Game Analysis

Strong Opinion – SOUTH FLORIDA (+4)  25   Marshall  24

Gasparilla Bowl – Thursday, 5 pm Pacific

South Florida started the season with 7 consecutive wins but only one of those victories was against a bowl-bound team (Georgia Tech) and the Bulls came crashing down to earth with 5 losses against bowl teams to end the regular season. Obviously, Marshall is a bowl team but the Thundering Herd are a worse than average team by FBS standards and are only a couple of points better than the average rating of the team’s that South Florida faced this season. In other words, the Bulls should be as good as they were overall this season in all areas except for quarterback with Blake Barnett reportedly available for emergency duty only. USF players are excited about being able to play another game in their home stadium after losing their final two home games of the regular season and I think they’ll play well even without Barnett.

South Florida’s offense is only average even with Barnett, who is actually not that good. Barnett was 0.5 yards per pass play worse than average after compensating for opposing defenses faced, but the backups were considerably worse in the two games that Barnett missed (Cincinnati and UCF). Chris Odadokun and Brett Kean combined for just 4.0 yards per pass play but they were without All-Conference TE Mitchell Wilcox in both of those games (he will play tonight based on his own words) and Cincy and UCF both have good pass defenses. Those two opponents would combine to allow just 5.4 yppp to an average quarterback while Barnett faced teams that would allow 7.3 yppp, so the difference between Barnett and his backups is not as extreme as the raw stats make it look. The difference, which Wilcox back, is about 0.8 yards per pass play, which equates to about 3 points.

With Barnett out the math model would still project decent production against a Marshall defense that wasn’t as good as their numbers suggest. The Herd stop unit allowed just 22 points and 4.9 yards per play but they faced a schedule of teams that would combine to average just 5.1 yppl against an average defense after factoring in the 3 backup quarterbacks that they faced. The run defense is good (4.1 yards per rushing play allowed to teams that would combine to average 4.5 yprp against an average team), which takes on more importance in what will be poor conditions for throwing the ball, but the Marshall pass defense is just average (5.5 yppp allowed to quarterbacks that would combine to average just 5.5 yppp against an average defense). South Florida is a good running team with two good backs in Jordan Cronkite and Johnny Ford, who combined for 1832 rushing yards at 6.6 ypr despite missing a combined 3 games. Having TE Wilcox back will also help in the windy conditions, as short passes take on more weight. South Florida may actually be better offensively than expected due to a change in play calling after Sterlin Gilbert left of the head coaching job at McNeese State. Gilbert was Charlie Strong’s OC at Texas too and I think he’s a terrible play-caller, so there is room for improvement in that area. The gusting winds could limit what new play-caller Justin Burke can do – although being the tight ends coach I expect Wilcox to see a lot of targets, which would be a positive given that he averaged 9.4 yards per target this season.

Marshall’s offense was 0.7 yards per play worse than average this season, (5.6 yppl against mostly bad defensive teams that would allow 6.3 yppl to an average attack) but the Thundering Herd are only 0.3 yppl worse than average with Isaiah Green at quarterback, as backup Alex Thomson was dreadful in the 3 ½ games that he played. Still, South Florida’s defense, which rates at 0.1 yppl worse than average (5.8 yppl allowed to teams that would average 5.7 yppl against an average team) has the slight advantage in this game. Marshall is projected to average just 5.3 yppl even if the weather were good but 15 mph winds with gusts into the 40s are expected and it will rain most of the day leading up to this game. Running the ball will be key and while Marshall has a better run defense, the Bulls have better running backs – especially if Marshall’s top runner Tyler King (655 yards at 6.1 ypr in just 7 games) misses his 6^th consecutive game (he’s questionable). Throwing will be tough and Green completed just 55.7% of his passes this season and is likely to be less accurate in the wind.

Playing at home may not seem like as much of an advantage in a bowl game but home teams are 19-11 ATS in bowl games since 1980 when not favored by 7 points or more and have covered the spread by an average of 6.8 points in those games – so I used a standard home field advantage for this game even though a good number of Marshall players have ties to southern Florida. There is value on South Florida even with Barnett not playing (the math favors USF by 4 with Barnett and by 1 points without him) and the only reason that they’re an underdog is because they’ve lost 5 straight games, which has lowered their stock. When a stock is low it’s time to buy and teams that have lost their previous 5 or more games are actually 4-0 straight up and 4-0 ATS in their bowl game with an average winning margin of 13.5 points and an average spread of +4.1 points. USF has something to prove and I think they’ll play well in this game. I’ll consider South Florida a Strong Opinion at +3.5 points or more.

The math projected 55 total points with Barnett out for USF but the projected weather conditions equate to 6.0 fewer points being scored. The market has mostly adjusted so I’ll pass on the total.

Note: The statistics in the math model projections included in the emails are not adjusted for weather. I only adjust the points scored for each team. In the case of this game, if you’re playing props, you should assume both teams to run the ball more often, throw with less efficiency, and for there to be fewer total plays (due to more running).