This season, Free Analysis sides were 287-273-8 overall and 99-81-2 on games with a difference or 4+ points from the spread. Totals are just 284-277-1 but they’re a profitable 37-29 on differences of 6 points or more from the line.


The record on all Free Analysis sides is now 1626-1431-62 ATS since 2013, which is very good picking nearly every game over 5 years. The Free Analysis sides in which my predicted margin was 4 points or more away from the line are now 569-473-22. Totals on the Free pages are now 1347-1294-16 in the 4-plus seasons I’ve been tracking them.


The spreads posted are the consensus lines at the time I posted each analysis.

Bowl Best Bets were 8-3 and I am now 52-35-2 on my Best Bets (75-49-3 on a Star Basis) and 22-18 on my Strong Opinions this season.

College Bowl Games

Memphis vs
Wake Forest

Sat, Dec 22 9:00 AM

Odds: Wake Forest +3.5, Total: 72.5

Game Analysis view matchup stats

Create an account to get analysis and updates sent to your inbox.

Best Bet – Under (72.5) – Wake Forest 34   Memphis  32

Strong Opinion – Wake Forest (+3.5 at -115)

Birmingham Bowl – Saturday, 9 am Pacific

I find it odd that the total on this game opened at 73 points and barely moved even though it was announced last week that the best running back in the nation, Memphis’ Darrell Henderson, was going to sit out to prepare for the NFL draft and that Wake Forest’s best offensive player, WR Greg Dortch, is also out. Henderson ran for 1909 yards at an incredible 8.9 yards per rush with 22 touchdowns – many of them long runs for quick touchdowns. Henderson’s backup Patrick Taylor doesn’t possess that sort of breakaway speed (just two runs of more than 35 yards on 178 runs this season) and his 5.7 ypr average (both this season and over his 3 years) is significantly lower than Henderson’s and not particularly impressive when you take into account that Memphis faced teams that would combine to allow 5.5 yprp to an average running back. Memphis goes from having one of the best rushing attacks in the nation (1.6 yprp better than average) to being just 0.4 yprp better than average without Henderson. The impact of Dortch should also be felt, as he lead the Demon Deacons in receiving yards with more than double any other receiver and his 8.6 yards per target was much higher than the 6.5 ypt average of the next 4 wide receivers. Dortch also was the punt and kick returner so the special teams should take a bit of a hit too. The combined value of those two players to the total is 6.3 points and the total has barely budged.

Wake Forest’s offense actually still rates better than their season rating even without Dortch because backup quarterback Jamie Newman was a significant upgrade over freshman Sam Hartman who was injured late in the season. Hartman was 0.7 yards per pass play worse than average while Newman has been 0.6 yppp better than average on his 111 pass plays this season. So, even without Dortch the Demon Deacons rate as average offensively on a compensated yards per play basis with Newman at quarterback (0.2 yppl better than their season rating). Wake is projected to move the ball at a decent 5.7 yards per play against an average Memphis defense that was better than their 31.5 points per game allowed would suggest. Memphis gave up that many points on the 5.7 yppl that they allowed because they faced a lot of fast-paced teams, which will be the case again today. Wake Forest was among the fastest paced teams in the nation for much of the season but they slowed things down with Newman was installed at quarterback, as the Demon Deacons went from averaging 3.0 plays per minute of possession time in their first 9 games with Hartman at quarterback to averaging just 2.4 plays per minute in Newman’s 3 starts, which is barely higher than the national average. I’m going to assume that the coaching staff will go back to being more up-tempo now that Newman has had a few weeks to work on that part of the offense – although it’s certainly possible that they’ll continue to play at a slower pace since the coach indicated that he slowed the pace to help his defense stay fresher after being thinned by injuries. The slower pace certainly worked better for the defense and the team and the case for the under even more compelling if Clawson continues to play at that slower pace (I assumed the average pace for the season in my math). If I use Wake’s average pace for the season then I’d project 476 yards at 5.7 yards per play for the Demon Deacons in this game.

As mentioned, the Memphis offense won’t be nearly as good without Darrell Henderson, as they go from being 1.0 yards per play better than average to 0.4 yppl better than average. I think part of the season the total on this game was high to start with is because how bad the defenses that Memphis faced this season was not fully being taken into account. The Tigers faced teams that would combine to allow 35.0 points and 6.3 yards per play to an average offensive team, which makes Memphis’ 43.6 points per game less impressive.

The Wake Forest defense seems pretty bad, as they allowed 33.3 points per game and 6.0 yards per play. However, they faced teams that would combine to average 30.7 points and 5.8 yppl against an average defense, so the Deacs were only 0.2 yppl worse than average and they gave up more points because they faced more possessions due to their fast-paced offense. Wake Forest actually wasn’t as bad as their compensated average stats would suggest as their 6.0 yppl allowed was skewed by the 667 yards at 9.5 yppl that they allowed to Clemson, including 469 rushing yards at 12.3 yprp. A change was made to the defense following that Clemson game (and after a bye week), as Luke Masterson was moved into the starting lineup as an inside linebacker. The Wake run defense was very good over the final 6 games after that change was made, as the Deacs yielded just 4.4 yprp to teams that would average 5.3 yprp against an average team. It’s unlikely that one player could make that much difference but I did make a small adjustment to the Wake Forest run defense rating to account for the variance of that Clemson game. Wake Forest was actually 0.4 yppl better than average over the final 6 games but I only made a 0.2 yppl adjustment for the new lineup and I have that unit rated as average heading into this game (rather than 0.2 yppl worse than average). The math projects 456 yards at 6.0 yards per play for a Memphis offense that will likely miss their star.

The absence of Henderson’s 8.9 yards per rush has not been properly adjusted for in either the side or the total and my math favors Wake Forest by 2 points with a total of just 66 points, even after adding 1.2 points for the good weather that’s expected today. The only bowl team that Memphis beat this season was Houston, who was without their star quarterback and All-American defensive linemen in that game. The other Tigers’ wins were against Mercer, Georgia State, South Alabama, Connecticut, East Carolina, Tulsa, and SMU, who are all terrible teams. Wake Forest and their improved defense beat NC State and Duke as double-digit underdogs in their final 3 games, so they’ve proven that they can beat a better than average team, which is something that Memphis has not proven (Houston without King and Oliver are a worse than average team). Wake Forest is a Strong Opinion at +3 points or more (-110 odds or better).

After going over the total in 5 of their first 6 games the change in the defensive lineup and the slower pace of play by the offense lead to 4 unders in Wake’s final 6 regular season games with one of those overs being in a game in which the defense allowed just 7 points. I’ll take the Under in a 1-Star Best Bet at 72 points or higher (Strong Opinion down to 71).

Houston vs

Sat, Dec 22 12:30 PM

Rotation: 223, Odds: Army -3.5, Total: 65.5

Game Analysis view matchup stats

Create an account to get analysis and updates sent to your inbox.

Best Bet – *Army (-3.5)  32   Houston  21

Best Bet – **Under (65.5)

The Best Bets on Army and Under are both mostly based on Houston being without their star quarterback D’Eriq King, who averaged an impressive 8.1 yards per pass play (against FBS opponents) and added 736 yards on 93 runs. King’s backup is freshman Clayton Tune, who has completed just 38 of his 85 passes (44.7%) despite having 23 of those attempts against horrible FCS team Texas Southern – a team that would allow over 10 yards per pass play against an average FBS quarterback. Tune took over midway through the second to last regular season game against Tulane and he completed just 24 of 58 passes (41.4%) for just 5.0 yards per pass play against Tulane and Memphis, who would combine to allow 6.7 yppp to an average quarterback. The 31 points that Houston scored in their 21 point loss to Memphis is misleading, as the Cougars averaged only 4.7 yards per play in that game. Houston has gone from having a well-balanced and very potent attack to being significantly worse than average with Tune at quarterback. I suspect that Tune may improve some with the extra bowl practices but a quarterback that completed only 45% of his passes, including just 52% against a horrible FCS defense, is simply inaccurate and I don’t think accuracy can be fixed over the course of a few weeks.

Army’s defense is 0.3 yards per play below average (5.7 yppl allowed to teams that would combine to average 6.0 yppl against an average defense) but their option offense possesses the ball for so long that opponents have fewer opportunities to score. The result is a unit that allowed just 18.0 points per game and held Oklahoma’s historic offense to just 21 points in regulation (lost by 7 in OT). If Tune plays at the level he played over those last two games (which is actually much better than if I added his compensated stats from the game against Texas Southern) then the Cougars would average only 5.3 yards per play with just 53 plays from scrimmage expected.

Army’s option attack averaged just over 300 rushing yards per game at 4.8 yards per rushing play and Kelvin Hopkins Jr. averaged 9.7 yards per pass play on the 6 to 10 passes per game that the Black Knights attempted. While the Army offense was worse than average on a yards per play basis (5.3 yppl against teams that would allow 6.2 yppl to an average team) they were so good at moving the chains and possessing the ball that the Knights averaged 20.4 more plays per game than their opponents. A big part of that was being smart about going for it on 4th down more than any team in the nation while being extremely successful in doing so (converted 31 of 36 4th downs).

Army’s option attack is likely to perform a bit better than normal against a Houston team that will be without all 3 starters along the defensive line. The Cougars began the season with one of the most dominating defensive lines in the nation, anchored by two time All-American Ed Oliver. Oliver sat out 4 games after getting injured against Navy, returned half-heartedly against Memphis in the regular season finale and has decided to bid adieu to his teammates to prepare for the NFL draft. Fellow starters Payton Turner and Jerard Carter are also out, as is #4 defensive linemen Isaiah Chambers, who has been out since mid-season. Houston’s dominating defensive front yielded only 3.4 yards per rushing play over the first half of the season (to teams that would combine to average 4.7 yprp) but then gave up 372 rushing yards at 5.8 yprp to Navy in the game Oliver was injured and surrendered an average of 276 rushing yards at 6.1 yprp in the final 5 games of the season (to teams that would average 5.5 yprp against an average defense). Houston went from having one of the best run defenses in the nation to now being worse than average after Oliver was injured. Carter was out for 3 of those games and Turner joined him on the sideline for the Memphis game in which they gave up 417 rushing yards at 7.6 yprp. Houston didn’t really perform that well against Navy’s option attack when they had all 3 defensive line starters and I suspect they’ll struggle to keep Army off the field in this game with 3 backups on the line. Army doesn’t have a lot of big rushing plays, which is another reason they have the ball for so long (38:51 average time of possession led the nation) but I project an efficient 5.3 yards per rushing play for the Cadets in this game and for them to run 23 to 24 more plays than Houston will.

Overall, I have Army at 443 total yards at 5.8 yards per play and the math predicts a 12 point win and just 48 total points. There is a chance that Tune will improve some with extra practice time but Army and the Under both look like very good plays in this game. It also doesn’t hurt that Military academies are 33-14 ATS in bowl games, including 20-4 ATS if they have a win percentage greater than .667. I suppose Air Force, Army and Navy are more disciplined in their preparation. I’ll go Under in a 2-Star Best Bet at 63 points or less and for 1-Star down to 62 points (Strong Opinion Under down to 60). I’ll also play Army for a 1-Star Best Bet at -4 points or less. That would have been a 2-Star play on Army had I not also played the Under but part of the reason I like the Under is the same reason I see value on Army (i.e Tune sucks) so the plays are correlated and 3-Stars is the most I wanted to risk. If the line on the side does not come back down to -4 or less, then Army would still be a Strong Opinion from -4.5 to -6.

Buffalo vs

Sat, Dec 22 4:00 PM

Rotation: 225, Odds: Troy +1, Total: 49.5

Game Analysis view matchup stats

Create an account to get analysis and updates sent to your inbox.

Buffalo (-1)  26   Troy  24

Dollar General Bowl – Saturday, 4 pm Pacific

My math model favors Buffalo by 1.3 points with a total of 50.1 points, so there is no value in this game. With that being the case I’ll keep my analysis relatively short.

Buffalo is led by veteran quarterback Tyree Jackson but the star of the team is future NFL WR Anthony Johnson, who averaged 18.2 yards per catch with 11 touchdowns and 11.2 yards per pass targeted at him. Johnson missed nearly two full games and the pass attack was worse than average without him but I rate the Bulls’ offense at 0.2 yards per play better than average with Johnson healthy and Jaret Patterson (946 yards at 5.6 ypr) as the main ball-carrier (he averaged just 5 runs per game over the first 5 games and 18 runs over the final 8 games).

Troy’s strength is a defense that allowed just 22.5 points per game and 5.1 yards per play and rates at 0.3 yppl better than average after compensating for opposing offenses faced. The Trojans’ run defense is very good (0.7 yprp better than average) but the pass defense is a bit worse than average and Buffalo should have good success through the air in very calm conditions that they aren’t used to playing in Buffalo. I project 399 yards at 5.7 yppl for the Bulls in this game.

Troy’s offense was 0.2 yppl worse than average for the season but an injury to highly accurate quarterback Kaleb Barker (73% completions and a 1.6% interception rate) has hurt the Trojans’ attack. Sawyer Smith has actually played more than Barker did and his compensated passing (6.5 yards per pass play against teams that would allow 7.0 yppp to an average quarterback) is only 0.2 yppp worse than the team’s season average. However, Barker also added 357 rushing yards at 8.1 yards per run and Smith is not nearly as nimble (4.9 yprp) and losing Barker actually had a bigger affect on the rushing numbers. Overall, Troy is 0.6 yppl worse than average with Smith at quarterback and he’s also more interception prone (6 picks on 186 passes, 3.2%).

Buffalo’s defense allowed just 25.9 points and 5.3 yppl but the Bulls faced a mostly easy schedule of opposing offensive units that would combine to average only 5.0 yppl against an average team. However, that unit still has an advantage over the Troy offense and the Trojans are projected at just 326 yards at 5.4 yppl in this game.

Troy’s advantage in this game is their excellent special teams against Buffalo’s worse than average special teams units but overall the math still favors Buffalo by 1.3 points. I’ll pass on this game.

Hawaii @
Louisiana Tech

Sat, Dec 22 7:30 PM

Odds: Louisiana Tech +1, Total: 61

Game Analysis view matchup stats

Create an account to get analysis and updates sent to your inbox.

Lean – Under (-61) – HAWAII (-1)  29   Louisiana Tech  28

Hawaii Bowl – Saturday, 7:30 pm Pacific

This game basically comes down to how well Louisiana Tech’s bad offense performs against Hawaii’s even worse defense. Hawaii allowed 35.4 points and 6.4 yards per play to a collection of mostly bad offensive teams that would combine to average only 23.9 points and 5.2 yppl against an average defensive team. I suspect the Warriors will allow less than normal in this game, as Louisiana Tech would average only 4.9 yppl against an average defensive team and they averaged just 21.5 points per game against FBS opponents that were mostly worse than average defensive teams. Louisiana Tech did not score more than 31 points against any FBS opponent and the Bulldogs faced a few defenses that are just as bad as Hawaii’s defense. In those games LA Tech managed just 31 points (against UTSA), 31 points (against UTEP) and 28 points (against Rice). The math projected 29.5 points for Louisiana Tech in this game before adjusting for the windy conditions, which seems about right given their scoring against other bad defensive teams.

Hawaii’s offense posted some big numbers, averaging 436 yards at 6.3 yards per play and 32 points per game but the Rainbow Warriors faced teams that would combine to allow 30.8 points and 6.1 yppl to an average offensive team, so that unit was just 0.2 yppl better than average. Louisiana Tech’s defense only allowed 24.4 points and 5.2 yppl to FBS opponents but those opponents would combine to average only 5.0 yppl against an average defense so the Bulldogs are a bit worse than average defensively. My math projects 417 yards at 6.3 yppl for Hawaii in this game.

Overall the math picks this game even at 28 points apiece. I have no opinion on the side but the 15 mph winds should drop the projected total by about 3 points, which gives me a prediction of 56 total points. I’ll lean Under the total.

Boston College @
Boise St.

Wed, Dec 26 10:30 AM

Rotation: 229, Odds: Boise St. PK, Total:

Game Analysis view matchup stats

Create an account to get analysis and updates sent to your inbox.

Best Bet – (230) *Under (54) Boston College vs Boise State, 1-Star Under 52 or higher.

20 mph winds (with gusts into the 40s) and rain are expected in Dallas on Wednesday afternoon, which are likely to make this game much lower scoring than expected. My model leans under even without the significant weather adjustment. I’ll play Under 52 points or higher in a 1-Star Best Bet (Strong Opinion down to 51 points). More analysis will be sent later.

Minnesota vs
Georgia Tech

Wed, Dec 26 2:15 PM

Rotation: 231, Odds: Georgia Tech -5.5, Total: 56.5

Game Analysis view matchup stats

Create an account to get analysis and updates sent to your inbox.

Best Bet – *Minnesota (+6)  28   Georgia Tech  25

Best Bet – *Under (60)

Rotation #231 – Wednesday, 2:15 pm Pacific

Minnesota will have some players suspended for this game, including at least one starter, but I was going to play Minnesota at +3.5 points before that news was announced and it’s likely that the market over-adjusted for the suspensions and for the absence of leading tackler Blake Cashman and OT Donnell Greene, who are both sitting out to protect themselves from injury leading up to the NFL draft. Cashman actually is worth about 1.5 points based on his very high tackles for loss number (15 TFL) but Minnesota played two games without Greene and their offense actually performed relatively better in those games, so I didn’t adjust for that. Regarding the unknown suspended players, there aren’t that many other players on the team that really make a significant difference aside from Cashman, who I already adjusted for, and quarterback Tanner Morgan, who has been quoted as saying he’s excited for the game (which I would assume means he’s playing). As far as the Under goes, the suspensions are just as likely to hurt us as they are to help us. I did take a bit of a risk with the side bet but my math favors the Gophers by 2.4 points without knowing which players are suspended and without adjusting for them – but I figure they can’t be worth more than 2 points given the lack of remaining stars on that team.

The reason I like Minnesota is two-fold. The offense went from bad to pretty good when inaccurate freshman Zack Annexstad was replaced by Tanner Morgan in the middle of the season and the defense improved dramatically when defensive coordinator Rob Smith and his complicated defense was replaced before week 11. Annexstad didn’t even start for his high school team his senior season and completed just 51.9% of his passes this season before getting replaced. Morgan completed 59% of his passes for 7.8 yards per pass play (excluding garbage time against backups) and rated at 1.1 yppp better than average after adjusting for opposing pass defenses faced. Annexstad was 0.4 yppp worse than average, so the offense is significantly better now than their season numbers due to how bad Annexstad was.

Georgia Tech’s defense, meanwhile, is good defending the run (5.2 yprp allowed to teams that would average 5.8 yprp against an average team) and struggled defending the pass (7.3 yppp allowed to quarterbacks that would combine to average 6.2 yppp), so Morgan should put up good numbers in this game in perfect dome conditions. Minnesota, however, is a run first team and the Gophers won’t have consistent success on the ground (4.5 yprp projected). I do expect Minnesota to pass more often than they normally do and to average 8.6 yppp, but only 58 total plays are projected for the Gophers, which should net them 367 yards based on the math.

The other reason there is value on Minnesota is their defense, which improved greatly after the chance of defensive coordinator. In 9 games with Rob Smith as the DC the Gophers allowed 31.8 points per game and 6.4 yards per play to teams that would combine to average 6.1 yppl against an average defensive team. In the 3 games after Smith was fired, and the defense was simplified, Minnesota yielded just 16.3 points and 4.9 yppl to 3 bowl teams (Purdue, Northwestern, and Wisconsin) that would combine to average 6.2 yppl against an average defensive team. Minnesota’s defense wasn’t actually as bad during those first 9 games as their average rating, as that average was skewed by 3 really bad performances against Maryland (8.5 yppl allowed), Nebraska (9.3 yppl allowed), and Illinois (11.0 yppl allowed). Minnesota’s median defensive performance, which is more accurate for the purposes of predicting future performance, was 0.5 yppl better than average in those first 9 games with the other 6 games all being at least 0.3 yppl better than average (compensated for opposing offense). The simplified scheme allowed the athletic Gophers to play faster and limited the defensive breakdowns that led to those 3 really bad games. Minnesota’s two best defensive performances of the season were in those final 3 games, as the Gophers held a potent Purdue offense to just 10 points and 3.8 yppl in week 11 and limited Wisconsin to 5.6 yppl in a 37-15 win that secured this bowl game. Even if the defensive performance in those final 3 games was a nothing more than positive variance and had nothing to do with the chance in scheme, the median defensive rating for all 12 games was still 0.5 yppl better than average, which is 0.5 yppl better than Minnesota’s season rating of average (6.1 yppl allowed to teams that would average 6.1 yppl), which was skewed by those 3 horrendous performances. I have no doubt that Minnesota’s defense is much better than their season rating, even without top-tackler Cashman. Most important for this game is Minnesota’s median run defense, which was 0.7 yards per rushing play better than average. In fact, Minnesota held opponents to 0.7 yprp or more below their offensive run rating (adjusted for game site) in 8 of their 12 games, so they should perform pretty well against Georgia Tech. I made a 0.4 yprp adjustment for Cashman being out and project 5.3 yprp for Georgia Tech in this game and 5.6 yards per play for the Yellow Jackets.

Both teams are projected to combine for 5.96 yards per play, which is a bit higher than the national average of 5.74 yppl (my stats exclude kneel downs and spikes). However, both teams play at a very slow pace and only 121.3 plays and 723 total yards are projected for this game, which is well below the national average for total yards in a game (yet the total is higher than average). I actually expect more yards than the math projects because this game is being played in perfect dome conditions, which will help the passing numbers. I add that adjustment to the predicted math model score and don’t adjust the projected yardage, so expect more yards than the math model projects. By the way, domes add more to passing, which obviously helps Minnesota more than it would Georgia Tech’s run-heavy option attack. The math still projects just 53 total points even with adding a few points for playing this game in a dome. The math also favors Minnesota by 2 ½ points in this game (before adjusting for suspensions, which are unknown at this time). As mentioned earlier, I don’t expect the Minnesota suspensions to be worth more than 2 points so there is still plenty of value on the Gophers here.

I’ll go Under 59 points or more in a 1-Star Best Bet and I’ll take Minnesota in a 1-Star Best Bet at +4.5 points or more.

California vs

Wed, Dec 26 6:00 PM

Odds: TCU +1, Total: 38.5

Game Analysis view matchup stats

Create an account to get analysis and updates sent to your inbox.

California (-1)  18   Texas Christian  17

Rotation #233 – Wednesday, 6 pm Pacific

The total on this game (38.5) is the lowest bowl total since 2006 when Georgia and Virginia Tech went over 38 points in a 31-24 game. The total is low because this game features two really good defensive teams and two offenses that have had trouble scoring. Cal games averaged just 44 total points this season despite 11 defensive touchdowns being scored in the Bears’ 12 games. Cal also had a kick return TD and their offensive unit only averaged 19.3 points per game while the defense allowed just 17.8 points per game if you exclude the 6 touchdowns that were scored while their offense was on the field. The Bears rate at 0.6 yards per play worse than average offensively with quarterback Chase Garbers in the game and the Bears’ stop unit ranks 7th in the nation in compensated yards per play (4.7 yppl allowed to teams that would average 6.0 yppl against an average defensive team).

TCU games, meanwhile, have averaged a combined 47.9 total points, which really low considering that the Horned Frogs play in the high-scoring Big 12 Conference. TCU’s opponents combine to average 62.5 total points per game, and their games were 14.6 points lower scoring than that. TCU actually moves the ball at a decent rate (5.5 yppl against teams that would allow 5.6 yppl to an average team) and 3rd string quarterback Grayson Muehlstein has performed a bit better in the last two games since taking over than #1 and #2 quarterbacks Shawn Robinson and Mike Collins did, Muehlstein completed 28 of 43 passes (65%) and averaged 6.6 yards per pass play with compensated numbers slightly higher than the team average but TCU’s offense is still average at best and those short passes that Muehlstein throws aren’t likely to do much damage against a sure tackling Cal team. TCU’s defense also ranks near the top of my rankings, as the Horned Frogs allowed just 5.2 yppl despite facing teams that would combine to average 6.3 yppl against an average team, and that unit should have no trouble limiting Cal’s impotent attack.

The math model projects just 35 total points and favors Cal by 0.4 points, with the edge being in special teams. Going under low totals is risky and my model assumes an average likelihood of a defensive touchdown being scored. But, Cal games featured 11 defensive touchdowns in 12 games and TCU’s games had 5 defensive touchdowns, which is also more than average. Even with that being the case it’s still unlikely that there will be a defensive touchdown, but having one would certainly make the under less likely. I’ll pass on this game.

Duke vs

Thu, Dec 27 10:30 AM

Rotation: 235, Odds: Temple -3, Total: 54

Game Analysis view matchup stats

Create an account to get analysis and updates sent to your inbox.

Note: Temple star running back Ryquell Armstead is suddenly listed as OUT for undisclosed reasons, which hurts our play on Temple. Star CB Rock Ya-Sin is also out. But, Duke’s injured linebackers, including star Joe Giles-Harris (and Humpries) are out. The math still favors Temple by 8.3 points after all of those adjustments – although I would have made Temple a 1-Star Best Bet at -3 or less only had all of this injury info been reported earlier.

Best Bet – **Temple (-3 at -115)  30   Duke  22

Rotation #236 – Thursday, 10:30 am Pacific

Temple is 8-1 straight up and 7-2 ATS with Anthony Russo at quarterback and the two spread losses were both close – a 12 point loss as an 11-point dog at UCF in a game that was neck-and-neck the entire was before UCF scored twice late to barely cover and a 10 point win laying 13.5 points to South Florida. Russo has been upgraded to probable for this game after missing the season finale against Connecticut and Owls are reportedly excited to play in this bowl despite losing head coach Geoff Collins to Georgia Tech. Unlike a few years ago when Matt Rhule left, all the assistant coaches have stayed behind to coach the Owls and interim head coach Ed Foley and his players have commented on the positive difference in preparation for this game.

With Russo at quarterback and RB Ryquell Armstead both healthy the Owls offense is 0.4 yards per play better than average, which gives them an advantage over a banged-up Duke defense that is just average even if linebackers Joe Giles-Harris and Ben Humphreys are able to play. Humphreys has barely played since getting injured against Virginia in week 8 (he missed week 9, played hurt in weeks 10 and 11, and tried to play in weeks 12 and 13 and was pulled in both games) and Giles-Harris, a two time All-ACC selection, has missed the last 3 games in which Duke allowed an average of 291 rushing yards at 6.8 yards per rushing play (to teams that would average 5.9 yprp against an average team). However, Duke’s run defense was leaking even before Giles-Harris was injured, as 305-pound defensive tackle Edgar Cerenord was lost for the season starting in week 8 and #2 tackler, SS Dylan Singleton was also lost to a season-ending injury in the first half of the week 11 game against North Carolina. In 3 games after losing Cerenord (weeks 8 through 10) and before losing Giles-Harris and Singleton, the Blue Devils allowed 329 rushing yards per game at 7.4 yprp (to teams that would average 5.9 yprp), so the issues with the run defense appear to have more to do with losing their run-stuffing defensive tackle than it does with losing their linebackers and strong safety since the troubles defending the run started in week 8 after Cerenord was lost.  I will assume that the run defense will be better if Giles-Harris can play in this game but I don’t think it will be better than their season rating of 0.2 yprp worse than average. In fact, it’s likely to be worse than that given that the Blue Devils couldn’t stop the run after Cerenord went down, even with Giles-Harris and Humphries. For now I just used Duke’s season run defense rating for the math on this game and I’d like Temple even more if Giles-Harris can’t play. Giles-Harris has practiced some but is said to be a game-time decision and certainly won’t be 100% if he does play on that injured knee. My guess is that the coaching staff wouldn’t want to risk the future of a player that will play in the NFL if he doesn’t damage that knee any further. But, I like Temple regardless.

The big edge for Temple in this game is their extremely stingy defense against Duke’s below average offense, which is led by extremely overrated quarterback Daniel Jones, who is said to be a possible first-round NFL draft pick despite completing a mediocre 59% of his passes and average only 5.5 yards per pass play (to teams that would allow 5.8 yppp to an average quarterback). If Jones is an NFL talent then his receivers must suck because there is nothing about his below average performance that suggests he’s even good enough to start for an average College team. The Duke rushing attack is average (5.1 yprp against teams that would allow 5.1 yprp to an average team) and overall the Blue Devils’ offense is 0.1 yards per play worse than average with Jones at quarterback (he missed the Baylor game and a game against FCS team NC Central).

Temple’s defense is among the best in the nation defending the pass, as the Owls yielded just 54.8% completions and 4.3 yards per pass play to quarterbacks that would combine to average 6.2 yppp against an average defense. Jones is worse than the average quarterback that Temple has faced and Duke is projected to average only 4.1 yppp in this game. The only 3 FBS quarterbacks to average more than 4.1 yppp against Temple were all well above average quarterbacks (Buffalo’s Jackson averaged 6.1 yppp, UCF’s Milton averaged 9.4 yppp, and Houston’s King averaged 5.9 yppp). The Owls absolutely dominated average or worse quarterbacks this season and I’d be really surprised if Jones averaged more than 5 yards per pass play in this game. The Temple run defense is 0.3 yprp better than average so I don’t expect Duke’s mediocre ground attack to do much damage.

Overall, Duke is projected at just 352 yards at 4.5 yards per play while Temple’s offense is projected to gain 429 yards at 5.8 yppl even if Duke linebackers Giles-Harris and Humphries play at 100% effectiveness. Overall, the math favors Temple by 11 points at the very least (up to 14 points if Giles-Harris and Humphries don’t play) and the math projects 51 to 54 total points depending on the status of Duke’s LB’s. I will play Temple in a 2-Star Best Bet at -3 at -115 odds or better and for 1-Star up to -4 points.

If I find out Giles-Harris is not going to play then Temple would be a 2-Star up to -4 points.

Miami Fla vs

Thu, Dec 27 2:15 PM

Rotation: 237, Odds: Wisconsin +3, Total: 47.5

Game Analysis view matchup stats

Create an account to get analysis and updates sent to your inbox.

*Wisconsin (+3.5 at -115)  24   Miami-Florida  23

Rotation #238 – 5:15 pm Pacific

There were 4 reasons for my play on Wisconsin when I released it. One was the fact that the Badgers ran the ball relatively better against better run defenses, which Miami has. The second reason was Miami’s best run defender and best all-around defensive player, DT Gerald Willis III, has decided to sit out this game to prepare for the NFL draft, which will make it easier for Wisconsin to run the ball. The third reason is that Wisconsin’s defense is underrated with freshman CB Rachad Wildgoose in the lineup, as he led the team in passes defended (7) despite barely playing in the first 5 games and the pass defense was much, much better after he became a starter. The fourth reason was the most important – Miami quarterback N’Kosi Perry is absolutely terrible, averaging just 4.9 yards per pass play despite facing teams that would allow 6.5 yppp to an average quarterback. Perry was expected to start until he posted a sexually explicit Snapchat, which has led to his benching and veteran Malik Rosier now starting at quarterback for the Hurricanes. Rosier isn’t great, as he was just average on his 142 pass plays against FBS competition this season (5.6 yppp against teams that would allow 5.6 yppp to an average QB), but that’s a significant upgrade from Perry’s -1.6 yppp rating. I still would favor Wisconsin to cover with Rosier at quarterback but I certainly wouldn’t have made the Badgers a Best Bet had Perry’s idiotic social media move had happened before I released my play.

I still would rather have Wisconsin though, as the Badgers rush attack, led by All-American Jonathan Taylor (1989 yards at 7.1 ypr), was relatively better against good run defenses. Wisconsin faced 5 good run defenses in the regular season (BYU, Iowa, Michigan, Northwestern, and Penn State) and the Badgers ran for an average of 6.1 yard per rushing play, which is 1.9 yprp more than what those teams would allow an average running team. Miami has a good run defense (4.5 yprp allowed to teams that would combine to average 5.3 yprp against an average team) but not having Gerald Willis significantly hurts that run defense. Willis, a two time All-American, led all defensive tackles in the nation in tackles for loss with 18, which is an incredible number for a defensive tackle that usually faces double-teams. With Miami’s run defense weakened, and with Wisconsin having proven that they can run successfully against good run defenses, the Badgers won’t have to throw as often into the dominating Hurricanes’ secondary. Miami yielded only 4.3 yards per pass play (to quarterbacks that would average 6.4 yppp against an average team) and Wisconsin freshman quarterback Jack Coan isn’t likely to have any success throwing against the Hurricanes in this game (only 3.9 yppp projected). However, I project 274 rushing yards at 6.5 yprp for Wisconsin and for the Badgers to total 357 yards at a decent 5.6 yards per play in this game.

Wisconsin wasn’t as good as expected this season because their defense was mediocre. Wisconsin allowed 6.0 yards per play over their first 5 games (to teams that would combine to average just 5.5 yppl against an average defense) and the pass defense was particularly bad (7.0 yppp allowed to quarterbacks that would combine to average 6.1 yppp against an average defense). A change was made when freshman CB Rachad Wildgoose was inserted into the starting lineup in week 7. Wildgoose barely played in the first 5 games but he led the team with 7 passes defended in just 7 starts and the pass defense in those final 7 regular season games improved significantly, allowing just 5.4 yards per pass play to quarterbacks that would combine to average 6.4 yppp against an average defensive team. Overall, the defensive was 0.6 yards per play better than average after that change in the secondary was made and that improvement in the defense is not in the line because it looks like the Badgers got worse defensively based on their points allowed in those final couple of games. Wisky gave up 81 points in their final two games against Purdue and Minnesota, but they only allowed 5.6 yppl, which was actually good considering Purdue and Minnesota (with Morgan at quarterback) would average 6.4 yppl against an average team. Wisconsin has a slight advantage over a Miami offense that is 0.3 yppl better than average with Rosier at quarterback and I project 365 yards at 5.7 yppl for the Hurricanes in this game.

Overall, the math picks this game even, so there is still value on Wisconsin even with Miami’s upgrade at quarterback, and I have a hard time believing that Miami’s players that are used to playing in warm weather are going to be fired up to play in 40-degree weather in New York City. Wisconsin will count as a 1-Star Best Bet but I’d just lean with the Badgers now and I have no opinion on the total.

Vanderbilt vs

Thu, Dec 27 6:00 PM

Rotation: 239, Odds: Baylor +4, Total: 55.5

Game Analysis view matchup stats

Create an account to get analysis and updates sent to your inbox.

Vanderbilt (-4)  30   Baylor  26

Rotation #239 – Thursday, 6 pm Pacific

I really have nothing interesting to say about this bowl game. My math model favors Vanderbilt by 4 points with a total of 55.7 points assuming the retractable roof is closed at NRG Stadium. Jalen Hurd being out for Baylor barely moves the needle, as Hurd averaged 8.8 yards per target and the rest of the Baylor wide receivers combine to average 8.4 YPT. There is not significant value on the side or total in this game between two mediocre teams. Rather than spend my valuable time writing up analysis of a game I have no opinion on I have decided to move on to this weekend’s more important bowl games.