cal and they will likely be worse unless Caminong starts and plays up to his talent rating. Both teams have new play-callers as Cal’s OC was fired and UNLV’s head coach and OC have left the program, so I’ll assume that’s a wash. I’m going to assume that Cal will run the ball more, as Harris is a running quarterback and the staff would likely also choose to run more if they give Caminong the nod to play.

This game is being played in a closed stadium, which adds to scoring, but UNLV played 5 of their 12 games against FBS teams in a dome in Vegas, so the impact of the dome isn’t as significant as it would be if both teams played their home games outdoors. I also factored in a slower pace of play for Cal based on the Bears likely running the ball more. They played at a much slower pace without Mendoza against SMU (about 7 total plays slower) but I only dropped the total plays projection by 3.0 (before adding more for playing in a dome, which generally leads to more passing).

The math favors UNLV by 4.9 points if I assume Cal’s quarterback is 50% Harris and 50% Caminong and I get 44.9 total points after adjusting for the perfect conditions.

Game Analysis view matchup stats

Lean – Sam Houston State (+6)  23   Georgia Southern  24

This line is up to Georgia Southern by 6 points (instead of the 2 points it would have been under normal circumstances) because of the high number of Sam Houston State players entering the transfer portal after head coach KC Keeler took the Temple job. However, the rumors are that most of those players are really excited about playing in Sam Houston’s first ever bowl game. The Bowl depth chart shows a couple of players that entered the portal not listed (backup QB Bauer and a reserve DE) but the rest of those intending to transfer are listed on the depth chart, which obviously has been updated given the absence of a couple of players. So, I’ll assume they’re playing if they’re on the depth chart.

Sam Houston’s offense is run-heavy as the Bearkats averaged over 200 ground yards per game at 5.1 yards per rushing play and only 149 pass play yards per game at 5.2 yards per pass play (5.3 yppp with Watson at QB). The Sam Houston attack should work at least as good as normal against a Georgia Southern defense that allowed 6.0 yprp this season. That number was skewed by the 375 yards at 11.7 yprp that they gave up to Boise State in their opener and I’ve taken that game out since it was an obvious outlier. However, the Eagles were still 0.4 yprp worse than average and they’re 0.7 yards per pass play worse than average defensively overall. The Eagles’ defense is better than the Bearkats’ offense, but SHSU should score around their season average of 23 points given that Georgia Southern’s defense is almost identical in yprp and yppp to the average defense that Sam Houston faced this season.

Georgia Southern’s offense was 0.5 yards per play worse than average in the regular season (5.4 yppl against teams that would allow 5.9 yppl to an average team) and they’ll likely struggle against a Sam Houston defense that was 0.5 yppl better than average, yielding just 4.8 yppl and 20.0 points per game to teams that would average 5.3 yppl against an average team. I rate that unit at 0.4 yppl better than average for this season if all the players on the depth chart play and project just 322 yards at 5.0 yppl for the Eagles even with perfect dome conditions factored in.

Sam Houston rates as the better team overall from the line of scrimmage but the Bearkats easily had the worse special teams units in the nation this season and overall the math favors Georgia Southern by 2.7 points. Once again, that’s assuming the players on the updated bowl depth chart will play. There is obviously a chance that some of them won’t play but Sam Houston applies to my best Bowl situation – a 77-15-1 ATS underdog angle. They also apply to a 70-27-2 ATS bowl underdog with a good defense situation and teams that apply to both of those are 15-0 ATS since 1980. Sam Houston was also just 3-9 last season and teams that lost 9 or more games the previous season tend to be very appreciative and focused for their bowl game. The Bearkats apply to a 56-16 ATS bowl situation that plays on teams that lost 9 or more games the previous season and the pride brought on by making the schools first ever bowl is likely the thing that will lead to some or all of the Sam Houston transfers playing in this game – much like the situation with James Madison last season when all of their transfers played in that school’s first ever bowl game.

If I knew for sure that Sam Houston’s transfers were all playing I’d have made them a Best Bet or Strong Opinion in this game. The situation alone is worth a Lean even if the line of Georgia Southern by 6 turns out to be the true line.

Game Analysis view matchup stats

Note: The line was -4 points when I released this lean earlier this week and I wouldn’t play it now at -6.5.

Lean – Ohio (-4)  33   Jacksonville State  25

Jacksonville State averaged 35.7 points per game with a run-oriented attack that averaged 440 yards at 6.4 yards per play from week 2 on after Tyler Huff took over at quarterback. Huff is a decent passer but his value is in his running, as he tallied 1462 yards on 199 runs. The Gamecocks were just 0.1 yppl better than average offensively with Huff after accounting for defenses faced and Ohio’s defense is better than nearly all the teams that Huff faced and in the same range as Louisiana Tech and Sam Houston State, who held the Jax State to a combined 4.4 yppl. The Bobcats defense rates as average on a national scale but their run defense is the strength of the unit (4.3 yprp allowed), which is key in defending the Gamecocks.

Ohio’s offense is similar, as the Bobcats have a dual-threat quarterback in Parker Navarro, who is a better passer than Huff and also topped 1,000 yards on the ground at 7.8 yards per run, which is actually better than Huff’s 7.3 yprp. The Bobcats’ attack is average on a national scale but they have the advantage in this game. Jacksonville State’s defense was 0.4 yppl worse than average in the regular season and they have a few key players that won’t be playing in this game.

Jacksonville State has 10 players, including 6 starters, that have entered the transfer portal but 5 of those 10 players are listed on the bowl depth chart, which is a good indication that they will play in this game. The offense appears to be intact, as top WR Cam Vaughn is listed on the depth chart and the only offensive transfer that isn’t is a backup offensive lineman. On defense, the transfers not listed on the depth chart are starting edge rusher/LB Reginald Hughes (sack leader), starting safety Zechariah Poyser (#4 tackler and 11 passes defended), backup S Tee Denson, and CB Jabari Mack, who lost his starting spot 6 games ago but led the team with 13 passes defended. Ohio, by the way, is the only team to have not had any players enter the transfer portal.

My math model would favor Ohio by 4.0 points with a total of 53.7 points before accounting for the transfers for Jacksonville State that are not on the depth chart. Losing the team’s leader in sacks (and 2nd in totals tackles for loss) along with two defensive backs that were the top two on the team in passes defended is a significant blow to the defense, which I project at 0.5 yards per play, which is 3.6 points to the side and 3.0 points to the total.

With the transfers factored in, I get Ohio by 7.6 points and 56.7 total points and it’s possible that some of the Jacksonville State players on the depth chart may not play too. Both head coaches have taken other jobs and the offensive coordinators for each team have been named as interim head coach.

Game Analysis view matchup stats

Note: This Best Bet was released when the line was 13.5 points. The number has come down to 10 and Tulane is now just a Lean.

2-Star Best Bet – **Tulane (+13.5)  23   Florida  28

My math model would have favored Florida by only 2.9 points in this game if both teams fielded the squads that they had at the end of the regular season. However, Tulane will be without star quarterback Darian Mensah, who had a great season, and two key defensive players are out for the Green Wave. However, it seems that the market has over adjusted for those losses.

Former Oregon backup quarterback Ty Thompson will get the start for Tulane, and he was a top-10 rated quarterback coming out of high school in 2021. Thompson was expected to be the starter, but Mensah beat him out in the summer and that decision certainly seems to have the right one given that Mensah averaged 8.3 yards per pass play against FBS opponents that would allow 6.5 yppp to an average quarterback. I’ll assume that Thompson will not be that good (although he has the talent to be better), but head coach Jon Sumrall and his OC Joe Craddock had much less talented quarterbacks at Troy that averaged 7.1 yppp in 2022 and 2023 and I’d expect Thompson to be at least that good in this system given his talent level. To be on the conservative side, I’ll assume Thompson would have averaged at 7.1 yppp this season had he been the starter, which is about a 3-point difference from Mensah.

I rate Tulane’s offense at 0.3 yppl better than average if Thompson plays at the level of the Troy quarterbacks the last 2 seasons under the Tulane coach/OC – although he’s much more talented and is likely to be better. Florida’s defense is 0.7 yppl better than average, so Tulane should move the ball at a reasonable level (I project 358 yards at 5.3 yppl).

Tulane will be missing two contributors from the defensive line, as Matthew Fobbs-White and Parker Peterson have already committed to other schools. Potential opt-out S Caleb Ransaw is on the updated bowl depth chart, so I’ll assume he’ll play. The impact of Fobbs-White and Peterson, who combined for 6.5 sacks, is just 1.2 points given the low number of plays that Florida is expected to run in this game. The Green Wave were 0.2 yppl better than average defensively this season and I rate that unit as average without the two defensive linemen that have transferred.

Florida’s offense is more boom or bust with DJ Lagway at quarterback than it was when Graham Mertz (77% completions) was getting most of the snaps. Overall, the two quarterbacks are the same on a compensated yards per pass play basis but Lagway is more interception prone (7 picks on 157 passes while Mertz threw just 2 interceptions on 94 passes). Florida rates at 1.1 yppl better than average offensively but they were also -14.2 in play differential this season, which was due to the big plays and 3-and-outs that come with Lagway at quarterback. I project a -12 play differential for Florida in this game and Tulane runs their offense at a very slow pace, which leads a projection of just 53 plays from scrimmage for the Gators (excluding kneel downs). They are projected to average 6.9 yppl, but that would lead to just 368 total yards.

Overall, the math favors Florida by just 7.1 points even with a point added for playing in Tampa. The line on this game opened at 9.5 points when it was already known that Mensah was going to transfer. The game was then bet up to 14 before coming back down to 13.5 points even though Mensah’s absence was already factored into the spread. Even if 9.5 was the fair line, which is about as high as I could make it, then Tulane would have a 58.9% chance of covering at +13.5 points and would still be a 54.7% bet at +11.

In addition to the line value, Florida applies to a 12-62-1 ATS minor bowl game big favorite situation and the Gators apply to a 17-59-2 ATS bow situation based on their 3 game straight up and spread win streak. The record is 0-3 ATS when those two situations apply to the same team.

Tulane is a 2-Star Best Bet at +13 or more and 1-Star down to +11.

Game Analysis view matchup stats

1-Star Best Bet – Notre Dame Team Total Under (29)

Lean – Indiana (+7 or more)  21   NOTRE DAME  24

Lean – Under (52)

A lot has been made of Indiana playing a weaker schedule than a typical Big  10 team and that they may not deserve to be in the playoffs. I can understand the point that the Hoosiers don’t have a win against a great team, but they only played one great team, and I think Ohio State is arguably the best team in the nation. So, most, if not all, of the teams in the playoffs playing Indiana’s schedule would have also lost at Ohio State and not all of them would have beaten the rest of the teams on Indiana’s schedule – and only a few probably would have beaten the other FBS teams that Indiana faced by an average score of 42.8 to 13.5.

Indiana is deserving simply because of how dominant they were in most of their games, which is more than can be said of Miami-Florida (lost to Georgia Tech and Syracuse) and Alabama, who lost to Vandy and Oklahoma. Alabama may have better wins than Indiana, but the Crimson Tide also have much worse losses and the Hoosiers mostly dominated teams at the level of Vandy and Oklahoma, as they beat better than average teams UCLA, Nebraska, Washington, and Michigan by an average margin of 24.3 points.

Despite all of the evidence that Indiana is a top-10 team, their offensive performance against Ohio State’s elite defense is a concern – particularly in this game against another elite defense. Indiana’s offense averaged 6.7 yards per play against FBS opponents that would combine to allow 5.5 yppl to an average attack but they managed just 3.1 yppl against the Buckeyes. Obviously, that doesn’t mean that they are not capable of moving the ball on a good defense, as drawing conclusions on a sample of one game is foolish. However, there is some merit to Indiana not being relatively as good when facing a better defensive team.

The regression analysis using the level of the opposing defense (adjusted by site) to predict Indiana’s yards per play in the 10 FBS games in which QB Kurtis Rourke played has a slope of 1.51, which means that they were 0.51 yards per play better, on a relatively basis, for every yard worse their opponent’s defense was (and visa versa). That is a better way to take into account their struggles against Ohio State’s defense and the Hoosiers also relatively struggled against Michigan’s solid defense – although they played very well against a couple of other better than average defensive teams. Using Notre Dame’s defensive level (adjusted for playing at home) would project Indiana to average 5.12 yppl in this game, which is 0.23 yppl less than my model would have predicted without using the regression equation. I will go with the 5.12 yppl prediction to give more weight to that Ohio State performance.

Notre Dame’s defense has been 1.2 yppl better than average this season but they struggled some against the two best offensive teams that they faced – giving up 434 yards at 5.8 yppl at home to Louisville in week 5 and yielding 557 yards at 7.2 yppl to USC a couple of weeks ago in a game in which they could have lost if not for a pair of 4th quarter endzone to endzone pick-sixes that turned likely USC points into 14 points for the Irish. Overall, however, Notre Dame’s defense was relatively as good against better than average offensive teams as against bad offensive teams.

Notre Dame’s pass defense hasn’t been as good in 6 games since Preseason All-American CB Benjamin Morrison was injured but part of that can be explained by the diminished pass run without DT Howard Cross III, who was second on the team in sacks despite missing the last 3 games. Morrison is still out but Cross is expected back for this game, and I don’t think any adjustment is needed for Morrison with Cross back.

Indiana can still compete in this game even if their offense struggles, as the Hoosiers’ defense is legit. Indiana’s defensive rating is actually better than Notre Dame’s defensive rating, as the Hoosiers have been 1.4 yppl better than average defensively, yielding just 15.7 points per game and 4.4 yppl to FBS teams that would combine to average 5.8 yppl against an average defense (which is only 0.1 yppl worse than the average offense that the Irish faced). That unit was just as good against better offensive teams, as they gave up just 5.7 yppl in that loss to the Buckeyes, which is 1.7 yppl better than an average defense would perform at Ohio State. The 38 points the Buckeyes scored in that game is a fluke given the 316 total yards they were held to by the Hoosiers.

Indiana is very good defending the pass (5.0 yppp allowed to quarterbacks that would average 6.5 yppp against an average defense) but their strength is defending the run, which they did better than any team other than Oklahoma this season. The Hoosiers gave up just 93 rush yards at 3.5 yards per rushing play to teams that would combine to average 5.0 yprp against an average team. Defending the run is a prerequisite for beating the Irish, who average more ground yards (228 per game) than air yards (189) this season. Riley Leonard is not a particularly good passer, as he settles for mostly short passes and was only 0.3 yards per pass play better than average, as he averaged 6.6 yppp against teams that would allow 6.3 yppp to an average quarterback.

Indiana isn’t the only offense that was relatively worse against better defensive teams, as the slope of the equation to predict Notre Dame’s offensive yards per play as a function of the opposing defensive rating (adjusted for site) is 1.65, which is even more extreme than the slope of the equation for the Indiana offense. Notre Dame faced 7 better than average defensive teams (Texas A&M, N Illinois, Miami-Ohio, Louisville, Georgia Tech, Florida State, and USC) and they were only 0.7 yppl better than average in those 7 games (6.1 yppl against teams that would allow 5.4 yppl), which is 0.4 yppl worse than their overall offensive rating of +1.1 yppl (6.7 yppl against teams that would allow 5.6 yppl), which was skewed upwards by the 10.1 yppl they gained against Army.

An argument can be made that two of those games against good defensive teams were the first two games of the season (Texas A&M and Northern Illinois) when new QB Leonard was getting his bearings. However, the Irish attack was relatively worse against better than average defensive teams even if you excluded those first 2 games (+0.9 yppl rating in the other 5 vs better than average defensive teams). A rating of +0.9 yppl is also what you get when you adjust for the Army game outlier by taking that game and their worst offensive game (vs N. Illinois) out – and that’s the level that I expect from Notre Dame’s offense for this game, which puts them at just 5.3 yppl against Indiana’s elite defense.

I don’t see either offense consistently moving the ball in this game and I see value on both Indiana and the Under. However, that value stems from Indiana’s underrated defense and I think the best way to play that edge is the bet the Notre Dame Team Total Under.

Notre Dame’s Team Total Under 28 points or more is a 1-Star Best Bet. The alternate play, if you don’t have Team Totals, is 1-Star on the game Under 52 points (at 51 or more).

Game Analysis view matchup stats

PENN STATE (-8.5)  30   Southern Methodist  23

SMU managed to lose to Clemson on a last-second field goal in the ACC Championship game, but the Mustangs outplayed the Tigers 458 yards at 5.3 yards per play to 319 yards at 4.4 yppl and certainly deserve to be playing this game.

SMU’s offense went from good to great when Kevin Jennings took over as the starting quarterback in week 4 and that undeserved loss to Clemson is their only blemish with Jennings at quarterback. The Mustangs have a big play aerial attack, as evidenced by Jenning’s 13.4 yards per completion but Jennings has also completed 66% of his passes and the Mustangs have averaged 6.7 yards per play in his 10 starts (against teams that would allow 5.3 yppl to an average team).

Penn State’s defense has been 1.1 yppl better than average, yielding 5.0 yppl to FBS  teams that would combine to average 6.1 yppl against an average defensive team. The only other elite defense that Jennings has faced is that of Duke, who rates just behind the Nittany Lions’ defense in compensated yards per play allowed, and SMU scored 28 points on 469 yards at 6.3 yppl in that game – so the Mustangs are capable of moving the ball against a good defense.

Penn State’s offense has been 1.4 yppl better than average this season (6.7 yppl against teams that would allow 5.3 yppl to an average team) and I rate them slightly better after accounting for the one game that star RB Nicholas Singleton (6.4 ypr) missed. SMU’s defense rates among the best in the nation at defending the run, having yielded just 3.9 yards per rushing play (to teams that would average 5.2 yprp against an average team) and the Mustangs also have a good pass defense (5.5 yppp allowed to quarterbacks that would average 6.2 yppp).

While the compensated yards per play numbers are close, the Nittany Lions have an overall advantage from the line of scrimmage because they’re expected to run more plays, as SMU’s big plays this season have resulted in fewer plays from scrimmage. SMU does have an edge in special teams and overall the math favors Penn State by 5.8 points at home. The cold weather expected (temps below 30 degrees) could have a negative impact on the visitors from Texas that are not used to playing in those conditions, so I’ve added an extra point of home field advantage to my projection.

Game Analysis view matchup stats

TEXAS (-13.5)  32   Clemson  14

Lean – Clemson Team Total Under (18.5)

Texas has been inconsistent offensively, but the Longhorns defense will dominate a Clemson offense that has struggled against better than average defensive teams again this season (now a 4-year pattern).

The Clemson offense was 0.7 yards per play better than average for the season (6.2 yppl against teams that would allow 5.5 yppl to an average attack) but the Tigers continue to struggle against better than average defensive teams after dominating worse than average defensive teams earlier this season with superior talent. In 5 FBS games against average or worse defensive teams (all between week 2 and week 8) the Tigers averaged 8.2 yppl and were 2.0 yppl better than average relative to what those teams would allow to an average FBS offense. The thought was that their horrible opening day performance against a good Georgia defense was an anomaly but that has proven to not be the case, as the Tigers started to struggle offensively when facing good defensive teams in their most recent 5 games – averaging just 4.9 yppl against Louisville, Virginia Tech, Pitt, S Carolina, and SMU – teams that would allow 5.0 yppl to an average offense. In 7 games this season against better than average defensive teams (Georgia, Florida State are the others) Clemson has averaged only 5.1 yppl, which is just 0.1 yppl better than the 5.0 yppl that those teams would allow to an average FBS offense. The 34 points they scored in the ACC Championship game was a mirage, as the Tigers tallied just 319 yards at 4.4 yppl and were outplayed by 12.6 points from the line of scrimmage in that lucky win.

Clemson is a bully and they’re about to get bullied right back by a Texas defense that is the best in the nation. The Longhorns have yielded just 12.5 points on 252 total yards per game at 4.1 yppl to teams that would combine to average 6.0 yppl against an average defense. Clemson’s offense, given how mediocre they’ve been in 7 games against better than average defensive teams, is no better than the average offense that Texas has faced this season, and I don’t see the Tigers getting more than 14 points in this game. Only Georgia and Vanderbilt have managed to score more than 17 points against the Longhorns and Vandy surpassed 17 points with just 46 seconds remaining in the game against a prevent defense and had just 200 total yards for the game before that late scoring drive. Clemson will struggle to move the ball consistently against the nation’s best defense.

The Texas offense hasn’t been as good as expected this season, as they’ve been just 0.9 yppl better than average with QB Quinn Ewers in the game after being 1.4 yppl better than average last season. The Longhorns have averaged 6.1 yppl with Ewers behind center while facing teams that would allow 5.2 yppl to an average attack (weighted by how many snaps he had against each defense) and Clemson’s defense (playing on the road) rates the same as the average defense that Ewers faced this season. The Tigers defend the pass well, but they’ve been mediocre against the run (5.3 yprp allowed) and the Longhorns should be able to establish their run game (I project 6.1 yprp and 6.0 yppp for Texas).

Outland Trophy winner LT Kelvin Banks Jr missed most of the week 14 game and didn’t play in the SEC Championship game due to a sprained ankle. The Longhorns gave up 6 sacks without their All-American protecting Quin Ewers’ blindside, but my guess is that he’ll play. It’s been 3 full weeks since Banks sustained the injury and even a high ankle sprain would likely be healed enough for him to play – although he may not be 100%.

The math model would favor Texas by 14.4 points if I weighed all of Clemson’s offensive performances the same but the evidence that they are significantly worse against better than average defensive teams is strong and I’m willing to bet that the Tigers’ offense is no better than average when not facing bad defensive teams. An average offensive team is highly unlikely to score more than 17 points against this Texas defense.

Game Analysis view matchup stats

OHIO STATE (-7)  28   Tennessee  19

Ohio State is coming off an inexcusable loss to Michigan in which they appeared to be playing not to lose than playing to win. I expect the Buckeyes to perform better in this game but Tennessee is capable of an upset.

The Volunteers have an elite defense that rates at 1.4 yards per play better than average, which is the same rating as the Ohio State offense. The Buckeyes struggled offensively against Michigan (4.3 yppl) but in 4 games against elite defensive teams (Oregon, Penn St, Indiana and Michigan) they were 1.3 yppl better than average so I don’t see a pattern of Ohio State’s offense playing worse against good defensive teams. I project 5.8 yppl for the Buckeyes in this game.

Ohio State’s defense is a bit better than the Vols’ defense, rating at 1.6 yppl better than average and they have a significant advantage over a Tennessee attack that I rate at just 0.4 yppl better than average (I excluded their game against FCS team Chattanooga and their game against Kent State, which was the worst FBS team this season). Tennessee is capable of playing well against a good defense, as evidenced by their 6.1 yppl against Alabama, but the Vols struggled in their other two games against very good defensive teams with just 4.8 yppl against Oklahoma and Georgia. I project just 4.5 yppl for the Volunteers against the best defense they’ve faced this season.

Overall, the math favors Ohio State by 8.2 with a total of 47 points and I bumped that up to Buckeyes by 9 points given Tennessee’s inexperience playing in cold temperatures (expected to be in the 20s throughout this night game).

Game Analysis view matchup stats

Texas-San Antonio (-10)  35   COASTAL CAROLINA  23

This game is being played on Coastal Carolina’s home field and I’m giving the Chanticleers half of a home field advantage (about 1.2 points). The math would have favored UTSA by just 1 point prior to adjusting for all of Coastal Carolina’s transfers, which include both quarterbacks that played this season.

Most of Coastal Carolina’s transfers and injuries are on the defensive side of the ball, but the biggest impact is having to rely on a pair of true freshmen quarterbacks that have never taken a college snap, as both starter Vasko and backup Kim have left the team. TE Kendall Karr (10.2 yards per target on 23 targets) and WR Carmeron Wright (7.8 YPT but only 37.5% success rate) are also gone and I expect a huge drop in production in the pass game going from Vasko to a pair of inexperienced freshmen. RB Christian Washington (537 yards at 5.3 ypr) has transferred to San Diego State but that just means more carries for star Braydon Bennett, who missed some time but still ran for 727 yards at 6.5 ypr and has 1900 rush yards at 6.4 ypr over his career.

The UTSA defense is not good, but they’ll be relatively better in this game given that Coastal Carolina will likely run more given the inexperience at quarterback. The Roadrunners have allowed 6.0 yppl to a schedule of FBS opponents that would combine to average only 5.4 ypp but they’re very good defending the run (4.4 yprp and 0.6 yprp better than average) and terrible against the pass (7.4 yppp allowed to quarterbacks that would average 5.8 yppp against an average defense). The Roadrunners have a couple of injured defensive players that have either missed most of the season (DT Evans) or were not impactful to the success of the defense (S Robinson and CB Harris), but starting DE Ronald Triplette has decided to transfer and is not on the bowl depth chart. Triplette’s 4 sacks were second on the team, but he did little (only 23 tackles and 1 other tackle for loss) and I value him at 0.4 points.

The UTSA offense averaged 5.7 yards per play against teams that would allow 6.1 yppl and they’re not as good as the 31 points per game would suggest, as that is a function of playing a fast pace and having an extra possession per game. The offense is intact for this game and should produce against a depleted Coastal Carolina defense that was 0.4 yppl worse than average in the regular season and could be without as many as 5 starters for this game. Top tackler LB Lubin is transferring along with top CB McDoom (13 of the team’s 42 passes defended), who has already committed to play at Cincinnati. Three other starters may also be out due to injury as linebackers Bruce and Gedeon missed the previous two games and safety Wiliams missed the regular season finale. There’s been no update on any of those players and I’ll assume that they’re all 50% likely to play – although their combined impact works out to just 0.3 yppl if none of them played (about 2 points). Not having Lubin and especially McDoom is very significant, as those two and the other transfers that affect depth project to be worth 4.7 points.

I have algorithms that project value of players based on their stats but there is a lot of variance in those projections so my advice is to stay away from this game. After making all the adjustments for current personnel, I get UTSA by 12 points with a total of 58 points.

Game Analysis view matchup stats

Lean – Northern Illinois (-3)  23   Fresno State  16

My math model would favor Fresno State by 1.5 points with everyone playing but the players that have left the Fresno State program are significantly more valuable than the players that have left Northern Illinois. Both starting quarterbacks will not play in this game but the NIU backup QB is experienced and has actually been a bit better than the departed starter but Fresno has to replace a two-year starter with two inexperienced quarterbacks. The Fresno defense has also been hurt more by the transfer portal and the line has not adjusted enough.

The Northern Illinois offense was 1.0 yppl worse than average and I don’t view the transfer of starting quarterback Ethan Hampton as a negative. Hampton averaged just 5.1 yards per pass play (against teams that would allow 7.1 yppp to an average QB), even with an 83-yard swing pass to a running back included. That’s 2.0 yppp worse than average even with that big play that was all yards after catch. Backup QB Josh Holst, meanwhile, was just 1.8 yppp worse than average on his 82 pass plays and the running of the two quarterbacks is comparable.

Not having top WR Trayvon Rudolph should not be a factor, as Rudolph averaged just 5.5 yards on his 52 targets against FBS opponents while the next 4 wide receivers combined for 6.4 YPT. The two starting wide receivers on the depth chart combined for 7.4 YPT on 79 targets against FBS opponents so not having Rudolph could be a plus. However, not having Antonio Brown is an issue, as Brown had 159 yards on 9 targets out of the backfield in FBS games – one of those being that 83 swing pass. Brown hasn’t been as effective in the run game, as his 4.65 yards per run against FBS teams is considerably less than the combined 5.33 ypr of Williams and Johnson Jr., who will split carries in this game. They will miss Rudolph’s 12.8 ypr average on his 14 runs this season (9.3 ypr against FBS teams) but overall, the run game should be a bit better with Williams and Johnson Jr. even without those jet sweeps that Rudolph runs effectively.

Fresno State had a very good defensive season, as the Bulldogs yielded just 5.0 yppl to teams that would combine to average 5.7 yppl against an average defense. However, that unit has been gutted with transfers, as the Bulldogs will be without two key defensive linemen (Lightfoot and Holmes), do-it-all LB Jackson (#5 tackler, team leading 4 sacks and 6 passed defended), and both starting cornerbacks (Lockridge and Hamilton), who combined for 7 interceptions and 27 total passes defended. I value those players at 0.8 yppl and 5.3 points. I still don’t expect a lot from an impotent NIU offense (just 360 yards at 4.8 yppl expected) but Fresno’s offense should be worse.

Fresno State struggled offensively this season, rating at 0.6 yppl worse than average with just 5.4 yppl against teams that would allow 6.0 yppl to an average offense. The Bulldogs can’t run the ball (1.1 yprp worse than average) and two-year starting quarterback Mikey Keene has left the program along with 4 of the top 5 wide receivers. The departed receivers accounted for 1292 receiving yards at 7.6 yards per target but #1 WR Mac Delena is on the depth chart and is apparently going to play in this game before heading to the NFL. Delena averaged 12.1 yards per target, which is significantly better than the rest of the receivers, but the experience is lacking at the two other receiver spots, and I value the departures in that position group as worth 0.7 yards per pass play. The absence of Keene, who is going to be replaced by a pair of quarterbacks with just 8 career pass attempts, is even more damaging.

The Northern Illinois defense allowed just 18.1 points and 287 total yards per game at 4.7 yppl in 11 games against FBS opponents (that would average 5.3 yppl), including holding Notre Dame to just 14 points and 4.8 yppl in their week 2 upset win in South Bend. That unit suffered a couple of injuries and will be missing two of the top four tacklers in S Banner and DL Gill-Howard, along with rush specialist Nevaeh Sanders (4.5 sacks). DE Thomas (2 sacks in 4 games before injury) and CB Knighton, who missed the last 4 games with an injury, are also not going to play. Those absences equate to 3.2 points based on my algorithms for assessing players that are out, but the Huskies still rate as an average defense on a national scale. I project Fresno with just 4.4 yards per pass play in this game and the coaching staff will likely call more run plays than usual. However, the Bulldogs are projected to average only 4.3 yprp, so moving the ball consistently is going to be challenging.

My math favors Northern Illinois by 6.5 points with 39.7 total points.

Game Analysis view matchup stats

Lean – Over (62.5) – San Jose State (-3/-3.5)  37   South Florida  31

San Jose State was 0.5 yards per play better than average offensively this season, averaging 6.6 yppl against teams that would allow 6.1 yppl to an average offense. That attack got better when Walker Eget took over at quarterback from Emmett Brown. Brown averaged 6.9 on 226 pass plays against teams that would allow 7.1 yppp to an average quarterback while Eget averaged 7.9 yppp against teams that would allow 6.6 yppp to an average QB. Eget is more interception prone but overall the Spartans were better offensively after the quarterback change.

WR Nick Nash, who led the nation in receptions, receiving yards and receiving touchdowns, has left the team to prepare for the NFL combine. That sounds like a major loss but it’s really not. Nash was targeted 176 times, which was 37 more than the second most in the nation, and his 7.9 yards per target is not only worse than average for a wide receiver but it is also 1.2 YPT worse than the rest of the wide receivers on the Spartans, who averaged 9.1 YPT. Nash had a slightly better success rate than the rest of the San Jose State receivers but if I value Nash strictly on success rate, compared to the rest of the wide receivers, and ignored his lower YPT he would still be only worth 0.6 points, which is what I decided to adjust.

San Jose State’s pass-heavy attack (66% pass plays) should work well against a South Florida defense that is decent defending the run (0.1 yprp better than average) but was 1.0 yppp worse than average against the pass (7.0 yppp allowed to quarterbacks that would combine for 6.0 yppp against an average defense). South Florida will be without one impact player, as S Tawfiq Byard has decided to transfer and is not on the depth chart for this game. Byard played closer to the line of scrimmage and will affect the run more than the pass. As he was 4th on the team in tackles, including 8 for loss (2 sacks) while breaking up just 1 pass. Byard is worth a bit less than a point taking into account that San Jose State throws twice as often as they run, and I made no adjustment for starting DT Doug Blue Eli leaving, as he had just 17 tackles and only 1.5 for loss.

The South Florida offense also had a change of quarterback mid-season, but the offense under Bryum Brown has the same rating as it did in the last 7 games with Bryce Archie at quarterback, as the offense was 0.2 yards per play worse than average under both starting quarterbacks.

San Jose State’s defense was 0.2 yppl better than average for the season (5.2 yppl allowed to teams that would average 5.4 yppl against an average defense), but both starting cornerbacks have hit the portal and will not play in this game. DJ Harvey, who signed with USC, is the biggest loss as he not only had 10 passes defended (4 of them interceptions) but also had 7 total tackles for loss. I value those two at 3.0 points in this game, which is enhanced by the 83 plays from scrimmage that I expect from USF.

After all the adjustments I get San Jose State by 6.2 points with a total of 68 points (assuming the better than average weather (+1.0 points) that is forecasted holds up). I’ll Lean Over at 63 points or less.

Game Analysis view matchup stats

Pittsburgh (-6.5)  27   Toledo  20

Pitt would be favored by 10 or 10.5 points with everyone available for both teams and I’d have favored Pitt in this game by 13 points, as the Panthers should have been better in scoring margin based on their metrics.

The Panthers’ offense was 0.1 yards per play worse than average this season in FBS games, as they managed just 5.6 yppl against teams that would allow 5.7 yppl to an average attack. Star RB Desmond Reid is expected to play after missing the season finale but starting QB Eli Holstein has been downgraded to not expected to play and that leaves third-string walk-on David Lynch at the helm of the offense. NFL bound WR Konata Mumpfield has declared for the NFL draft, but he is listed on the bowl depth chart, which has been updated given that others that have transferred are not on the depth chart. It’s not uncommon for players that have declared for the NFL draft to play in the bowl game, but you can’t be sure that will be the case for Mumpfield until he announced it one way for the other or can be seen before the game suited up or not. Mumpfield averaged 9.5 yards on 79 targets against FBS opponent while the next 4 wide receivers combined for just 6.3 YPT. Mumpfield would be worth 2.7 points is he doesn’t play (which I’m assuming he won’t for now). Backup receivers Censure Lee and Daejon Reynolds are not on the depth chart so the depth at that position is an issue even if Mumpfield plays and I am docking the Panthers another 0.7 points for that. TE Gavin Bartholomew has also declared for the NFL draft, which is not a detriment if he skips this game given his 5.6 yards per target (the other two TEs combined for 7.0 YPT in FBS games). I expect more running with Reid given the inexperience at quarterback, which would slow down the pace of the Pitt offense and take some of the pressure off of Lynch.

Toledo’s defense allowed just 21.1 points and 5.0 yppl against FBS teams, although it was against a mostly weak schedule of opposing offensive units that would combine to average only 5.2 yppl against an average defense. The Rockets have a potential opt-out in All-MAC DT Darius Alexander but I’m going to assume he’ll play since he’s listed on the depth chart. I project 374 total yards at 4.8 yppl if Mumpfield is out for Pittsburgh (and Alexander plays for Toledo), adjusted for perfect dome conditions.

Toledo’s offense really struggled this season, as the Rockets managed just 5.4 yppl against FBS teams despite facing a collection of defensive units that would allow 6.2 yppl to an average attack. I’ll assume that potential opt-out WR Jerjuan Newton plays, but Toledo is going to struggle against a good Pitt defense. The Panthers’ defense was 0.8 yards per play better than average against FBS teams this season (5.2 yppl allowed to teams that would average 6.0 yppl against an average defense) and that unit is intact for this game, as the top 18 players in tackles are all listed on the depth chart. I project just 329 yards at 4.4 yppl for Toledo in this game.

The math favors Pitt by 6.5 points with 46.5 total points, but there is a chance the Mumpfield might play for the Panthers. This game is certainly a pass, as it’s hard to know what to expect from Lynch in this game.

Game Analysis view matchup stats

Lean – Kansas State (-6.5)  33   Rutgers  22

Lean – Over (50.5)

Before adjusting for opt outs and transfers my math model would have favored Kansas State by 13.3 points in this game against a lucky Rutgers team that was out played 5.4 yards per play to 6.3 yppl in games against FBS teams but managed to outscored opponents by 0.6 points per game because they only turned the ball over 8 times all season, which is quite random for a team with an inaccurate quarterback (just 54.8% completions.

The Rutgers’ offense averaged a modest 5.5 yards per play in games against FBS opponents, excluding the Nebraska game, which was played in heavy winds (the lost 7-14). Their opponents in those games would combine to allow 5.3 yppl to an average attack, so the Scarlet Knights have been 0.2 yppl better than average offensively but have scored more than they should have because of their extreme turnover luck (just 6 offensive turnovers in those 10 games). Star RB Kyle Monangai has announced his intention of entering the NFL draft and he announced on Tuesday that he would not play in this game. He is worth 1.1 points.

Kansas State’s defense was 0.8 yppl better than average this season but the Wildcats will be without NFL-bound CB Jacob Parrish, who led the team with 8 passes defended and supported the run very well for a corner (50 tackles is 5th on the team). Parrish is worth 1.6 points based on my algorithm, but the Wildcats still have a significant advantage over the mediocre Rutgers’ attack, which I project to gain 377 yards at 5.4 yppl even in perfect dome conditions (the Chase Field roof has historically been closed for this game)

Kansas State’s offense was 1.2 yppl better than average in the regular season but they’ll be without star RB DJ Giddens, who ran for 1343 yards at 6.6 ypr this season and also contributes some to the passing game with 7.4 yards per target on 34 targets, compared to just 4.0 YPT on 33 targets for the other backs. The other running backs combined for 6.0 ypr, so they’re more than capable of running the ball effectively against a bad Rutgers’ run defense. However, Giddens averaged 7.4 yards per target on 34 targets in the pass game, compared to just 4.0 YPT on 33 targets for the other backs and his value in the pass game is just as much as it is in the rush attack. Kansas State is also without one of their starting offensive linemen and their #2 WR Keagan Johnson, whose 7.6 yards per target will be replaced by 4 guys (not including leading WR Brown) that combined for just 6.6 YPT. Overall, the Wildcats’ offense projects to be 3.3 points worse without the 3 starters.

Rutgers’ defense is not as good as the 25.4 points per game they allowed to FBS opponents, as one game was played in a windstorm (I also took that game out of their offensive numbers) and Rutgers allowed 406 yards at 6.5 yppl in the other 10 FBS games while facing teams that would combine to average 5.9 yppl against an average defense. I rate the Scarlet Knights’ defense at 0.5 yppl worse than average after taking out the garbage time yards they allowed against Akron and Kansas State’s offense is still 0.7 yppl better than average without those 3 starters. I project 451 yards at 7.2 yppl for the Wildcats in this game.

Kansas State lost their regular season finale 21-29 at Iowa State, but the Wildcats are 34-9 ATS in their next game after a loss and spread loss since 2009, including 16-3 ATS under coach Klieman (2-0 in bowls). Kansas State is the better team and I’ll lean with the Wildcats at -7 or less and will Lean Over 51 or less.

Game Analysis view matchup stats

Lean – Bowling Green (-7.5)  34   Arkansas State  22

I was surprised to see that Arkansas State was in a bowl game because they’ve been a bad team in my mind the whole season. But, I guess 7-5 gets you a bowl even if you were outscored by 7.4 points per game (they were a lucky 6-0 in games decided by 7 points or less). Bowling Green isn’t a good team either but they’re good enough to beat Arkansas State.

Bowling Green’s offense was 0.3 yards per play worse than average this season but veteran quarterback Connor Bazelak rarely throws interceptions (just 5 this season and 12 in 23 games since last season). Star RB Terion Stewart entered the transfer portal and has committed to Virginia Tech, although he’s also stated that he’s going to play in this bowl game, which I suppose he can if he just committed to sign with the Hokies rather than actually signing with them. Stewart averaged 5.6 ypr on 160 runs, but he got 161 yards on 14 runs against FCS team Fordham while averaging a modest 5.0 ypr against FBS teams. The Falcons’ two backup running backs combined for 5.0 ypr against FBS teams and the run offense was relatively better in the two games that Stewart missed this season.

Arkansas State’s defense is horrible. The Red Wolves gave up an average of 467 yards per game at 7.0 yards per play against teams that would average 5.6 yppl against an average defensive team. The defense has been worse without DE Jayden Jones, who has missed most of the last 5 games with injury (and has now transferred) and it’s probable that star CB Dontay Joyner will not play after entering his name in the portal and committing to Maryland. Star TE Harold Fannin will be headed to the NFL, but he says that he intends to play in this game and he is listed on the depth chart that was released on Monday morning. Bowling Green should have a good day offensively and I project 440 yards at 6.9 yppl for the Falcons in this game.

The Arkansas State offense was 0.7 yppl worse than average, as their 5.4 yppl was against teams that would combine to allow 6.1 yppl to an average FBS attack. Bowling Green’s defense was just 0.2 yppl worse than average, allowing 20.3 points per game at 5.4 yppl to teams that would combine to average 5.2 yppl against an average defense. The Falcons’ defense could be without All-MAC LB Joseph Sipp Jr., who has committed to Kansas. Sipp Jr. led the team in tackles, sacks (6) and total tackles for loss (10.5) and is worth 1.3 points according to my algorithm and I have him as out for now – although he is listed on the depth chart. I project 360 yards at 5.3 yppl for the Red Wolves.

I will Lean with Bowling Green at -8 or less.

Game Analysis view matchup stats

Georgia Tech (-2.5)  25   Vanderbilt  24

Georgia Tech’s offense was very good until starting quarterback Haynes King was injured late in week 7 and couldn’t throw the ball for the next 4 games (he missed 2 and played sparingly as a runner in two others). King was obviously healthy in their Yellow Jackets’ 42-44 overtime loss to Georgia, as he averaged 8.2 yards per pass play and ran for 111 yards. Georgia Tech’s offense was 1.1 yppl better than average in the 7 games against FBS teams with Haynes healthy, averaging 6.5 yppl against teams that would allow 5.4 yppl to an average offense. That unit will be without top WR Eric Singleton Jr but his mediocre 8.0 yards per target and 47% success rate likely won’t be missed, as the rest of the available wide receivers, including #2 Malik Rutherford (took his name out of the portal and is listed on the depth chart), combined to average 8.1 yards per target with a 48% success rate. Starting TE Jackson Hawes is also out, but his 9.9 YPT on 19 targets is below the 10.8 YPT on 13 targets by the other tight ends. Left tackle Corey Robinson (transfer) and Jordan Brown, who started the season as the LT, are both absent from the depth chart and I’ve adjusted 0.8 points for that position. Overall, the Georgia Tech offense get a bit of an upgrade from their season rating with King healthy.

Vanderbilt’s defense was better in point allowed (25.2 ppg against FBS teams) than they were in yards per play allowed, as the Commodores gave up 6.4 yppl to FBS teams that would average 6.2 yppl against an average defense. Vandy is a bit better than average defending the run, but they were often torched through the air by good quarterbacks and allowed 7.5 yards per pass play to FBS quarterbacks that would combine to average 6.9 yppp against an average defense. It looks like star S De’Rickey Wright wants to put more of himself on tape to help his NFL stock, as he’s listed on the recently released depth chart and will also take part in the East-West Shrine game after the bowl season. Assuming Wright plays the entire starting defense will take part in this game. However, that defense isn’t good enough to keep Georgia Tech from moving the ball consistently.

Vanderbilt’s offense was impressive at times, but the Commodores were just a bit better than average overall this season with 5.5 yppl coming against FBS teams that would allow 5.3 yppl to an average team. Vandy will be down one starter, as tackle Gunnar Hansen has decided to transfer and is not on the depth chart.

Georgia Tech’s was 0.3 yards per rushing play and 0.3 yards per pass play better than average this season and rates at about the same level as the average defense that Vanderbilt faced this season. The Yellow Jackets will be without edge rusher Romello Height, who started every game and led the team in total tackles for loss (6.5 TFL, 2.5 of them sacks). Safety Taye Seymore, who started 7 games, is also out but the defense played at about the same level in the games he didn’t play this season. I value those absences at 0.9 points. The Vandy offense and the Georgia Tech defense rate about the same.

Georgia Tech is no doubt a better team from the line of scrimmage, but Vanderbilt has outstanding special teams units while the Yellow Jackets are worse than average in special teams. Overall, the math favors Georgia Tech by 4 points but Vanderbilt applies to a 57-16 ATS bowl angle that plays on teams that lost 9 or more games the previous season. Those teams tend to be more excited about playing in a bowl game after being so bad the previous season and that angle has already won with Sam Houston State this year.

Game Analysis view matchup stats

Oklahoma (-1)  25  Navy  21

Note: Oklahoma WR Deion Burks, who was expected to play, has been ruled out with a concussion. I value him as worth 1.0 points. The line has moved down to -1.

My math would favor Oklahoma by just 4.2 points before factoring in all the of the opt outs and transfers that the Sooners have had. But, one of those departures should help the team more than the rest will hurt the team.

Oklahoma’s offense will be without starting quarterback Jackson Arnold, their leading rusher Jovantae Barnes, 6 of the 11 wide receivers that caught a pass this season, and their starting tight end. Not having Arnold is an upgrade, as Arnold averaged only 3.8 yppp against FBS competition while Michael Hawkins Jr. averaged 5.0 yards on 83 pass plays (against FBS teams) despite 55 of those coming against Texas and Tennessee, who are both elite in pass defense.  Hawkins averaged his 5.0 yppp against teams that would allow 4.7 yppp to an average quarterback while Arnold’s 3.8 yppp came against teams that would allow 5.6 yppp. Hawkins is an upgrade mostly because he at least looks down the field occasionally (10.6 yards per completion) while Arnold settled for mostly short passes and averaged just 8.6 ypc against FBS teams.

It looks like starting WR Deion Burks will be able to play in this game after missing 7 of the last 8 games, as he’s listed as a starter on the depth chart. The other two starters and two backups combined for just 5.6 yards per target and Burks only averaged 5.7 YPT – although 40 of his 43 targets came from Arnold. The wide receivers that are injured and/or have left the team combined for 7.5 yards on 74 targets and are worth 0.4 yards per pass play (1.4 points). However, that’s less than the upgrade from having Hawkins at quarterback instead of Arnold. The Oklahoma offense should be significantly better with Hawkins, even with the downgrade in the receiving group. As far as RB Barnes being out, I don’t see that being an issue, as Barnes averaged just 3.6 ypr against FBS teams (he had 203 yards on 18 runs against FCS team Maine) and the run game improved in the final 3 games of the regular season with Barnes out. Hawkins’ compensated yards per run isn’t quite as good as Arnold’s running numbers but I do rate the Sooners’ run attack better without Barnes.

Navy’s defense was 0.1 yards per play better than average for the season but that included two games against fellow option teams Air Force and Army and teams that run the option know how to defend the option. In games against non-option teams, which is what I’ll apply for this game, the Midshipmen were 0.1 yppl worse than average defensively. Oklahoma should move the ball at a decent rate with Hawkins at quarterback against Navy’s defense, which is 0.9 yppl worse than the average FBS defense that the Sooners faced this season.

Navy’s option attack averaged 6.5 yards per play against teams that would allow 6.0 yppl to an average offense and they’ll be tested in this game by a very good Oklahoma defensive front. The Sooners’ had the nation’s best run defense in the regular season, as they allowed just 3.7 yards per rushing play to FBS teams that would combine to average 5.4 yprp against an average defense. Overall, the Oklahoma defense was 1.2 yppl better than average in the regular season. However, leading tackler LB Danny Stutsman (110 tackles) and safety Billy Bowman (#4 tackler) have opted out of this game to prepare for the NFL. The other defenders that have declared for the draft are on the depth chart and I’ll assume that they’ll play.

The Sooners still project to be a very good defense even without two stars, as I still rate that unit at 0.8 yards per play better than average and 1.3 yards per rushing play better than average. The option offense, however, is designed to neutralize talent discrepancies and this year’s Navy team was relatively better running the ball against good run defenses than they were against bad run defenses. The Midshipmen averaged 9.3 yprp against a very good Memphis run defense and they averaged 5.7 yprp against Notre Dame’s stout defensive front while also ripping apart a good ECU run defense for 5.9 yprp. To further prove my point, Navy ran for 5.9 yprp for the season but averaged just 5.7 yprp against the 3 worst run defenses that they faced (Bucknell, Temple, and Charlotte). Despite the inverse correlation between Navy’s yards per rushing play and the level of the opposing run defense, I’m still going to assume that the Midshipmen run for less than their season average in this game. Obviously, I have a formula to account for option teams being relatively better against better defensive teams (and relatively worse against worse defensive teams) and applying it to this game would project 5.3 yards per rushing play and 5.9 yards per play, which is comparable to the 5.8 yppl they gained on a neutral field against Notre Dame.

My math favors Oklahoma by 5.5 points with a total of 47.5 points (assuming neutral weather) and the Sooners apply to a 71-27-2 ATS bowl angle that plays on teams that averaged a relatively low number of points in the regular season. However, the military academies have historically been very good in bowl games, as Army, Navy, and Air Force are a combined 40-14 ATS in bowl games since 1981, including 26-4 ATS if they have a win percentage of greater than 0.667 (11-0 ATS since 2014). The trends cancel each other out and I’d rather have the Sooners, although not enough to recommend a bet.

Game Analysis view matchup stats

Game Analysis view matchup stats

Syracuse (-17)  40   Washington State  23

Washington State some key players to the portal and they’ll be without their head coach and both coordinators.

The biggest loss on the Washington State offense is obviously QB John Mateer, who was 0.5 yards per pass play better than average throwing the ball but was even more valuable as a runner with 978 yards on 145 runs (6.7 yprp) in 11 games against FBS teams. Not having WR Kris Hudson is not a negative given that Hudson averaged just 6.6 yards on his 88 targets against FBS teams while the 4 wide receivers listed as starters on the depth chart for this game combined for 10.2  YPT on 171 targets. Not forcing the ball to Hudson could actually be a positive, but I’ve settled on no adjustment for the receiving group. The absence of top RB Parker is detrimental given his 5.0 ypr against FBS teams is being replaced by two running backs that combined for just 3.9 ypr this season.

Zevi Eckhaus is slated to start at quarterback in place of Mateer and he has 3 years of starting experience at the FCS level at Bryant, where he averaged 7.0 yards per pass play in his final two seasons there while adding 744 yards on 109 runs (6.8 yprp). Eckhaus obviously faced less talented defenses and my adjustment for schedule strength (and better talent to work with at WSU) equates to 6.3 yppp had he faced Washington State’s schedule of defenses this season, which is 0.8 yppp worse than Mateer. However, Eckhaus has proven that he can run at about the same level as Mateer. The question is if he’ll run as often as Mateer ran this season (13.2 run per game against FBS team), as Eckhaus only ran it about 5 times per game at Bryant.

Syracuse was horrible defending the run this season (6.1 yprp allowed to teams that would average 5.3 yprp against an average defense) and they were just 0.2 yppp better than average defending the pass (average without CB Bellamy, who was #2 on the team in passes defended and won’t play after transferring). The Washington State running backs only project to run for 4.6 yards per run but Eckhaus should average around 7 yards per run when he takes off. I’ll assume he runs less than Mateer ran but more than he ran it at Bryant. The WSU offense projects to be about 6 points worse than if everyone were playing but they still should move the ball at a decent rate, as Eckhaus is a decent passer and a good runner, and the Syracuse defense is worse than average.

The Syracuse offense was 0.3 yppl better than average this season (6.0 yppl against teams that would allow 5.7 yppl to an average team) but they’re better than that metric because they average a lot of plays per drive and averaged 17.6 more plays per game than their opponents this season. The Washington State defense was 0.4 yppl worse than average this season, although a pass-heavy Syracuse offense isn’t as likely to take advantage of the Cougars’ defensive weakness, which is stopping the run. WSU has some starters that have transferred and won’t play, and I project the Cougars’ defense to be 0.7 yppl worse without the departed players, which equates to 5.4 points given the number of plays that Syracuse is likely to run in this game.

After all the number crunching to account for Washington State’s attrition, the math comes out to Syracuse by 17.0 points with a total of 62.9 points.

Game Analysis view matchup stats

1-Star Best Bet – *USC Team Total Over (24 -115)

Or

Strong Opinion USC (+3.5) and Strong Opinion Over (52.5)

Southern California (+3.5)  30   Texas A&M  29

USC is easily the best 6-6 team in the nation, as Trojans lost 5 of their games by 7 points or less (leading in the 4th quarter in most of those) while their other loss was by 14 points to Notre Dame in a game in which they had two balls intercepted at the Notre Dame endzone that were both returned for touchdowns. USC rated at 1.6 yards per play better than average while Texas A&M has been 1.3 yppl better than average (+1.5 yppl with Reed at QB). The Trojans were the better team from the line of scrimmage in the regular season but overall the math model would have favored A&M by 0.3 points (with Reed at QB for the Aggies) before adjusting for opt outs and transfers. Those adjustments favor USC and the market adjusted the opposite direction, which I believe gives us value on the Trojans.

The USC offense was 1.3 yppl better than average this season, averaging 6.5 yppl against teams that would allow 5.2 yppl to an average team. Jayden Maiava took over for Miller Moss at quarterback for the final 3 regular season games and Moss has transferred. The two were nearly identical in compensated yards per pass play and interception rate. Wide receivers Branch, Hudson, and Robinson have all left the team, but that could be a plus. Those 3 combined for 1361 yards at an average of 7.1 yards per target and a 44% success rate while the rest of the wide receivers combined for 1429 yards at 8.55 YPT and a 51% success rate. I didn’t make a positive adjustment for the receiver group, although there is some evidence that they might be better off without the departed trio. Top two running backs Marks and Joyner are also not playing in this game, and the Trojans will also be down 2 offensive line starters. USC has plenty of talent at receiver and the running backs and offensive linemen combine to be worth only 2.3 points in my estimation.

The Texas A&M defense was 0.7 yppl better than average in the regular season (5.5 yppl allowed to FBS teams that would average 6.2 yppl against an average defense) but that unit will be without 3 starting defensive linemen that have opted out to prepare for the NFL draft. I value those losses at 3.1 points.

The Texas A&M offense was 0.5 yppl better than average against FBS teams this season (5.8 yppl against teams that would allow 5.3 yppl to an average attack) but they were better with Marcel Reed at quarterback, as he was a better passer than opening day starter Conner Weigman as well as being a more dangerous runner (555 yards on 84 runs). The absence of WR Cyrus Allen is not an issue, as his 39% success rate on his 33 targets was the worst among all wide receivers. However, the loss of top RB Le’Veon Moss does hurt the attack, as Moss’s 6.1 ypr on 112 runs against FBS opposition is much better than the rest of the running backs. Amari Daniels was next best at 4.6 ypr against FBS teams and he’ll get the bulk of the work. The run offense has been 0.4 yprp worse (compensated for opposing defenses) in the 3 games Moss missed at the end of the season. A&M still rates at 0.5 yppl better than average with Reed at QB and without Moss (worth 1.3 points).

USC’s defense was 0.3 yppl better than average in the regular season and I don’t think they’ll be much worse for this game. Losing CB Jaylin Smith is a bit of a blow, but the Trojans are getting back their best defensive player, LB Eric Gentry, who led the team with 6.5 total tackles for loss (2 of them sacks) in just 4 games before sitting out the rest of the regular season with a concussion. He could have come back earlier but they want to redshirt him and now he can play since bowl games do not affect the redshirt status. Gentry has been practicing with the team and has hinted that he’ll play. I won’t upgrade the defense because of Gentry, since it’s unknown how much he’ll play, but losing Smith is only worth 1.0 points.

I favor the underrated Trojans by 0.9 points, with 59.1 total points (adjusted for playing in a dome), and they showed last year how deep they are by winning their bowl game 42-28 as a 7.5-point underdog. The market adjusted more for USC’s absent players than they did A&M’s and I value the Aggies’ losses as more costly – most likely because I didn’t adjust for USC’s 3 missing receivers.

With value on USC and the Over the best way to play this game is the USC Team Total Over 24 (at 24 -120 or better). If you don’t have Team Totals, or if your TT is higher than 24 at -120 odds, then play Strong Opinions on USC (+3.5) at +3.5 or more and on the Over (52.5) at 53 or less. Strong Opinions are recommended as 0.5 Star plays.

Game Analysis view matchup stats

Lean – North Carolina (-2.5)  33   Connecticut  26

Lean – Over (52.5)

Connecticut is not a good team and I would have favored North Carolina by 9.4 points before adjusting for opt outs and transfers that won’t play in this game.

The North Carolina offense ended the regular season rated at 0.5 yards per play better than average and were +0.6 yppl with quarterback Jacolby Criswell at quarterback. Running back Omarion Hampton was the star of the offense with 1660 yards at 5.9 ypr but he opted out to prepare for the NFL draft. I value Hampton, and one starting offensive lineman as worth a combined 3.8 points.

The UConn defense was 0.1 yppl worse than average, as the 5.4 yppl that the Huskies allowed was against a schedule of mostly bad offensive teams that would combine to average only 5.3 yppl against an average defense. The Huskies will be without starting DL Gourdine and CB Malcolm Bell, who started 4 games. Those two are worth 1.2 points.

The Connecticut offense was 0.6 yppl worse than average in the regular season but the coaching staff finally wised up and benched horrible quarterback Nick Evers, who was 2.0 yards per pass play worse than average. Joe Fagnano started the last 2 games and was better than Evers throughout the season (on 187 pass plays). His best game was against FCS team Merrimack, who has a horrible defense. I didn’t count that game in my ratings but I still rate Fagnano at just 0.6 yppp worse than average (1.4 yppp better than Evers, who had 190 pass plays against FBS teams). Leading receiver Skyler Bell has entered the transfer portal but he’s listed on the bowl depth chart so I’ll assume that he’ll play. Running back Durell Robinson, who ran for 731 yards and averaged 6.6 ypr against FBS opponents, has transferred and he is not on the depth chart and will not play. The other running backs combined for 5.6 ypr, which equates to a difference of 0.7 points.

North Carolina’s defense was 0.1 yppl worse than average this season (5.6 yppl allowed to teams that would combine to average 5.5 yppl against an average team) and they’ll be without 4 key starters in DE Rucker (6 sacks), DT Ritzie (tied for team lead in sacks with 6.5), LB Echols (#2 tackler and #2 in passes defended) and FS Lane, who led the team in tackles. I value those 4 players at 4.3 points in this matchup.

After all the adjustments I get UNC by 3.8 points with a total of 59.3 points with -0.7 points for the projected weather (rain but not windy). The Tarheels ended the season with 3 straight losses (0-3 ATS) but that negative run sets them up in a 60-16-4 ATS bowl situation and they also apply to a 67-24-3 ATS bowl situation.

I’ll Lean with North Carolina at -3 or less and I’ll Lean Over 54 points or less.

Game Analysis view matchup stats

Lean – Nebraska (-4)  29   Boston College  20

Boston College is mediocre offensively, averaging 5.5 yppl against FBS teams that would combine to allow 5.5 yppl to an average attack. The Eagles’ coaching staff made a quarterback change for the final 3 games by inserting Grayson James to take the place of Thomas Castellanos, who abruptly left the team the week he was told he wasn’t the starter. James was horrible as a starter at FIU and was terrible in a week 5 start against Western Kentucky (4.9 yards per pass play against a bad pass defense), but he played better in the 3 starts at the end of the season and overall was about the same as Castellanos. The rushing attack got better over the last 4 games with RB Jordan McDonald running for 308 yards on 45 runs (6.8 ypr) after not running the ball at all in the team’s previous 7 games against FBS teams. If McDonald had been in the running back rotation all season the Eagles’ run game would have been 0.25 yprp better (0.8 points per game). All-ACC tackle Ozzy Trapilo has opted out and won’t play and that hurts the offense a bit (0.6 points).

Nebraska’s defense was 0.3 yppl better than average in the regular season (I did not include the 14-7 game against Rutgers, which was played in heavy wins and made both teams ineffective throwing the ball). That unit will be missing a few players that contributed to the defense (Butler, Gbayor, Williams) who combine to be worth 1.7 points. I did not make an adjustment for CB Tommi Hill because the defense was better in the 5 ½ games he missed.

Nebraska’s offense wasn’t as good as hoped with highly touted freshman Dylan Raiola at quarterback, as the Cornhuskers were just 0.1 yards per play better than average while Raiola was only 0.2 yppp better than average (not including that Rutgers game in heavy wind). Leading rusher Dowdell (614 yards but only 4.3 ypr) hit the transfer portal and is not on the depth chart, but that just means more carries for Emmett Johnson (5.3 ypr), who recently announced that he’s staying with the team. WR Isaiah Neyor has opted out to apparently prepare for the NFL draft. I doubt any NFL team will be drafting a receiver that averaged a sub-par 7.0 yards per target with a lowly 39% success rate and not having Neyor for this game could be a plus for the Huskers – although I’ll make no adjustment given that the next 5 wide receivers were not that much better (7.2 YPT but a much better success rate).

The Boston College defense allowed 5.8 yppl to teams that would combine to average 5.9 yppl against an average team, although I rate that unit at 0.3 yppl better than average after adjusting for the outlier against Virginia Tech (9.3 yppl allowed). The big loss for the BC defense is the news that All-American DE Donovan Ezeiruaku will not play in this game. Ezeiruaku led the nation in sacks with 16.5 (55% of BC’s sacks) while also leading the Eagles in tackles, which is very rare for a defensive lineman. My algorithm suggests that he’s worth 2.4 points, but it could be more than that given his dominance.

The math favors Nebraska by 6.1 points and BC applies to a 13-63-1 ATS bowl situation that plays against teams that were playing better than expected at the end of the season. The Eagles are on a 5 game spread win streak but bowl underdogs that have covered the spread in 5 or more consecutive games are just 2-21-1 ATS. I’ll Lean with Nebraska at -4 or less.

Game Analysis view matchup stats

Texas Christian (-10)  31   Louisiana-Lafayette  24

Louisiana-Lafayette had a very good offense this season, as the Ragin’ Cajuns averaged 7.0 yards per play with either of their top two quarterbacks on the field (against FBS teams that would allow 6.2 yppl to an average team). Starting QB Ben Wooldridge was injured in week 12 and veteran former starter Chandler Fields was even better throwing the ball the last 3 ½ games. Fields was injured in the Sun Belt Conference championship game, but Wooldridge is back practicing with the team and is now expected to start. Wooldridge won’t have #2 WR Harvey Broussard and possibly #4 WR Robert Williams (injured in the last game) to throw to, but those two combined for just 648 receiving yards (against FBS teams) at a modest 8.0 yards per target and the receiver group is fine with Lance LeGendre on the field (12.4 YPT on 64 targets). The loss of TE Terrance Carter, however, is an issue, as Carter averaged 10.4 yards per target on 63 targets against FBS teams while the other two tight ends combined for just 148 yards at 5.9 YPT. Some of Carter’s targets will go to the wide receiver group but I value him at 1.8 points. Not having big play RB Dre’lyn Washington also hurts, as Washington averaged 6.7 ypr on his 68 runs against FBS teams while adding 107 receiving yards on just 7 targets. Washington is worth 1.2 points. The Ragin’ Cajuns are also weaker up front with starting LT McGowen and starting LG Gillie both absent from the bowl depth chart. I rate the ULL offense at 0.1 yppl worse than average with those key losses factored in, with TE Carter being the big loss.

The TCU defense gave up 28.8 points per game to FBS opponents but that average was skewed up by the 66 points they allowed to SMU (on just 6.3 yppl) and the Horned Frogs are actually 0.8 yppl better than average defensively – yielding 5.4 yppl to FBS teams that would combine to average 6.2 yppl against an average defense. That unit appears to be intact now that safety Jamel Johnson has removed himself from the transfer portal – although he’s now listed as a backup on the depth chart and I adjusted a bit for that.

The TCU offense was 0.9 yppl better than average in the regular season, averaging 6.4 yppl (with QB Hoover in the game) against teams that would allow 5.5 yppl to an average team. The top two pass catchers, Jack Bech and Savion Williams will both miss this game, but I don’t think that’s going to be an issue given that they combined for 9.1 yards per target against FBS teams while the rest of the wide receivers combined to average 10.1 YPT on 145 targets. Bech, however, was certainly the star (1023 yards at 11.2 YPT in FBS games) and demanded attention from opposing defenses, but Williams was horrible (6.8 YPT on 81 targets) compared to the other receivers. Williams was valuable in the running game, however, as he tallied 322 yards at 6.3 yards per run. Leading rusher Cam Cook has transferred but his 3.8 ypr in FBS games won’t be missed. Overall, I project the Horned Frogs’ attack to be only slightly worse than they were in the regular season.

Louisiana’s defense was average this season, (5.5 yppl allowed to FBS teams that would combine to average 5.5 yppl against an average team), but they are down what amounts to 2.5 starters, as S Tyree Skipper is still injured and LB Carmycah Glass and CB Justin Agu have both entered the transfer portal and are not on the depth chart. Glass was #2 in tackles but had very few impact plays and Agu is solid in coverage. That trio is worth 2.4 points based on my algorithm and ULL’s defense projects to be 0.3 yppl worse than average with available defenders.

After all the adjustments I favor TCU by 11.3 points with a total of 55.5 points. However, Louisiana-Lafayette applies to my best bowl situation – a 78-15-1 ATS angle that has already won once this year with Sam Houston State winning straight up as a dog.

Game Analysis view matchup stats

Lean – Over (57.5) – Miami-Florida (-4.5)  35   Iowa State  29 (With Ward and Restrepo in for Miami)

The offense carried Miami to 10 wins this season and it looks like star QB Cam Ward is going to play one more game for the Canes before heading to the NFL. WR Restrepo is another possible opt-out but there’s been no word on that and the fact that Restrepo has accepted an invitation to the Senior Bowl means that he’s not concerned about playing more football. Starting WR Isaiah Horton has left the team, but Horton averaged only 7.6 yards per target, which was much lower than the rest of the team. If the rest of the Hurricanes’ offense plays, then the Hurricanes would be 1.8 yards per play better than average and should move the ball well in this game.

Iowa State’s defense was gutted in the Big 12 Championship game by Arizona State (45 points and 8.7 yppl) but the Cyclones were still 0.3 yppl better than average for the season on the stop side of the ball (5.7 yppl allowed to teams that would average 6.0 yppl against an average defense). The Cyclones missed safety Malik Verdon against Kansas (8.1 yppl allowed) and most of that game against ASU and he is expected to miss this game too along with DT Onyedim, who was honorable mention All-Big 12. Those two are worth 1.3 points and the Cyclones look like an average defense without them. ISU’s defense, while not great, does force the opposing quarterback to be patient, as the Cyclones only allowed 52.8% completions and 11.7 yards per completion. They would rather put an extra defender in coverage than pressure the quarterback, which is why they averaged only 1.1 sacks per game. Ward will have time to throw and I think he’s got enough experience to be patient and wait for a receiver to get open before firing.

The Iowa State offense was just 0.4 yppl better than average this season (6.0 yppl against teams that would allow 5.6 yppl to an average team) and that unit appears to be intact for this game and gets TE Brahmer back from injury (he averaged 10.5 yards on his 17 targets in 8 games).

The Miami defense got a lot of heat from the media but was actually 0.3 yppl better than average. That unit may be without defensive backs Richard (has missed the last 3 games) and Hill (missed the last 2) but they got DE Elijah Alston back from injury in their last game and it appears as if all the vital defensive players are playing – although Miami didn’t release a depth chart.

Miami is clearly the better team and the line is lower than it should be because it’s possible that Ward and/or Restrepo may not play. The math comes out with Miami by 6.4 and a total of 64.4 points (with -1.2 points for weather included). I am going to Lean Over 58 or fewer points and hope that those two do play.

Game Analysis view matchup stats

Miami-Ohio (-2.5)  23   Colorado State  22

The Miami-Ohio defense allowed just 18.9 points per game and only 4.8 yards per play while facing teams that would combine to average 5.3 yppl against an average defense. The Redhawks will be without star CB Raion Strader, who transferred to Auburn, and backup corners Yharbrough and Hardrick have also left the team. Strader had 19 passes defended and he’s worth 1.5 points by himself. Overall, the cluster losses at the cornerback position are worth 2.1 points but it could be more given the inexperienced backups that will probably be forced to play.

Colorado State’s defense was better than expected this season, as the Rams allowed just 24.3 points per game and 5.8 yppl (against teams that would average 5.5 yppl against an average defense). The Rams will be without leading tackler LB Buom Jock, who has entered the transfer portal and is not on the depth chart. However, starting DE Mukendi Wa-Kalonji is on the depth chart despite stating his intentions to enter the portal, and I’ll assume he’ll play. I’ll also assume the DT Cam Bariteau will be back on the field after missing the season final a few weeks ago. The absence of Jock and rotational DL Gatkuoth (4 TFL) is worth a point.

The Colorado State offense averaged 6.0 yppl, against teams that would allow 6.0 yppl, but they’ll be without 3 of their 4 best receivers, as star Tory Horton is still injured while Jamari Person and Caleb Goodie have entered the transfer portal and are not on the depth chart. Goodie is big loss out of the two transfers, as he averaged 20.8 yards per reception 10.9 yards per target while the most experienced receiver available, Armani Winfield, averages just 5.5 YPT despite leading the team in targets with 61. Dane Olson has shown that he might be able to fill in nicely as the deep threat in place of Goodie, as he’s averaged 20.8 yards per catch and 15.0 YPT on his 15 targets. In all, the rest of the receiving group combined for 7.6 YPT while the departed trio combined for 8.6 YPT. I valued the missing receivers at 1.2 points.

Miami-Ohio only scored 22.6 points per game, but the Redhawks’ big play offense averaged 6.2 yppl (against teams that would allow 6.0 yppl to an average team). Miami will be without their top two wide receivers Tracy and Virgil and starting LT Jados. Tracy and Virgil combined for 1634 yards and 9.5 yards per target, which is significantly better than the 7.9 YPT averaged by the other 3 wide receivers and it also means that at least one inexperienced receiver is likely to see the field. Overall, I rate the Miami offense at 0.3 yppl worse than average with current personnel (about a 3 point adjustment).

After all the adjustments I get Miami-Ohio by 1.7 points with a total of 45.1 points (with +1.2 for the good weather expected in Tucson). The Redhawks also apply to a negative 12-33-1 ATS bowl situation that applies to teams that lost in the conference championship game.

Game Analysis view matchup stats

Strong Opinion – NC State (-7)  37   East Carolina  24

I would have favored NC State by 8 points if each team had the available roster that they neded the season with but East Carolina’s transfers and opt outs are damaging to the Pirates while NC State’s lone loss, WR KC Concepcion, should make the Wolfpack offense better.

East Carolina finished the regular season 4-1 after firing their head coach and naming Blake Harrell the interim coach and he was rewarded with the full-time job. The Pirates were a below average team on both sides of the ball, rating at 0.2 yppl worse than average on offense and 0.3 yppl worse than average on defense while also being -7.5 in play differential.

The ECU offense got better when Katin Houser took over at quarterback, as he is 0.6 yards per pass play better than Jake Garcia (adjusted for opposing pass defenses faced) while also throwing fewer interceptions (Garcia has 3 more picks on 35 fewer passes). The offense will miss receiver Chase Sowell, who averaged 11.9 yards per target in 9 games against FBS opponents. Leading pass catcher Winston Wright Jr. is also out but his 7.4 yards per target will not be tough to replace. However, Wright is also an All-American kick returner (26.8 KR average) and his special teams value is worth 0.4 points. I rate the Pirates’ attack at 0.3 yppl worse than average with Houser but without Sowell.

NC State’s defense is 0.5 yppl worse than average but all the players that started the second half of season are expected to play as potential opt-out Davin Vann (6.5 sacks and 14 total tackles for loss) is listed on the freshly released depth chart. LB Caden Fordham, who started the first 6 games, is still out with an injury but the defense was better in the 6 games without him (0.2 yppl worse than average) – although I’ll use the season rating in my model.

NC State’s offense was 0.3 yppl better than average in FBS games this season and quarterback CJ Bailey did a good job filling in for injured veteran transfer Grayson McCall. The Wolfpack offense will be without leading pass catcher KC Concepcion but that is addition by subtraction, as maybe they’ll stop running all those ridiculous short routes and let Bailey throw the ball down the field more. Concepcion averaged only 7.7 yards per catch against FBS competition and a horrible 4.4 yards per target with less than a 40% success rate. The rest of the wide receivers combined to average 8.5 yards per target and a 50% success rate so not forcing the ball to Concepcion should be a positive, as he managed to catch only 57% of the passes thrown to him, which is particularly bad when considering how short most of those throws were. I project the NC State pass attack to be 0.6 yards per pass play better without Concepcion, who also averaged just 1.9 yards on 19 runs this season.

The East Carolina defense will be without 2 starters in CB Brown-Murray, who led the team in passes defended, and LB Barker, who led the Pirates in total tackles for loss with 12. Those are significant losses, and I rate the ECU defense at 0.6 yppl worse than average heading into this game.

I favor NC State by 13 points with 61.5 total points and NC State is a Strong Opinion at -7 -115 odds or better.

Game Analysis view matchup stats

1-Star Best Bet – *Brigham Young (+3 -105)  28   Colorado  24

Colorado’s offense isn’t as good as they are perceived to be, as they were just 0.5 yards per play better than average this season (6.2 yppl against teams that would allow 5.7 yppl to an average team). Starting right tackle Phillip Houston missed the final regular season game and his not listed on the depth chart, which is a slight negative for an otherwise healthy group with no opt outs or transfers reported.

BYU’s defense has a significant advantage in this game, as the Cougars allowed just 20.1 points per game and 5.0 yppl while facing a schedule of teams that would combine to average 6.1 yppl against an average defensive team. The Cougars will be missing one starter in safety Crew Wakley but he is nothing special and projects to be worth just 0.5 points.

BYU’s offense was 0.8 yppl better than average in the regular season and 11 of the 12 starters will be playing in this game with the exception being the right tackle Keim, who is injured along with his backup. The offense moved the ball better than normal against Houston with the 3rd-string RT playing (relative to the level of Houston’s defense), but I will dock the offense a bit more than I penalized the CU offense for being without their right tackle.

The Colorado defense was 1.2 yppl better than average in the regular season but I project that unit at 1.0 yppl better than average with the transfer of LB Nikhai Hill-Green, who was 2nd on the team in total tackles, 2nd on the team in tackles for loss and also defended 6 passes (2 being interceptions), which is very good for a linebacker. I value Hill-Green at 1.2 points.

BYU was a better team than Colorado in the regular season from the line of scrimmage and the Cougars are better in special teams while projected turnovers are basically even in this matchup. And the adjustments for this game slightly favor the Cougars,

As far as wins and losses, both teams lost to a hot Kansas team near the end of the season (BYU by 4 points and CU by 16 points) and BYU’s only other loss was by just 5 points to Big-12 Champion Arizona State while Colorado’s other two losses were to a mediocre Nebraska team by 18 points and to Kansas State by 3 points.

Colorado was better at turning yards into points, however, because BYU was -6.3% in 3rd-down conversion percentage while the Buffaloes were +3.6% on 3rd-downs. Colorado was also better in points per redzone opportunity, averaging 5.1 PPRZ on offense and just 4.2 PPRZ on defense, while BYU was even in PPRZ differential. The question is how much of the 3rd-down difference and redzone efficiency difference is due to variance and how much is real, and thus likely to continue.

Colorado projects to be 6.2% better in offensive 3rd-down conversion percentage because of the huge advantage in quarterback completion percentage and the actual difference is 6.1% (41.3% for CU and 35.2% for BYU). However, there is no explanation for why BYU’s defense has allowed opponents to convert on 41.6% of 3rd-downs given that they defend the run well and have allowed just 54.5% completions. The 37.7% conversions that Colorado’s defense has allowed is what would be expected given their defensive stats (37.9% expected), but BYU is better than Colorado in every defensive statistic that projects 3rd-down conversions and the Cougars should be allowing 35.3% 3rd-down conversions rather than 41.3%. That’s why there is a discrepancy between a model based on a projected box score (which is what my model is) and a model based on points, which would favor Colorado by 2.6 points.

BYU is projected to outgain Colorado 368 yards at 5.7 yppl to 348 yards at 5.4 yppl (adjusted for playing in a dome) and the Cougars, as previously mentioned, have an advantage in special teams while turnovers are projected to be only 0.1 in favor of the Buffaloes. The math model based on the projected box score and special teams ratings favor BYU by 5.1 points, but there is a chance that the Cougars continue to be just as bad in 3rd-down defense as they’ve been. I think a fair line is BYU by 2 points and the Cougars apply to my best bowl situation, which is now 79-15-1 ATS with 2 wins this year already (both dogs winning straight up).

I’ll take BYU in a 1-Star Best Bet at +3 -120 odds or better.

Game Analysis view matchup stats

Army (-14)  27   Louisiana Tech  13

Louisiana Tech got invited to this bowl game after Marshall dropped out due to lack of players and the Bulldogs are excited to be playing in this game. A good number of players did not return to the team but Louisiana Tech has one of the nation’s best defensive coordinators and I suspect that they’ll do a decent job defending Army’s option attack.

Louisiana Tech defensive coordinator Jeremiah Johnson has never had a worse than average run defense in 18 seasons as a defensive coordinator – most of those being at Northern Iowa, which had a better than average run defense on an FBS scale even though they had FCS talent. Johnson even got Kent State’s normally horrible defense to be better than average defending the run in his one season there in 2022 and he took a Bulldogs’ defense that had been at the bottom of the nation in run defense the previous two seasons and turned them into a unit that allowed just 4.2 yards per rushing play (to teams that would combine to average 4.7 yprp against an average defense). That unit is down a couple of impact starters in DT David Blay (led team in sacks and total TFLs) and LB Kolbe Fields (#2 in TFL and in total tackles) along with 5 other contributors to the defense who had a combined 11 starts. I value those players as worth 0.6 yards per play, which equates to 4.1 points in this game.

Army’s offense averaged 31.4 points per game and 6.2 yppl (against teams that would allow 6.1 yppl to an average attack) but the Black Knights did lose top back Kanye Udoh to the transfer portal (he signed with ASU). Udoh ran for 1117 yards at 6.2 ypr in the regular season and I value him as worth 1.4 points. Army is projected to gain 351 yards at 6.0 yppl against a weakened but still well coached Louisiana Tech defense.

The Louisiana Tech offense is led by Evan Bullock, who completed 67.4% of his passes and threw only 2 interceptions on 273 passes but sits in the pocket too long and takes too many sacks instead of throwing the ball away. Bullock averaged only 5.5 yards per pass play (against teams that would allow 7.2 yppp to an average quarterback) and he’ll be without top WR Tru Edwards, who entered the transfer portal and wasn’t practicing with the team in the early bowl practices (I’ll assume he’s still not with the team). Edwards averaged 9.0 yards on his 100 targets and easily led the team with 897 receiving yards and is worth 1.2 points. Fellow receivers Lewis and Jackson are also missing from the depth chart. Those two combined for 8.3 YPT on 53 targets while the available wide receivers combined for just 6.8 YPT and are worth another 0.8 points. Bullock will likely look for TE Eli Finely more and he averaged 8.5 yards per target. The top 3 running backs are on the depth chart but they combine for just 4.5 ypr (against a schedule that would allow 5.5 ypr to an average back) and the Bulldogs are down one starting offensive lineman. Louisiana Tech’s offense projects to be 1.8 yppl worse than average but Bullock doesn’t put the ball in harm’s way often, which will help a bit. Louisiana Tech also does not have their place kicker with them, which will hurt on special teams because he booted 51 of 56 kickoffs for touchbacks.

Army’s defense was 0.1 yppl worse than average in the regular season, allowing 5.4 yppl to teams that would combine to average 5.3 yppl against an average team and they were worse over the last 4 games without edge rusher Kyle Lewis, who was #2 on the team in sacks with 4 in 9 games (he’s still out and worth 0.5 points).

After all the adjustments the math favors Army by 13.8 points with a total of 39.9 points.

Game Analysis view matchup stats

Game Analysis view matchup stats

Alabama (-13.5)  28   Michigan  16

I would have favored Alabama by 9 points on a neutral field over Michigan at the end of the regular season, but the injuries, transfers and opt-outs have hurt the Wolverines more than they’ll likely hurt Bama.

Alabama’s offense averaged 6.6 yards per play (against FBS teams that would allow 5.3 yppl to an average team) but that unit struggled down the stretch against good defensive teams, averaging just 4.4 yppl against South Carolina, Tennessee, and Oklahoma in the second half of the season. The Tide will be without 5 wide receivers, but they’ll have stars Ryan Williams and Germie Bernard, who combined for 1571 yards on 149 targets (10.5 YPT) while the unavailable receivers combined for just 475 yards on 61 targets (7.9 YPT). A couple of inexperienced receivers will have to fill in at times, but they should produce at the same level as the missing receivers and the tight end group is intact and averaged 10.9 YPT.

Michigan’s defense was 1.2 yppl better than average this season (5.0 yppl allowed to teams that would combine to average 6.2 yppl against an average team), but that unit will be missing 3 starters along the defensive line in Mason Graham, Josaiah Stewart, and Kenneth Grant, who are all headed to the NFL and are not expected to play in this game. I value those 3 as worth 3.5 points.

Michigan’s offense struggled this season with just 22.3 points per game and 4.9 yards per play, but they were at least decent when Davis Warren was at quarterback. Warren started the season at quarterback but was replaced by Alex Orji and then Jack Tuttle. Those experiments failed and Warren reclaimed the job in week 9. The offense rated at average for an FBS team with Warren at quarterback, which is obviously bad for a Big 10 team, and now they’ll be without their top two running backs, star TE Loveland, and their starting left tackle, who have all declared for the NFL draft and are not expected to play. Top WR Tyler Morris is also expected to miss this game after entering the transfer portal. Morris and Loveland combined for just 830 yards on 128 targets (6.5 YPT), but that’s better than the remaining wide receivers and tight ends, who combined for just 639 yards at 5.4 YPT. I value the losses to Michigan’s offense at 3.1 points.

Alabama’s defense allowed just 18.2 points and 5.0 yppl against FBS opponents and was 0.9 yppl better than average after adjusting for opposing offenses faced. The defensive back 7 has been hit hard by injuries and defections, as the Tide will be without starting CB DeVonta Smith, All-American S Malachi Moore, starting S Keon Sabb (injured in week 8), and linebackers Robinson and Lawson, who are both injured. I value those losses as worth 3.3 points to the Bama defense.

Michigan will be without their punter Tommy Doman, who also did a good job on kickoffs, and I value him as worth 0.9 points in field position, which enhances Alabama’s special teams advantage in this game despite Michigan having the much better place kicker.

My math favors Alabama by 12.8 points with a total of 44.4 points and Michigan applies to a 32-11-2 ATS big bowl underdog situation.

Game Analysis view matchup stats

Washington (+2.5)  27   Louisville  26

Washington was better this season than their 6-6 record, as the Huskies lost two games that they should have easily won based on the stats (Washington State and Rutgers) and averaged 6.0 yards per play while allowing just 5.1 yppl to a schedule of teams that was a combined 0.6 yppl better than average. My math model would have favored Louisville by just 4.6 points on a neutral field at the end of the regular season but the adjustments for opt-outs, transfers and injuries favor the Huskies in this game.

Louisville’s offense averaged 34.3 points and 6.7 yards per play against FBS teams that would combine to allow 5.3 yppl to an average team, but the Cardinals will be without starting QB Tyler Shough and two of their top 3 wide receivers. The missing receivers (Brooks, Huggins-Bruce) combined for 1160 yards on 111 targets against FBS teams (10.5 YPT) and the available wide receivers in this game combined for 1025 yards on 105 targets (9.8 YPT), which equates to 0.24 yppp and 0.7 points. Of course, the big loss is Shough, who rated at 1.4 yards per pass play better than average and threw just 6 interceptions. Veteran Harrison Bailey, who started a few games for NC State a few years ago, is the backup and I’ll assume he plays at the same level as Jack Plummer did last season in coach Brohm’s system, which is 0.7 yppp better than average.

On a positive note, the Cardinals should be able to run the ball well, as back Isaac Brown and Duke Watson combined for 1379 yards at 7.1 ypr (vs FBS team) and won’t be sharing runs with departed backs Keyjuan Brown and Donald Chaney (356 yards at 4.0 ypr). I project more yards per run than yards per pass play and I suspect Brohm will use his two stud running backs to lead the offense. In all, I don’t think that Louisville’s offense will be hurt that much if they commit to the run, as Washington was only 0.2 yprp better than average defending the run while being 1.6 yppp better than average against the pass.

Washington’s defense was 1.0 yppl better than average overall this season and that unit is basically intact, as just a handful of players that were injured early in the season are listed as out for this game. I will adjust 0.2 yppp for injuries to pass rushers Durfee and Davis, who combined for 5.5 sacks early in the season and are among the players that won’t play.

Washington’s offense averaged only 22.5 points, but the Huskies averaged close to 400 yards of offense per game at 6.0 yppl and should have scored more than they did given that they also averaged less than one turnover per game. Coach Fisch decided to turn to Demond Williams to quarterback his offense in the season finale against Oregon and he’s sticking with the youngster over veteran Will Rogers to start this game. Williams played at least one series in every game, and he completed 77% of his 73 passes with zero interceptions. However, he also took 16 sacks, including 10 against Oregon. Williams’ compensated passing metrics were not as good as Rogers, but he adds to the run game and was the team’s second leading rusher. Overall, I rate Williams as about half a point worse than Rogers after also taking into account that Williams rarely throws into coverage (just 4 of his 73 passes were defended against with zero intercepted).

Louisville’s defense was 0.6 yppl better than average this season (5.6 yppl allowed to FBS teams that would combine to average 6.2 yppl against an average team) but the Cardinals will be without impact starters DE Gillotte and CB Riley, who both opted out to prepare for the NFL draft process. Gillotte led the team in sacks and total tackles for loss while Riley is a shutdown cornerback that had 15 passes defended in just 10 games. Rotational DT Dawson is also out, and he was #2 on the team in sacks and #3 in total TFLs. The Cardinals’ pass defense was 0.3 yards per pass play better than average this season, but they gave up 9.0 yppp in two FBS games without Riley to teams (ND and SMU) that would combine to average 7.6 yppp against an average defense. The losses to Louisville’s defense are worth 0.5 yppl and 3.7 points.

After all of the adjustments, the math favors Louisville by 1.8 points with 53.2 total points – even after lowering the total number of plays by 7 to account for the slower paced offense with Williams at quarterback for Washington (partially because more of his pass plays end in sacks and fewer pass plays end with incomplete passes). Washington applies to a 71-28-2 ATS bowl situation that plays on low scoring teams while Louisville applies to a negative 30-95-2 ATS bowl situation that applies to teams that averaged 34 or more points per game. It should also be noted that ACC teams are 0-10 ATS so far with the only straight up win being Syracuse (as an 18.5-point favorite) against a depleted Washington State team. Washington would be a Lean at +3 or more.

Game Analysis view matchup stats

South Carolina (-10)  29   Illinois  19

South Carolina’s offense started slowly under new starting quarterback LaNorris Sellers, as he tallied just 82 yards on 27 pass plays against Old Dominion in the opening game. Sellers got batter as the season progressed and averaged 6.7 yppl against FBS teams that would combine to allow 5.8 yppl to an average quarterback. Sellers also ran for 786 yards at 7.0 yards per run (not including sacks) in 10 games against FBS opponents. More of the offense will be on Sellers in this game with RB Rocket Sanders opting out. Sanders only averaged 4.8 ypr but the two backs taking his place in this game combined for just 3.7 ypr on 81 runs against FBS teams and Sanders also contributed to the pass game with 277 yards on 25 targets against FBS teams. I rate the Gamecocks’ offense at 0.3 yppl better than average without Sanders, who is worth a total of 2.6 points to this offense relative to his backups.

The Illinois defense is nothing special, as the Illini yielded 5.7 yppl against FBS teams that would combine to average 6.0 yppl against an average defense. However, I’ve adjusted for the negative outlier against Purdue (536 yards allowed at 9.2 yppl) and rate the Illini stop unit at 0.5 yppl better than average. The only possible absence of note would be LB Seth Coleman, who missed the regular season finale with a shoulder injury. But I’m going to assume he’ll be able to play with a month off to heal.

The Illini offense was 0.3 yppl better than average in the regular season, averaging 5.8 yppl against FBS teams that would combine to allow 5.5 yppl to an average attack. The run game improved with an injury to RB Kaden Eagin, who averaged just 3.9 ypr on 51 runs in 4 games against FBS competition. His injury gave Aidan Laughery and Josh McCray more touches and those two combined to average 5.4 ypr for the season against FBS teams while getting the large majority of their runs against Big-10 opponents. The Illinois rushing attack was 0.3 yards per rushing play better than average for the season but was +0.8 yprp after the injury to Eagin. The big loss for the Illini attack is star WR Pat Bryant, who has opted out of playing in this game to turn his attention towards the NFL. Bryant averaged 11.8 yards on his 78 targets against FBS opponents while the rest of the wide receivers combined to average 7.9 YPT and Zakhari Franklin, the other highly targeted WR, averaged only 7.0 YPT on 84 targets. The absence of Bryant equates to 0.9 yards per pass play and 2.8 points.

The South Carolina defense yielded just 18.4 points per game and 4.9 yards per play to FBS teams that would combine to average 6.1 yppl against an average defense. That unit is obviously worse without DE Kyle Kennard, who led the team with 11.5 sacks (but didn’t make many tackles against the run). There is still plenty of talent along the Gamecocks’ defensive front, but Kennard is worth 1.2 points based on my algorithm.

After adjusting for the players that are out, I get South Carolina by 10.3 points with a total of 49.1 points. Pass.

Game Analysis view matchup stats

Lean – Over (59.5) – Baylor (-4)  35   Louisiana State  31

Baylor was a good team based on their metrics when they were sitting at 2-4 for the season and the Bears proved the math right as they won their last 6 games while going 5-0-1 ATS. The math would have favored Baylor by 2 points at the end of the regular season and favors them by more after factoring in the players not playing for LSU.

Baylor’s offense struggled the first two weeks of the season with Dequan Finn at quarterback, but the attack took off when Sawyer Robertson was inserted at quarterback in week 3. In the final 10 games of the regular season the Bears averaged 35.9 points per game on over 450 yards of offense at 6.7 yards per play (against teams that would allow 5.5 yppl to an average team). The only players absent from the depth chart are Finn and wide receivers Hal Presley and Ketron Jackson, whose combined 7.5 yards per target on 74 targets was lower than the rest of the receivers, who combined to average 9.0 YPT.

The Tigers allowed just 24.0 points per game against FBS teams, but they weren’t that good. LSU’s defense was only 0.3 yppl better than average this season, surrendering 6.1 yppl to teams that would combine to average 6.4 yppl to an average team. The defense did a good job defending the pass (6.3 yppp allowed to quarterbacks that would combine to average 7.3 yppp against an average defense) but the Tigers struggled to stop the run – allowing 5.9 yards per rushing play (to teams that would average 5.5 yprp). Baylor has two good running backs that combined for 1612 yards at 6.0 ypr and the Bears should top 6.0 yards per run in this game. LSU is thin in the secondary, as starting safeties Sage Ryan (transferred) and Major Burns (opt out) aren’t playing while backups Allen and Jackson are injured. That status of NFL bound LB Greg Penn is up in the air, as there’s been no word on if he’ll opt out or not. I value the LSU defections as worth 2.4 points (not including Penn, who I assume will play for now but has been injured and could sit out).

LSU’s offense wasn’t nearly as good this season as it was with Jaylen Daniels in 2023, but the Tigers were still 1.0 yppl better than average offensively with Garrett Nussmeier (6.1 yppl against teams that would allow 5.1 yppl to an average team). Nussmeier is returning for another season, but he’ll be missing 3 of his top 4 pass-catchers in this game, a WRs Kyren Lacy and CJ Daniels have opted out along with TE Mason Taylor. Losing Taylor shouldn’t be much of a problem, as his 6.7 yards per target on 76 targets against FBS teams is not hard to replace. Lacy and Daniels both averaged 8.4 YPT against FBS teams on a combined 144 targets. Aaron Anderson, who averaged 10.2 YPT on 71 targets, will see more balls coming his way, which should offset the losses, and Chris Hilton, who missed the first half of the season, should be healthier now and ready to take on a bigger role. Hilton averaged 8.7 YPT on just 15 targets but has the talent to replace Daniels’ 8.4 YPT. I really don’t see much of a drop in production from the receiver group, which tends to have quality depth, as Lacy and Daniels were nothing special and Anderson is by far the best of the group and will get more targets. The avalable receivers combined for 8.8 YPT on 120 targets, which is better than the guys that have left the team. The offense will be missing both tackles, which could become an issue in pass protection. There is good talent waiting to play but depth now becomes an issue, and I rate the losses of the two offensive linemen at 1.7 points.

The Bears’ defense will likely be without part-time starting safety Corey Gordon, who is in the transfer portal, but his role would have been as a backup with starter Carl Williams back healthy after missing the final 3 games (he’s been spotted practicing) and is atop the bowl depth chart. Steve Linton, who is tied for 2nd on the team in TFL and missed the last couple of games, is also back at practice and expected to play. That unit rats at 0.4 yppl better than average and should get back to playing at a good level after struggling in their final against Kansas with some starters out.

The math favors Baylor by 5.9 points with 65.8 total points (6.9 and 66.6 total points if LSU LB Penn doesn’t play) but LSU applies to a 63-26-2 ATS bowl situation. I’ll pass the side but Lean Over at 60 or less.

Game Analysis view matchup stats

Game Analysis view matchup stats

Note: I released this game when ASU was a 14 point underdog. The Sun Devils are just a Lean at lean than +14.

Strong Opinion – Arizona State (+14)  20   Texas  28

Arizona State is 11-2 ATS this season, including 6 straight spread wins entering this game, and the Sun Devils are still underrated. Everyone knows about dual threat running back Cam Skattebo but it’s the ASU defense that continues to be underrated.

Arizona State has yielded just 20.8 points per game and only 5.3 yppl to a schedule of opponents that would combine to average 6.2 yppl against an average defensive team. Texas scored a lot of points early in the season against worse than average defensive teams (48.5 ppg against CSU, UTSA, ULM, and Miss State) but the Longhorns have averaged just 28.1 points and a modest 5.8 yppl in their other 10 games against average or better defensive teams. Those opponents would combine to allow 5.1 yppl to an average offense and the Arizona State defense rates at 0.2 yppl better than that. I still included those early season games for Texas in my model, and I project 28 points on 396 yards at 6.0 yppl for the Longhorns while taking into account the perfect dome conditions (assuming WR Isaiah Bond will play, although he may not).

The strength of the Longhorns is obviously their defense, which has yielded just 13.3 points and 4.2 yppl to teams that would combine to average 6.0 yppl against an average team. That defense showed some chinks in the armor last week against Clemson (417 yards at 6.2 yppl), but I expect that unit to play at their normal strong level. Arizona State has a unique offense and a unique player in Skattebo, who not only ran for 1568 yards at 6.0 ypr with 19 touchdowns but also had 506 yard receiving at 11.8 yards per target. Skattebo missed one game, and QB Leavitt missed the Cincy loss, but the Sun Devils averaged 6.8 yards per play (against teams that would allow 5.7 yppl to an average team) in the 11 games in which Leavitt and Skattebo both played while scoring 22 points or more in all of them. ASU didn’t face a defense as good as the one they’ll face in this game (as there are only a couple in the nation as good as Texas) but are playing their best brand of offense recently and just raced up 45 points and 8.7 yppl against a good Iowa State defense in the Big 12 Championship game. Top WR Jordyn Tyson is out with injury but his 9.7 yards per target isn’t that much higher than the 8.9 YPT of the next 3 wide receivers and the Sun Devils just played at an elite level offensively against Iowa State without Tyson. I still docked the offense 0.3 yards per pass play, which equates to 0.6 points. My model projects 324 yards at 5.0 yppl, which should be enough to stay within a couple of touchdowns of a Texas offense that is unlikely to put up a large number of points against a good ASU defense.

Arizona State is a Strong Opinion at +14 -115 odds or better. Lean on ASU at +12 or more.

Game Analysis view matchup stats

Ohio State (-2.5)  27   Oregon  26

Ohio State’s offense averaged 36 points and 7.1 yards per play (with QB Howard in the game) against a schedule of opponents that would allow 5.5 yppl to an average team. I rate that unit a bit worse for the season (at +1.5 yppl) after adjusting for the outliers against Marshall (10.5 yppl) and on the low side against Michigan (4.3 yppl). The Buckeyes had one of their best performances in the first meeting with Oregon, averaging 7.1 yppl against a Ducks’ defense that would yield just 4.5 yppl at home to an average FBS attack. However, their performances against other very good defenses were mixed (great against Tennessee but just 5.8 yppl against Penn State and Indiana).

Ohio State has lost two elite offensive linemen in LT Josh Simmons (after 6 games) and C Seth McLaughlin. The run game got a bit worse when Simmons went down, and the Buckeyes have averaged just 4.1 yards per rushing play in 3 games without McLaughlin (although vs 3 teams that would allow just 4.0 yprp to an average team). I don’t think the run game is as bad as that, but I do value the OL losses as worth 0.3 YPRP, which equates to 0.9 points. The pass protection and pass attack have been equally good without those two linemen.

The Buckeyes should still be able to run the ball against an Oregon defense that was just 0.1 yards per rushing play better than average defending the run (5.2 yprp allowed to teams that would average 5.3 yprp against an average defense) and the Ducks just gave up 303 ground yards at 9.2 yprp to Penn State in the Big 10 Championship game. That’s obviously an outlier, but I project 172 rush yards at 5.4 yprp and 387 total yards at 6.2 yppl for the Buckeyes in this game (adjusted for good weather expected, just 3 to 4 mph winds).

Oregon’s offense was just as good as the Ohio State offense, as the Ducks averaged 36 points on 6.8 yppl while facing teams that would allow 5.4 yppl to an average team. The passing numbers weren’t as good in the two games without top WR Tez Johnson and I’ve adjusted for that in my model. The Ohio State defense is the best in the nation, as the Buckeyes yielded just 11.4 points per game and 4.2 yppl to teams that would average 5.9 yppl against an average defense. However, Oregon torched that defense for 32 points on 496 yards at 7.6 yppl. That’s likely not going to happen again, as I project 377 total yards at 5.8 yppl for the Ducks in this game.

Oregon is a bit better in special teams and the math favors Ohio State by 1.1 points with 53.0 total points. The opening line was also Ohio State by 1 and has gone up a bit. Despite what appears to be a bit of value on the Ducks my feeling is that Ohio State is the better team, and I want no part of this game.

Game Analysis view matchup stats

Notre Dame (+1)  23   Georgia  22

Georgia’s offense averaged 6.1 yppl against FBS teams that would combine to allow just 5.1 yppl to an average attack, but the Bulldogs are without two-year starting quarterback Carson Beck, who was injured on the last series of the first half in the SEC Championship game against Texas. Gunner Stockton led the team in the second half and the Bulldogs ran the ball more with Stockton behind center (24 runs and 22 pass plays) while averaging just 4.5 yards per play. That’s not bad considering that Texas would allow just 4.0 yppl to an average offense, but it’s 0.5 yppl worse than Georgia’s season offensive rating. I expect the offense to better after weeks of Stockton practicing with the first team and finding plays that he’s comfortable with, as the coaching staff did not have him throwing the fall down the field against the Longhorns. Stockton was 12 of 16 passing, but he averaged just 5.9 yards per completion. I’m sure the coaches realize that short passes and the run game are not going to beat Notre Dame, and I expect a more normal offense from the Bulldogs in this game – although with more running than normal given that Stockton likes to run with the ball (6 runs for 26 yards in one half against Texas). It’s also relatively easier to run against the Irish, especially with one of their star defensive tackles out, than it is to throw against their elite secondary.

Stockton doesn’t have enough of a track record to assess his ability based on his stats and I’m certainly not going to base his passing level on 18 passes for just 61 yards against Texas given that the staff were extremely conservative with the plays they were letting him run. I will lower Georgia’s pass play rating by 0.5 yards per pass play, which is about normal for a backup quarterback with similar recruiting rating as the starter. I do expect more running from the Bulldogs in this game and a much slower pace of play, as Georgia averaged just 1.80 plays per minute of possession with Stockton in the game against Texas. I don’t think the pace will be that slow, but it should be slower (I’ve lowered the projected total plays in the game by 5).

Notre Dame’s defense is one of the best in the nation with an elite defensive backfield that resulted in a pass defense that was #2 in my compensated yards per pass play ratings (4.5 yppp allowed to quarterbacks that would combine to average 6.6 yppp against an average defense). That Irish trust their defensive backs enough to stack the box to help out the run defense, which was 0.7 yards per rushing play better than average. However, the Irish will be without DT Rylie Mills, who led the team with 7.5 sacks, which is a lot for an interior lineman. I value Mills at 1.7 points but the Irish still have stud Howard Cross III in the middle of that line and quality linebackers. I project 339 yards at 5.3 yppl for Georgia in this game (adjusted for dome conditions).

Notre Dame’s offense averaged 38.9 points and 6.6 yards per play this season against teams that would allow 5.5 yppl to an average attack. However, the Irish racked up huge offensive numbers against a few mediocre to bad defensive teams and they were relatively worse against better defensive teams. Adjusting for that leads to a rating of +0.8 yppl for this game and they were +0.9 yppl in their playoff win over Indiana.

The key to beating the Irish is to stop the run and force Riley Leonard to beat you with his arm. However, the Irish have an elite rushing attack (RB Love average 7.4 ypr) and Georgia isn’t as good defending the run (0.7 yprp better than average) as they are defending the pass (1.2 yppp better than average). Overall, the Georgia defense was 0.9 yppl better than average but the Bulldogs have had a very low interception percentage if 5 seasons under defensive coordinator Glen Schuman and they had just 8 interceptions in 12 games against FBS teams this season. Riley Leonard isn’t likely to turn the ball over against that defense (Leonard has thrown just 6 interceptions in 13 games). I project 349 yards at 5.9 yppl for the Irish in this game (adjusted for playing in a dome).

The math favors Notre Dame by 2.2 points with 45.5 total points, but Kirby Smart has a very good history in post-season games (12-6 ATS), including 8-1 ATS when not favored by more than 4 points (6-3 SU in those games). I still trust the Notre Dame defense, which is the best unit on the field, to control a Georgia offense with a backup quarterback and a mediocre ground game (just 5.0 yprp for the Bulldogs this season).

Game Analysis view matchup stats

Mississippi (-17.5)  34   Duke  16

The Ole Miss offense looked unstoppable early in the season against weaker competition but stumbled at times against better competition. The Rebels were 1.4 yppl better than average against FBS teams this season (7.0 yppl against teams that would combine to allow 5.6 yppl to an average attack)  I still rate the Rebels’ attack at +1.2 yppl better than average if I exclude their games against worse than average defensive teams and adjust for the outlier against Arkansas (9.1 yppl) that skewed their rating up more than their worst performance dragged it down. The likely absence of star WR Tre Harris drops that offense a bit more, although he played in only 4 of the games against better competition. Harris missed 4 of 5 games at the end of the regular season and had just 2 targets in the other. He averaged 12.3 yards on 68 targets against FBS teams while the other wide receivers combined for 10.9 YPT. The absence of RB Henry Parrish is not much of a negative, but I did adjust a bit for the transfer of starting RT Micah Pettus.

The Rebels will be challenged by a very good Duke defense that yielded just 4.9 yppl this season to FBS teams that would combine to average 5.8 yppl against an average defense. That unit should be intact for this game, with the exception of a backup cornerback, but despite how good their pass defense was this season (5.1 yards per pass play allowed to quarterbacks that would combine to average 6.5 yppp against an average team) the Blue Devils did give up 10.3 yards per pass play to SMU and 9.1 yppp to Miami – so they have been susceptible to elite quarterbacks.

Duke’s offense was sluggish this season, as the Blue Devils averaged only 5.4 yards per play against FBS teams that would allow 5.7 yppl to an average attack. Starting quarterback Malik Murphy has transferred and Henry Belin IV will get the start. Belin has one career start, against NC State last season, and he was just 4 for 12 for 107 yards passing (one big play for 69 yards) with a sack and had 4 runs for 31 yards. I’m going to assume that he’s worse than Murphy was with the typical backup vs starter adjustment, which is 2.4 points in this case. Leading rusher Star Thomas has entered the transfer portal, but I haven’t seen any statement from him about playing in this game and he is listed on the depth chart. It doesn’t really matter, as Thomas’ 4.1 ypr is not tough to replace.

The Blue Devils are going to have a tough time moving the ball on an elite Ole Miss defense that yielded just 15.2 points per game and 4.7 yppl (with starters in) to FBS opponents that would combine to average 6.0 yppl against an average defense. Leading tackler LB Chris Paul Jr. has opted out to prepare for the NFL combine, which is a big loss, and starting defensive back Jadon Canady is in the transfer portal and I’ll assume he won’t play (no official word though). Canady was second on the team with 11 passes defended, so that’s another significant loss. I value those two defenders as worth 2.1 points with most of that being Paul, who aside from being the leading tackler has 11 total tackles for loss and 5 passes defended, which is good for a linebacker.

After all the adjustments, I get Mississippi by 18.1 points with 50.8 total points. Ole Miss does apply to a negative 12-31 ATS double-digit favorite situation that applies to Alabama, who lost straight up to Michigan, so it’s possible that the Rebels may not be fully focused for this game. I’ll pass.

Game Analysis view matchup stats

Texas State (-13.5)  40   North Texas  26

Texas State’s offense averaged 37.1 points and close to 500 yards per game at 6.6 yards per play (against teams that would allow 6.2 yppl to an average team) and they should feast on a bad North Texas defense as long as quarterback Jordan McCloud is in the game. There is speculation that McCloud won’t play the entire game in order to give freshman Brad Jackson some experience. The Bobcats will be without top RB Ismail Mahdi, but his 991 yards at 5.4 ypr should be replaced adequately by Pare, Burgess and Hankins, who combined for 1000 yards rushing at 5.5 ypr with Pare and Burgess (listed as co-starters) combining for 5.9 ypr. The remaining running backs also averaged a combined 10.7 yards on 19 targets while Mahdi averaged 7.5 YPT on 26 targets. I’m making no adjustment for Mahdi not playing although I’ll make an adjustment for the transfer of RT Alex Harkey.

The Mean Green are not so mean on defense, as that unit allowed 6.3 yppl despite facing teams that would combine to average just 5.3 yppl against an average defense. Opponents with better than average passing quarterbacks (South Alabama, Texas Tech, Tulane, and Houser of ECU) combined for 10.1 yards per pass play against the Eagles and McCloud should light that secondary up.

North Texas was good offensively with former TCU quarterback Chandler Morris leading the way. Morris averaged 7.0 yards per pass play against teams that would allow 7.1 yppp to an average FBS quarterback and the run game was good (5.7 yprp against teams that would allow 5.3 yprp to an average team). Morris has transferred along with top WR DT Sheffield and TE Oscar Hammond. Sheffield averaged 8.6 yards on his 96 targets, which wasn’t much better than the 8.2 yards per target that the next 6 wide receivers averaged, and Hammond averaged 9.2 yards on his 26 targets while the other tight ends combined for just 26 yards on 6 targets the entire season. Backup quarterback Drew Mestemaker threw just 5 passes and completed 4 of them for 69 yards in mop-up duty this season. I value the Mean Green offensive losses at 5.6 points but there’s a lot of variance in that projection given the lack of data on Mestemaker.

Texas State was good defensively for s Sun Belt team and 0.1 yppl better than average on a national scale, as the Bobcats yielded just 5.2 yppl to teams that would combine to average 5.3 yppl against an average defense. A few defenders are no longer with the team, most notably DE Parker, who missed the end of the season with injury, and CB Eaton, and the total value of the missing players is projected at 3.5 points.

The math favors Texas State by 14.6 points with 66.5 total points if Texas State McCloud plays the entire game, but that may not be the case. Without knowing how much McCloud will play I prefer to pass this game. There was heavy action on the under so maybe those rumors are true.

Game Analysis view matchup stats

Minnesota (-9)  26   Virginia Tech  16

Virginia Tech is missing a ton of key players for this game but it appears as if the market has adjusted correctly and I don’t see enough value to play this game.

Minnesota’s offense was 0.1 yppl worse than average in the regular season, averaging 5.3 yppl against teams that would allow 5.4 yppl to an average team. That unit should be intact for this game aside from a couple of offensive linemen that combined for 16 starts – most notably LT Ersery (12 starts). I rate the Gophers’ attack at -0.3 yppl heading into this game barring any other late defections.

The Virginia Tech defense was 0.7 yppl better than average this season, allowing 5.2 yppl to teams that would combine to average 5.9 yppl against an average stop unit (adjusted for facing the worst of the quarterbacks against Marshall, Stanford and Georgia Tech). That unit will look nothing like the one from the regular season, as 7 starters are expected to miss this game, including All-American DE Powell-Ryland (16 sacks!). My algorithm projects the Hokies’ defense to be 7.3 points worse in this game (based on Minnesota’s projected 63 plays).

The Virginia Tech offense was a disappointment this season, as they were significantly worse than last season despite having nearly everyone back. The Hokies were just 0.2 yppl better than average, gaining 5.9 yppl against teams that would allow 5.7 yppl to an average FBS attack. All indications are that starting quarterback Kyron Drones is still out with injury and backup Collin Schlee is questionable to doubtful to play, which leaves Pop Watson in line to potentially get his second start. Watson struggled off the bench against Duke in week 13 (98 yards on 33 pass plays) but after a week of practice with the first team offense Watson was sharp in a 37-17 win over Virginia – completing 14 of 21 passes for 254 yards (10.3 yards per pass play with the 2 sacks included) while running for 65 yards on 9 runs.

I still don’t think either quarterback would be as good as Drones (based on their lower completion percentages) and the Hokies are also missing their star running back Tuten, their top two receivers (Da’Quan Felton and Jaylin Lane) and 3 starting offensive linemen. The absence of Tuten (1159 yards at 6.3 ypr) and three starting linemen should significantly hurt the offense but not having Lane and Felton could be a plus. Those two may have been first and second on the team in catches (38 and 32 receptions) but they combined for just 826 yards on 121 targets (6.8 yards per target) while the rest of the wide receivers combined for 1028 yards on 100 targets, which is 10.3 YPT. Tight end Benji Watson also averaged 7.9 YPT on 42 targets. I think not having Felton, in particular, is a plus given his horrible 5.8 YPT and 35% success rate. However, despite what I think is an upgrade at receiver, the rest of the offense is much worse, and I rate that unit at 0.4 yppl worse than average for this game.

Minnesota’s defense was elite this season, as the Gophers yielded just 17.5 points per game and 4.7 yppl against teams that would combine to average 5.9 yppl against an average team (adjusted for facing Michigan with Orji at QB). Star CB Justin Walley has declared for the NFL draft, and I’ll assume he won’t play (although he is scheduled to play in the East-West Shrine game, which sometimes is an indication of playing in the team’s bowl game too). Walley led the team in passes defended (12) despite missing two games and he is projected to be worth 1.0 points.

Overall the math favors Minnesota by 10.0 points with a total of 42.0 points. Virginia Tech does apply to a 60-16-4 ATS bowl situation that plays on teams that underperformed against the spread in the regular season. However, I don’t really think that has any bearing given that so many of the Hokies’ starters won’t be playing in this game.

Game Analysis view matchup stats

Strong Opinion – Liberty (+4)  26   Buffalo  24

The math would have favored Liberty by 7.7 points at the end of the regular season, but the Flames have been hit by defections – most notably veteran starting quarterback Kaidon Salter, who has committed to Colorado. Buffalo is intact aside from their kicker, who made 14 of 15 field goals inside of 50 years (0-1 at 50+) and transferred to Texas Tech (worth 0.6 points).

The Liberty offense wasn’t nearly as good as expected, as Salter had a horrible season throwing the football, averaging just 5.9 yards per pass play in 10 games against FBS teams that would combine to allow 7.3 yppp to an average quarterback. That shouldn’t be hard to replace (I’ll adjust 0.5 yppp) but Salter’s running (661 yards on 86 runs, not including sacks) will be tough to replace. Leading rusher Quinton Cooley (1144 yards at 6.1 ypr in 10 games against FBS teams) hurts a bit too, but backups Lucas and Blue combined for 778 rush yards at 5.9 ypr against FBS competition. Reaching that average is not likely with two starters out along the offensive line and overall offense projects to be 0.7 yppl worse than they were in the regular season (now -0.9 yppl), which equates to 5.3 points.

The Flames should still move the ball at a decent level as the Buffalo defense is 0.8 yppl worse than average (5.8 yppl allowed to teams that would combine to average 5.0 yppl against an average defense). The Bulls defend the run at a decent level (just 0.2 yprp worse than average) but Liberty QB Ryan Berger (formerly Appalachian State’s backup) should post decent numbers in the pass game, as Buffalo gave up 6.7 yards per pass play to quarterbacks that would combine to average only 5.4 yppp against an average defense.

Buffalo’s offense was also horrible, as the Bulls averaged only 5.4 yppl against teams that would combine to allow 6.5 yppl to an average FBS attack. Liberty was just 0.4 yppl worse than average defensively in the regular season but the Flames will be missing some pieces from that unit. The team’s second leading tackler LB Jerome Jolly is not with the team but he’s nothing special (55 tackles, 4.5 TFL). The defense is also without DT Nairne (7 starts and just 1.5 total TFL), injured starting safety Green (only 32 tackles with no TFL and only 2 passes defended) and backup NB Bodnar, who played just 5 games before getting injured. Most of Liberty’s best defensive players are playing in this game and I would rate Jolly (not playing) as the 5th most important defender. My algorithm still projects the Liberty defense to be 0.3 yppl worse without the missing players (2.2 points), which puts that unit at 0.7 yppl worse than average for this game, which is 0.4 yppl better than Buffalo’s offense.

Buffalo is a horrible team that was outgained by 62 yards per game and 0.5 yards per play despite facing a schedule of teams that was 12.2 points worse than an average FBS team from the line of scrimmage. Overall, the Bulls were 17 points worse than an average FBS team from the line of scrimmage this season (but were lucky in turnovers) and even with Liberty’s losses they should not be favored by 3 points or more in this game. In addition to the line value (my math model favors Liberty by 0.5 points even after adjusting 7.8 points for their absent players; and 0.6 points for Buffalo not having their accurate FG kicker). Liberty also applies to my best bowl situation, an 81-16-1 ATS situation that is 4-1 this year.

Liberty is a Strong Opinion at +3 points or more.

Game Analysis view matchup stats

Penn State (+2.5)  23   Notre Dame  22

Note: The line was +2.5 when I sent the analysis to subscribers.

The market has overreacted to Notre Dame beating a flawed Georgia team that was without their starting quarterback. The Irish have an elite defense but with the injury to star DT Rylie Mills they rate a bit worse than Penn State’s defense. The Notre Dame offense is not as good as the Nittany Lions’ attack and Irish quarterback Riley Leonard has struggled to throw the ball against better defensive teams.

Leonard has averaged just 6.3 yards per pass play this season against teams that would combine to allow 6.1 yppl to an average quarterback and he’s been relatively worse against teams with a good pass defense. Notre Dame has faced 9 teams that were at least 0.3 yards per pass play better than average defensively (Texas A&M, Northern Illinois, Miami-Ohio, Louisville, Georgia Tech, Florida State, USC, Indiana, and Georgia) and he’s been 0.1 yppp worse than average in those games – averaging just 5.6 yppp against teams that would allow 5.7 yppp to an average QB, weighted by how may pass plays he had against each team, as is always the case. That includes averaging just 4.5 yppp in playoff games against Indiana and Georgia, who would both allow 5.1 yppp to an average quarterback.

The Irish have an elite run offense that averaged 6.5 yards per rushing play against teams that would combine to allow 4.9 yprp to an average team. The run game was also relatively worse against better defensive teams and overall the Irish offense, which rated at 0.9 yards per play better than average in all games, was just 0.5 yppl better than average against better than average defensive teams (5.8 yppl against 10 better than average defensive teams that would allow 5.3 yppl to an average attack). I used regression analysis to arrive at a negative adjustment of 0.2 yppl for this game (rather than the 0.4 adjustment had I just used their games against better than average defensive teams).

Penn State’s defense is elite against the run and the pass, having yielded just 4.2 yprp, 5.5 yppp and 4.9 yppl to teams that would combine to average 5.3 yprp, 6.9 yppp, and 6.2 yppl against an average stop unit. Since defending Leonard in the pass game should not be an issue this game will likely come down to how well the Nittany Lions defend an elite Irish rush attack. Penn State has faced 4 good rush attacks this season (USC, Ohio State, Oregon, and Boise State) and they held those 4 teams to 5.2 yprp, which is 0.8 yprp better than an average team. However, that is skewed by the 8.4 yprp they allowed to USC. Penn State yielded just 4.6 yprp to the other 3 good running teams since that USC game, including 3.6 yprp against Boise last week while limiting Heisman Trophy finalist Ashton Jeanty to just 3.5 yards per run. I project 5.4 yprp for Notre Dame in this game but just 5.2 yppp for Leonard.

Penn State’s offense averaged 6.6 yppl with quarterback Drew Allar in the game (against teams that would allow 5.3 yppl to an average offense), but the Nittany Lions were also relatively worse against better defensive teams as they surpassed 8.0 yppl three times against worse than average defensive teams (W Virginia, Bowling Green, and Purdue), which skewed their offensive rating higher. As with Notre Dame, the pass attack was relatively worse against better defensive teams but overall the offense wasn’t nearly as affected by facing good defensive teams as Notre Dame’s offense was (the adjustment came out to -0.1 yppl).

Notre Dame’s defense has been 1.3 yppl better than average and their worst game was also against USC (7.2 yppl allowed). That unit played relatively well in other games against good offensive teams and overall their was very little correlation between the level of the opposing offense and Notre Dame’s defensive game ratings. The Irish are still without sack leader DT Rylie Mills, whose 7.5 sacks are 3 more than any other player on the defense. Mills is worth 1.2 points and I expect his absence to matter more in this game than it did against Georgia playing with a backup quarterback.

The math model favors Penn State by 0.7 points with 46.4 total points before applying the adjustments for each offense being relatively worse against better defensive teams. I get Penn State by 1.1 with 45.1 total points with Penn State having a clear edge from the line of scrimmage but Notre Dame being 2.3 points better in special teams (despite their bad place-kicking) and with a 0.8 points advantage in projected turnovers.

Game Analysis view matchup stats

Ohio State (-5.5/-6)  30   Texas  22

The math model, weighing all games equally, favors Ohio State by 5.5 points in this game, but the Buckeyes have been relatively better against better teams while Texas has been relatively worse against good teams because the Longhorns’ offense has been significantly worse against better defensive teams than they have overall.

Ohio State’s offense has averaged 7.2 yards per play with QB Will Howard in the game while facing teams that would allow 5.4 yppl to an average team and the Buckeyes have been even better in two playoff games – averaging 8.2 yppl with Howard in the game against Tennessee and Oregon defenses that would allow 4.7 yppl to an average attack. Ohio State did have that one bad offensive game against Michigan (4.3 yppl) when their play-calling indicated that they were playing not to lose the game rather than playing to win it. The coaching staff has obviously learned from their mistake in that game, and overall the Buckeyes’ offense has been relatively better against better defensive teams even with that Michigan debacle included. In fact, the regression equation projects that the Buckeyes would be 0.4 yppl better, relatively, against an elite Texas defense than they would overall. I decided to adjust 0.2 yppl for that.

The Texas defense is the best in the nation in compensated yards per play allowed, edging out Ohio State by 0.05 yppl. The Longhorns have given up 4.3 yppl to teams that would combine to average 6.1 yppl against an average defense and that unit was a bit better, relatively, against better offensive teams and I made a slight adjustment for that.

The Texas offense has been the problem, as the Longhorns racked up huge numbers against weaker defensive teams but have struggled, relatively, against better stop units. Texas has run 792 plays from scrimmage (not including “TEAM” plays) with quarterback Quinn Ewers in the game and they’ve averaged 6.1 yards per play while facing teams that would allow 5.1 yppl to an average offense. The offense was relatively better against weaker teams, as the slope of the regression equation to project the Texas offense (in terms of yppl) based on the defensive level of the opposing defenses (based on compensated yppl allowed), weighted by how many plays Ewers was in the game for against each team, is +1.59. That means that for every yard better or worse the defense was the Longhorns’ offense was 0.59 yppl better or worse, relatively. A team that plays on the same relative level against both good and bad defensive units would have a slope of 1.00. The equation would project a performance that is 0.4 yppl worse than normal for Texas facing an elite Ohio State defense, but I’ve decided on an adjustment of 0.2 yppl, which equates to 1.3 points.

Ohio State’s defense has allowed just 4.2 yppl to teams that would combine to average 5.9 yppl against an average defense and that unit was also a bit better, relatively, against better offensive teams even with the 7.6 yppl they allowed to Oregon in Euguene in week 7. The Buckeyes also yielded just 4.0 yppl to the Ducks last week while giving up only 3.7 yppl and 3.1 yppl, respectively, to very good offensive teams Tennessee and Indiana. Those 3 recent performances more than made up for the outlier in the first meeting against Oregon. As with the Texas defense, I made a slight adjustment for Ohio State being a bit better against better offensive units.

After the adjustments for how each unit performed, relatively, against better teams I get Ohio State by 8.4 points and 51.8 total points (adjusted for perfect dome conditions – they keep the roof closed at AT&T Stadium).

Game Analysis view matchup stats

Ohio State (-8)  28   Notre Dame  18

If all games played by each team were weighed the same the math would favor Ohio State by 5.5 points, but Ohio State has been relatively better against better teams (especially on offense) while Notre Dame has been relatively worse against better teams (particularly on offense).

For the season, the Ohio State offense was 1.8 yards per play better than average with QB Will Howard in the game, averaging 7.2 yppl against teams that would allow 5.4 yppl to an average team. However, the Buckeyes’ attack has raised their level of play in the playoffs, averaging 7.6 yppl in 3 games against defensive units that would allow 4.5 yppl to an average team and over the course of the season the Ohio State offense has been 2.0 yppl better than average in 7 games against very good defensive teams Oregon (twice), Penn State, Indiana, Michigan, Tennessee, and Texas, who are all at least 1.0 yppl better than average. The Buckeyes averaged 6.6 yppl in those games.

Notre Dame’s defense has been 1.4 yppl better than average this season, which is 0.2 yppl better than the average of those good defensive teams that the Buckeyes’ faced, but the Irish are without star DT Rylie Mills, who still has 3 more sacks than any other player on the Notre Dame defense despite missing 2 games. That unit has played well in two games without Mills, but I still value him at a bit more than 0.1 yppl and 0.8 points. The two best offensive teams that Notre Dame faced this season are Louisville and USC and the Irish gave up and average of 496 total yards at 6.5 yppl and 29.5 points per game against those potent offensive teams and Ohio State’s offense most resembles that of the Trojans, who scored 35 points on 557 yards at 7.2 yppl against the Irish and should have scored more given that they had two passes intercepted in or at the end zone in that game. This game is also being played in a dome, which elevates the passing numbers for each team and usually leads to more passing. I project 420 yards at 6.6 yppl for Ohio State.

The Notre Dame offense was 0.9 yppl better than average over the course of the season, but that unit racked up huge numbers against much less talented teams – averaging 485 yards per game at 7.6 yppl against 5 average or worse defensive teams (Purdue, Stanford, Navy, Virginia and Army). The Irish offense was 0.7 yppl better than average in 10 games against better than average defensive teams, which is think is a reasonable level to expect from a unit that clearly has some deficiencies in the pass game. Riley Leonard did throw the ball well in the semifinal win over Penn State but his also manage just 79 yards on 25 pass plays against Georgia the game before that an has generally struggled against better than average pass defenses (5.8 yppp against teams that would allow 5.6 yppp to an average QB). The injury sustained last week by LT Anthonie Knapp should also have a negative impact on that offense, as he’s projected to be worth 0.1 yppl.

I don’t think the Notre Dame offense is nearly good enough against an Ohio State defense that is elite defending the run (3.9 yprp against teams that would average 5.3 yprp against an average defense) and the pass (4.6 yppp allowed to quarterbacks that would average 6.8 yppp). Oregon, which ranks 4th best in the nation in compensated yards per play on offense, is the only team to score more than 17 points against the Buckeyes’ defense and the list of teams that did not surpass 17 points includes 3 teams that are better offensively than Notre Dame, whose average points per game was enhanced by 6 defensive touchdowns scored (against none allowed). Ohio State has only allowed more than 5.1 yppl in one game all season (the game at Oregon) and I project just 294 yards at 4.8 yppl for the Irish in this game, even with perfect conditions.

I can see the total going either way, as the math projects 47.3 total points but has ND scoring 19 points, which I don’t think will happen. I’m not recommending a play on the side but I’d rather than Ohio State if I was inclined to make a small wager on the game.