Dr. Bob College Football
Monday Dec. 29
[h=3]***West Virginia (pk) 40 Texas A&M 29[/h]Mon Dec-29-2014 at 11:00 AM Pacific Rotation: 237 Over/Under 65.0 - Matchup Stats
West Virginia is much better than their 7-5 record suggests and Texas A&M isn’t as good now as they were earlier in the season when Kenny Hill was their quarterback. Hill started out the season in a fashion that made Aggies’ fans quickly forget about Johnny Maziel but Hill cooled off, was suspended for a couple of games and never got his job back. Coach Kevin Sumlin decided to go with freshman Kyle Allen even when Hill was available to play and Allen is simply not as good. Hill averaged 7.3 yards on 310 pass plays against FBS opponents that would allow 5.8 yppp to an average quarterback. Allen, meanwhile, has averaged 5.6 yards on his 151 pass play while facing teams that would allow 5.4 yppp to an average quarterback. The numbers are about the same in his 4 starts (5.5 yppp against 5.2 yppp) but I’ve decided to use Allen’s final two games only, which are a bit better (5.2 yppp against teams that would allow 4.7 yppp to an average quarterback). The Aggies’ rushing attack has been 0.7 yards per rushing play better than average (a bit worse without Hill’s 264 yards on 35 runs) and overall I rate the Texas A&M at 0.7 yards per play better than average with Allen at quarterback, which is 0.3 yppl worse than their season rating. West Virginia’s defense is equally good defending the run (0.5 yprp better than average) and the pass (0.8 yppp better than average) and overall the Mountaineers have been 0.6 yppl better than average, allowing 5.6 yppl to a schedule of FBS teams that would combine to average 6.2 yppl against an average team. This is a pretty even match up and Texas A&M is projected to gain 420 yards at 5.65 yards per play.
West Virginia’s advantage is when the Mounties have the ball. West Virginia averaged 495 yards per game at 5.9 yards per play while facing teams that would combine to allow 5.2 yppl to an average team. Quarterback Clint Trickett suffered a concussion and missed the final 1 ½ games but backup Skyler Howard was a bit better that Trickett on a compensated yards per pass play basis and hasn’t throw an interception on 66 pass plays while also adding a running dimension that Trickett doesn’t have. Howard ran 7 times for 69 yards in the finale against Iowa State and has averaged 10.6 yards on his 11 runs (excluding his one sack, which goes into passing stats), so he very well might be a better overall option than Trickett, who has announced his retirement from football. What’s telling about Howard’s 11 runs is that his 10.6 yards per run average is not skewed by a big run. In fact, Howard’s longest run is just 24 yards, which shows he has consistently run well and I expect that to continue, as does the coaching staff, who has mentioned what an addition Howard’s running would be to the offense. Word is that coach Dan Holgorsen was leaning towards starting Howard in this game even if Trickett had been cleared to play and Holgorsen’s quarterbacks always play at the same high level, which was the case at Houston and Oklahoma State as an offensive coordinator, where his backups played just as well as his starters. Last season both his quarterbacks had identical compensated numbers and again this season Howard has filled in without the offense taking a dip. West Virginia’s pass offense was 1.4 yards per pass play better than average this season (6.9 yppp against teams that would allow 5.5 yppp to an average team) and Howard has been 1.5 yppp better than average in his 66 pass plays (7.2 yppp against teams that would allow 5.7 yppp to an average quarterback). I actually downgraded Howard’s pass rating because he’s averaging 13.4 yards per completion, which is higher than the team average ypc, but his running makes up for that reduction in passing – and it might be that Howard simply is better at throwing the ball down the field than Trickett was, as Trickett had a tendency to make the safe underneath throw, which is why he averaged a sub-par 11.7 yards per completion. Howard isn’t making risky throws necessarily, as he has not been intercepted on his 65 pass attempts. Overall, the offense appears to be at least as good with Howard at quarterback and the Mountaineers’ attack has a higher compensated yards per play rating in his 1 ½ games at the helm.
Texas A&M’s defense wasn’t as bad as they looked, as the 464 yards per game they allow is a function of their fast pace on offense, which leads to more defensive possessions too, and the quality of opponent they faced. The Aggies were only 0.3 yppl worse than average defensively on a national scale, allowing 6.2 yppl to teams that would combine to average 5.9 yppl against an average defensive unit. I decided to take out their bad defensive game against Missouri (591 yards at 6.8 yppl against a mediocre Tigers’ offense) since that was the game that star LB Myles Garrett missed. Even with that boost to the defensive numbers the math still projects 576 yards at 6.6 yppl for West Virginia in this game.
Not only does West Virginia have a 1.0 yppl advantage in this game but the Mountaineers are #2 in the nation in plays per game and are projected to run about 13 more plays than Texas A&M, which results in a predicted advantage of 155 total yards. West Virginia is significantly better than Texas A&M from the line of scrimmage and are an underrated team because they haven’t been able to efficiently turn their 82 yards per game advantage into more than a 2.8 points advantage. The Mounties’ point margin should be much better based on their combination of yards and yards per play, which project a scoring margin of +7.3 points per game when you add in special teams. The disparity is almost 100% due to fumble variance. As you probably know by now fumbles are 90% random in college football, which is to say that only 10% of the past fumble differential translates into future fumble differential. West Virginia leads the nation is negative fumble differential at -16, as the Mounties have lost 18 of their 27 fumbles while recovering only 2 of their opponent’s 12 fumbles. That’s bad luck in the number of fumbles for and against and particularly bad luck that they’ve lost 67% of their fumbles but have recovered only 17% of their opponent’s fumbles. A -16 in fumbles translated into over 5 points per game of negative variance, which is the main reason why West Virginia is underrated heading into this game. West Virginia had proven more than capable of competing with really good teams (although I do not consider A&M to be a really good team), as the Mountaineers beat Baylor 41-27 despite a -3 in turnovers, lost by just 1 point to TCU despite also being -3 in turnovers, and only lost by 6 to Kansas with a -2 turnover difference. West Virginia could very easily be 10-2 and going to a major bowl game if not for the fumbles and they should beat up on Texas A&M as long as their negative fumble luck doesn’t carry over, which is most likely won’t. My math model favors West Virginia by 12 points based on the projected stats and to have a +0.26 edge in turnovers based on being less likely to throw interceptions (they’re +1 in interceptions while Texas A&M is -9 interceptions, which is far more predictive than fumbles but still regresses towards the mean). I was going to make West Virginia a Best Bet at -3 ½ with the assumption that Howard would start and now we have extra line value because Trickett has retired. I suspected the line might go down to -3 if Trickett didn’t play but the line has dropped to a pick, as apparently people have no concept about Holgorsen’s history of plugging any quarterback into his system without a drop in production. The underrated Mountaineers have a very profitable 60.7% chance of covering and I’ll take West Virginia in a 3-Star Best Bet at -3 points or less and for 1-Star up to -4 points.
The fumbles that kept West Virginia from scoring as many points as they should have scored with a normal amount of fumbles also has served to keep the total on this game lower than it should be. My model predicts 71 total points and the total dropped 2 points from 67 to 65 on the news of Trickett’s retirement, which is 2 points of free line value with two key numbers (66 and 65) involved. I’ll consider the Over a Strong Opinion at 65 points or lower and a lean at higher than 65 points.
[h=3]**Oklahoma (-4) 27 Clemson 14[/h]Mon Dec-29-2014 at 02:30 PM Pacific Rotation: 240 Over/Under 52.5 - Matchup Stats
**UNDER (52 ½)
I have two plays on this game for the same reason. Clemson is significantly worse offensively with Cole Stoudt at quarterback. Oklahoma is certainly better with Travis Knight back at quarterback, even if star WR Sterling Shepard is not 100%, which aides our side bet on Oklahoma but doesn’t take enough away from the value on the Under. Clemson had two quarterbacks this season, one being Stoudt and the other being freshman superstar DeShawn Watson, who will be among the best quarterbacks in the nation next year. Watson was in and out of the lineup due to a knee injury this season but he was incredibly good when he played, averaging 9.2 yards on 137 pass plays against FBS opponents that would combine to allow 6.5 yppp to an average quarterback. Cole Stoudt, meanwhile, averaged only 4.8 yppp on this 247 pass plays against FBS opponents that would allow 5.9 yppp to an average QB, and he also threw 10 interceptions (Watson threw just 2). Clemson is a good offensive team with Watson playing and they’re horrible with Stoudt in the lineup because he’s terrible and the Tigers’ running game sucks (4.3 yards per rushing play against teams that would allow 5.1 yprp to an average team). The rushing attack is actually worse with Stoudt playing because opponents aren’t worried about being beat with a deep pass (Stoudt averages only 9.5 yards per completion) and they can bring a safety closer to the line of scrimmage to help defend the run. Clemson’s offense averaged only 5.2 yards per play this season (against teams that would allow 5.6 yppl to an average team) and the Tigers are 1.0 yppl worse than average with Stoudt at quarterback. Oklahoma’s defense is 0.7 yppl better than average (5.3 yppl allowed to teams that would combine to average 6.0 yppl against an average team) and they are equally good defending the run and the pass. The math model projects just 310 yards at 4.0 yards per play for Clemson in this game. The Tigers only averaged 20.7 points in the 7 games against FBS teams when Stoudt was the main quarterback and those opponents were not nearly as good defensively as Oklahoma. Clemson isn’t likely to reach 20 points and the math projects just 14 points for the Tigers.
The reason I didn’t release this game earlier is because I was waiting on an update on the status of Oklahoma’s star WR Sterling Shepard. Shepard is a difference maker, as passes targeted for Shepard have averaged 12.3 yards while passes thrown to the other 3 main wide receivers have averaged just 6.7 yards. That’s a major difference and the Oklahoma pass offense struggled for two games without Shepard even before starting quarterback Travis Knight was injured and it was even worse for the final 3 games without Knight. Shepard basically missed the final 5 games of the season (he caught one pass for 46 yards early against Iowa State in week 10 before being injured) and Knight averaged only 6.0 yards per pass play on 62 pass play without Shepard. That’s a major drop from the 8.2 yppp on 223 pass plays that Knight averaged when Shepard was healthy. Shepard did practice this week and expects to play but even he’s not sure how his groin will hold up when playing at game speed. Clemson has a great pass defense (4.5 yppp allowed to quarterbacks that would average 6.5 yppp against an average defense) and the math would project only 6.1 yppp for Oklahoma if Shepard were 100% and playing at his old level. If Shepard didn’t play then the projection would be just 4.8 yppp for Knight in this game. The battle between Oklahoma’s great rushing attack (6.4 yards per rushing play against teams that would allow 5.0 yprp) and Clemson’s great run defense (4.2 yprp allowed to teams that would average 5.4 yprp) is pretty even and the math model is projecting 4.9 yprp for the Sooners in this game. I decided to use Knight’s overall pass rating for this game, which is based on 223 pass play with Shepard and 65 pass plays without Shepard. That takes into account the chance that Shepard may not be 100%. With that assumption the math would project 340 yards at 4.9 yards per play for Oklahoma and 29 points.
Oklahoma’s edge in total yards is amplified by the much better field position that they should enjoy with their great special teams and Clemson’s sub-par special teams. The Sooners have huge edges in net punting and net kickoff yard line average and the better field position should make a difference in this game. Oklahoma also has an advantage in projected turnovers and overall the math favors the Sooners by 15 points and 42.7 total points based on the projected stats. If Shepard doesn’t play then the math would favor the Sooners by 11.3 points with a total of just 39.7 points and if Shepard is 100% and he and Knight play at their pre-injury level then I would get Oklahoma by 15.9 points with 43.7 total points. The math significantly likes Oklahoma and the Under in all 3 scenarios. I’ll take Oklahoma in a 2-Star Best Bet at -6 points or less (3-Stars at -3) and I’ll take the Under in a 2-Star Best Bet at 51 points or higher.
[h=3]Arkansas (-6) 23 Texas 16[/h]Mon Dec-29-2014 at 06:00 PM Pacific Rotation: 242 Over/Under 45.0 - Matchup Stats
The Texas Bowl executives have got to be thrilled to have secured two high profile programs whose fans are likely to make the short trip to Houston in good numbers. While the stands should be pretty full the action on the field doesn’t figure to be all that exciting given the conservative nature of both offenses and the good defense being played by both sides. These teams are actually pretty even from a yards per play perspective but Texas doesn’t move the chains nearly as well as Arkansas does and the Razorbacks are expected to have a significant possession and play advantage.
The Arkansas offense is a run-oriented attack that features two backs that have topped 1000 yards in Jonathan Williams and Alex Collins. Despite the lofty yardage totals the Razorbacks are actually pretty mediocre running the ball, as they’ve averaged a modest 5.0 yards per rushing play in 11 games against FBS opponents that would combine to allow 4.9 yprp to an average team. Quarterback Brandon Allen, meanwhile, has averaged only 6.0 yards per pass play (6.2 yppp is the national average) but he’s faced a schedule of teams that would combine to allow just 5.3 yppp to an average quarterback, so Allen has actually been significantly better than average and has thrown just 5 interceptions all season. Allen will be tested again by a Texas defense that has the 2nd best pass defense in the nation. The Longhorns have allowed just 4.6 yppp to a collection of quarterbacks that would combine to 6.6 yppp against an average team. Texas held some of the best quarterbacks in the nation in check as UCLA, Baylor, Oklahoma, Kansas State, and TCU combined to average only 5.6 yppp against the Horns. Allen is projected to average 5.1 yppp while the Arkansas rushing attack is expected to gain 4.7 yards per rushing play against a good but not great Texas run defense (0.5 yprp better than average). Overall the Razorbacks are projected to gain 347 yards at 4.9 yards per play.
Arkansas was one of the most improved defensive teams in the nation, as the Hogs struggled on that side of the ball last season but have been 0.8 yards per play better than average defensively this year (5.6 yppl allowed to teams that would combine to average 6.4 yppp against an average defense). I actually rate the Hogs at 1.0 yppp better than average since their pass defense with CB Tevin Mitchel was 0.2 yppp better if you exclude the opener against Auburn that he missed. Auburn took advantage of Mitchel’s absence in that game and beat the Razorbacks deep several times. The Arkansas run defense, meanwhile, was a bit better if I exclude the two games that #2 tackler Brooks Ellis missed in the middle of the season. The Arkansas defense faced the toughest schedule of opposing FBS offenses in the nation and that unit should not have any trouble limiting a sub-par Texas attack that 0.2 yppl worse than average in 11 games with Tyrone Swoopes at quarterback. Swoopes replaced David Ash after 1 game and was decent (5.7 yppp against teams that would allow 5.8 yppp to an average QB) while the rushing attack was also a bit below average (4.6 yprp against teams that would allow 4.8 yprp to an average team). That attack is a full 1 yard worse than that average rating of the FBS teams that Arkansas faced this season and the Hogs are projected to limit the Longhorns to just 276 yards at 4.7 yppl in this game.
Arkansas has an overall edge from the line of scrimmage due to being projected to run 13 more plays than Texas and the Razorbacks also have better special teams and an edge in projected turnovers of 0.3. Overall the math favors Arkansas by 7 ½ points, which is actually not enough of an advantage to recommend a side since the chance of covering is just 51.4% based on the historical performance of my model. However, the model also projects just 39.4 total points, which would historically be a 53.7% play on the Under (44.5 points). That is a profitable percentage, so I’ll lean Under and I have no opinion on the side.
Monday Dec. 29
[h=3]***West Virginia (pk) 40 Texas A&M 29[/h]Mon Dec-29-2014 at 11:00 AM Pacific Rotation: 237 Over/Under 65.0 - Matchup Stats
West Virginia is much better than their 7-5 record suggests and Texas A&M isn’t as good now as they were earlier in the season when Kenny Hill was their quarterback. Hill started out the season in a fashion that made Aggies’ fans quickly forget about Johnny Maziel but Hill cooled off, was suspended for a couple of games and never got his job back. Coach Kevin Sumlin decided to go with freshman Kyle Allen even when Hill was available to play and Allen is simply not as good. Hill averaged 7.3 yards on 310 pass plays against FBS opponents that would allow 5.8 yppp to an average quarterback. Allen, meanwhile, has averaged 5.6 yards on his 151 pass play while facing teams that would allow 5.4 yppp to an average quarterback. The numbers are about the same in his 4 starts (5.5 yppp against 5.2 yppp) but I’ve decided to use Allen’s final two games only, which are a bit better (5.2 yppp against teams that would allow 4.7 yppp to an average quarterback). The Aggies’ rushing attack has been 0.7 yards per rushing play better than average (a bit worse without Hill’s 264 yards on 35 runs) and overall I rate the Texas A&M at 0.7 yards per play better than average with Allen at quarterback, which is 0.3 yppl worse than their season rating. West Virginia’s defense is equally good defending the run (0.5 yprp better than average) and the pass (0.8 yppp better than average) and overall the Mountaineers have been 0.6 yppl better than average, allowing 5.6 yppl to a schedule of FBS teams that would combine to average 6.2 yppl against an average team. This is a pretty even match up and Texas A&M is projected to gain 420 yards at 5.65 yards per play.
West Virginia’s advantage is when the Mounties have the ball. West Virginia averaged 495 yards per game at 5.9 yards per play while facing teams that would combine to allow 5.2 yppl to an average team. Quarterback Clint Trickett suffered a concussion and missed the final 1 ½ games but backup Skyler Howard was a bit better that Trickett on a compensated yards per pass play basis and hasn’t throw an interception on 66 pass plays while also adding a running dimension that Trickett doesn’t have. Howard ran 7 times for 69 yards in the finale against Iowa State and has averaged 10.6 yards on his 11 runs (excluding his one sack, which goes into passing stats), so he very well might be a better overall option than Trickett, who has announced his retirement from football. What’s telling about Howard’s 11 runs is that his 10.6 yards per run average is not skewed by a big run. In fact, Howard’s longest run is just 24 yards, which shows he has consistently run well and I expect that to continue, as does the coaching staff, who has mentioned what an addition Howard’s running would be to the offense. Word is that coach Dan Holgorsen was leaning towards starting Howard in this game even if Trickett had been cleared to play and Holgorsen’s quarterbacks always play at the same high level, which was the case at Houston and Oklahoma State as an offensive coordinator, where his backups played just as well as his starters. Last season both his quarterbacks had identical compensated numbers and again this season Howard has filled in without the offense taking a dip. West Virginia’s pass offense was 1.4 yards per pass play better than average this season (6.9 yppp against teams that would allow 5.5 yppp to an average team) and Howard has been 1.5 yppp better than average in his 66 pass plays (7.2 yppp against teams that would allow 5.7 yppp to an average quarterback). I actually downgraded Howard’s pass rating because he’s averaging 13.4 yards per completion, which is higher than the team average ypc, but his running makes up for that reduction in passing – and it might be that Howard simply is better at throwing the ball down the field than Trickett was, as Trickett had a tendency to make the safe underneath throw, which is why he averaged a sub-par 11.7 yards per completion. Howard isn’t making risky throws necessarily, as he has not been intercepted on his 65 pass attempts. Overall, the offense appears to be at least as good with Howard at quarterback and the Mountaineers’ attack has a higher compensated yards per play rating in his 1 ½ games at the helm.
Texas A&M’s defense wasn’t as bad as they looked, as the 464 yards per game they allow is a function of their fast pace on offense, which leads to more defensive possessions too, and the quality of opponent they faced. The Aggies were only 0.3 yppl worse than average defensively on a national scale, allowing 6.2 yppl to teams that would combine to average 5.9 yppl against an average defensive unit. I decided to take out their bad defensive game against Missouri (591 yards at 6.8 yppl against a mediocre Tigers’ offense) since that was the game that star LB Myles Garrett missed. Even with that boost to the defensive numbers the math still projects 576 yards at 6.6 yppl for West Virginia in this game.
Not only does West Virginia have a 1.0 yppl advantage in this game but the Mountaineers are #2 in the nation in plays per game and are projected to run about 13 more plays than Texas A&M, which results in a predicted advantage of 155 total yards. West Virginia is significantly better than Texas A&M from the line of scrimmage and are an underrated team because they haven’t been able to efficiently turn their 82 yards per game advantage into more than a 2.8 points advantage. The Mounties’ point margin should be much better based on their combination of yards and yards per play, which project a scoring margin of +7.3 points per game when you add in special teams. The disparity is almost 100% due to fumble variance. As you probably know by now fumbles are 90% random in college football, which is to say that only 10% of the past fumble differential translates into future fumble differential. West Virginia leads the nation is negative fumble differential at -16, as the Mounties have lost 18 of their 27 fumbles while recovering only 2 of their opponent’s 12 fumbles. That’s bad luck in the number of fumbles for and against and particularly bad luck that they’ve lost 67% of their fumbles but have recovered only 17% of their opponent’s fumbles. A -16 in fumbles translated into over 5 points per game of negative variance, which is the main reason why West Virginia is underrated heading into this game. West Virginia had proven more than capable of competing with really good teams (although I do not consider A&M to be a really good team), as the Mountaineers beat Baylor 41-27 despite a -3 in turnovers, lost by just 1 point to TCU despite also being -3 in turnovers, and only lost by 6 to Kansas with a -2 turnover difference. West Virginia could very easily be 10-2 and going to a major bowl game if not for the fumbles and they should beat up on Texas A&M as long as their negative fumble luck doesn’t carry over, which is most likely won’t. My math model favors West Virginia by 12 points based on the projected stats and to have a +0.26 edge in turnovers based on being less likely to throw interceptions (they’re +1 in interceptions while Texas A&M is -9 interceptions, which is far more predictive than fumbles but still regresses towards the mean). I was going to make West Virginia a Best Bet at -3 ½ with the assumption that Howard would start and now we have extra line value because Trickett has retired. I suspected the line might go down to -3 if Trickett didn’t play but the line has dropped to a pick, as apparently people have no concept about Holgorsen’s history of plugging any quarterback into his system without a drop in production. The underrated Mountaineers have a very profitable 60.7% chance of covering and I’ll take West Virginia in a 3-Star Best Bet at -3 points or less and for 1-Star up to -4 points.
The fumbles that kept West Virginia from scoring as many points as they should have scored with a normal amount of fumbles also has served to keep the total on this game lower than it should be. My model predicts 71 total points and the total dropped 2 points from 67 to 65 on the news of Trickett’s retirement, which is 2 points of free line value with two key numbers (66 and 65) involved. I’ll consider the Over a Strong Opinion at 65 points or lower and a lean at higher than 65 points.
[h=3]**Oklahoma (-4) 27 Clemson 14[/h]Mon Dec-29-2014 at 02:30 PM Pacific Rotation: 240 Over/Under 52.5 - Matchup Stats
**UNDER (52 ½)
I have two plays on this game for the same reason. Clemson is significantly worse offensively with Cole Stoudt at quarterback. Oklahoma is certainly better with Travis Knight back at quarterback, even if star WR Sterling Shepard is not 100%, which aides our side bet on Oklahoma but doesn’t take enough away from the value on the Under. Clemson had two quarterbacks this season, one being Stoudt and the other being freshman superstar DeShawn Watson, who will be among the best quarterbacks in the nation next year. Watson was in and out of the lineup due to a knee injury this season but he was incredibly good when he played, averaging 9.2 yards on 137 pass plays against FBS opponents that would combine to allow 6.5 yppp to an average quarterback. Cole Stoudt, meanwhile, averaged only 4.8 yppp on this 247 pass plays against FBS opponents that would allow 5.9 yppp to an average QB, and he also threw 10 interceptions (Watson threw just 2). Clemson is a good offensive team with Watson playing and they’re horrible with Stoudt in the lineup because he’s terrible and the Tigers’ running game sucks (4.3 yards per rushing play against teams that would allow 5.1 yprp to an average team). The rushing attack is actually worse with Stoudt playing because opponents aren’t worried about being beat with a deep pass (Stoudt averages only 9.5 yards per completion) and they can bring a safety closer to the line of scrimmage to help defend the run. Clemson’s offense averaged only 5.2 yards per play this season (against teams that would allow 5.6 yppl to an average team) and the Tigers are 1.0 yppl worse than average with Stoudt at quarterback. Oklahoma’s defense is 0.7 yppl better than average (5.3 yppl allowed to teams that would combine to average 6.0 yppl against an average team) and they are equally good defending the run and the pass. The math model projects just 310 yards at 4.0 yards per play for Clemson in this game. The Tigers only averaged 20.7 points in the 7 games against FBS teams when Stoudt was the main quarterback and those opponents were not nearly as good defensively as Oklahoma. Clemson isn’t likely to reach 20 points and the math projects just 14 points for the Tigers.
The reason I didn’t release this game earlier is because I was waiting on an update on the status of Oklahoma’s star WR Sterling Shepard. Shepard is a difference maker, as passes targeted for Shepard have averaged 12.3 yards while passes thrown to the other 3 main wide receivers have averaged just 6.7 yards. That’s a major difference and the Oklahoma pass offense struggled for two games without Shepard even before starting quarterback Travis Knight was injured and it was even worse for the final 3 games without Knight. Shepard basically missed the final 5 games of the season (he caught one pass for 46 yards early against Iowa State in week 10 before being injured) and Knight averaged only 6.0 yards per pass play on 62 pass play without Shepard. That’s a major drop from the 8.2 yppp on 223 pass plays that Knight averaged when Shepard was healthy. Shepard did practice this week and expects to play but even he’s not sure how his groin will hold up when playing at game speed. Clemson has a great pass defense (4.5 yppp allowed to quarterbacks that would average 6.5 yppp against an average defense) and the math would project only 6.1 yppp for Oklahoma if Shepard were 100% and playing at his old level. If Shepard didn’t play then the projection would be just 4.8 yppp for Knight in this game. The battle between Oklahoma’s great rushing attack (6.4 yards per rushing play against teams that would allow 5.0 yprp) and Clemson’s great run defense (4.2 yprp allowed to teams that would average 5.4 yprp) is pretty even and the math model is projecting 4.9 yprp for the Sooners in this game. I decided to use Knight’s overall pass rating for this game, which is based on 223 pass play with Shepard and 65 pass plays without Shepard. That takes into account the chance that Shepard may not be 100%. With that assumption the math would project 340 yards at 4.9 yards per play for Oklahoma and 29 points.
Oklahoma’s edge in total yards is amplified by the much better field position that they should enjoy with their great special teams and Clemson’s sub-par special teams. The Sooners have huge edges in net punting and net kickoff yard line average and the better field position should make a difference in this game. Oklahoma also has an advantage in projected turnovers and overall the math favors the Sooners by 15 points and 42.7 total points based on the projected stats. If Shepard doesn’t play then the math would favor the Sooners by 11.3 points with a total of just 39.7 points and if Shepard is 100% and he and Knight play at their pre-injury level then I would get Oklahoma by 15.9 points with 43.7 total points. The math significantly likes Oklahoma and the Under in all 3 scenarios. I’ll take Oklahoma in a 2-Star Best Bet at -6 points or less (3-Stars at -3) and I’ll take the Under in a 2-Star Best Bet at 51 points or higher.
[h=3]Arkansas (-6) 23 Texas 16[/h]Mon Dec-29-2014 at 06:00 PM Pacific Rotation: 242 Over/Under 45.0 - Matchup Stats
The Texas Bowl executives have got to be thrilled to have secured two high profile programs whose fans are likely to make the short trip to Houston in good numbers. While the stands should be pretty full the action on the field doesn’t figure to be all that exciting given the conservative nature of both offenses and the good defense being played by both sides. These teams are actually pretty even from a yards per play perspective but Texas doesn’t move the chains nearly as well as Arkansas does and the Razorbacks are expected to have a significant possession and play advantage.
The Arkansas offense is a run-oriented attack that features two backs that have topped 1000 yards in Jonathan Williams and Alex Collins. Despite the lofty yardage totals the Razorbacks are actually pretty mediocre running the ball, as they’ve averaged a modest 5.0 yards per rushing play in 11 games against FBS opponents that would combine to allow 4.9 yprp to an average team. Quarterback Brandon Allen, meanwhile, has averaged only 6.0 yards per pass play (6.2 yppp is the national average) but he’s faced a schedule of teams that would combine to allow just 5.3 yppp to an average quarterback, so Allen has actually been significantly better than average and has thrown just 5 interceptions all season. Allen will be tested again by a Texas defense that has the 2nd best pass defense in the nation. The Longhorns have allowed just 4.6 yppp to a collection of quarterbacks that would combine to 6.6 yppp against an average team. Texas held some of the best quarterbacks in the nation in check as UCLA, Baylor, Oklahoma, Kansas State, and TCU combined to average only 5.6 yppp against the Horns. Allen is projected to average 5.1 yppp while the Arkansas rushing attack is expected to gain 4.7 yards per rushing play against a good but not great Texas run defense (0.5 yprp better than average). Overall the Razorbacks are projected to gain 347 yards at 4.9 yards per play.
Arkansas was one of the most improved defensive teams in the nation, as the Hogs struggled on that side of the ball last season but have been 0.8 yards per play better than average defensively this year (5.6 yppl allowed to teams that would combine to average 6.4 yppp against an average defense). I actually rate the Hogs at 1.0 yppp better than average since their pass defense with CB Tevin Mitchel was 0.2 yppp better if you exclude the opener against Auburn that he missed. Auburn took advantage of Mitchel’s absence in that game and beat the Razorbacks deep several times. The Arkansas run defense, meanwhile, was a bit better if I exclude the two games that #2 tackler Brooks Ellis missed in the middle of the season. The Arkansas defense faced the toughest schedule of opposing FBS offenses in the nation and that unit should not have any trouble limiting a sub-par Texas attack that 0.2 yppl worse than average in 11 games with Tyrone Swoopes at quarterback. Swoopes replaced David Ash after 1 game and was decent (5.7 yppp against teams that would allow 5.8 yppp to an average QB) while the rushing attack was also a bit below average (4.6 yprp against teams that would allow 4.8 yprp to an average team). That attack is a full 1 yard worse than that average rating of the FBS teams that Arkansas faced this season and the Hogs are projected to limit the Longhorns to just 276 yards at 4.7 yppl in this game.
Arkansas has an overall edge from the line of scrimmage due to being projected to run 13 more plays than Texas and the Razorbacks also have better special teams and an edge in projected turnovers of 0.3. Overall the math favors Arkansas by 7 ½ points, which is actually not enough of an advantage to recommend a side since the chance of covering is just 51.4% based on the historical performance of my model. However, the model also projects just 39.4 total points, which would historically be a 53.7% play on the Under (44.5 points). That is a profitable percentage, so I’ll lean Under and I have no opinion on the side.
| Washington Wizards vs. Houston Rockets (NBA) - Dec 29, 2014 8:05 PM EST | |
| Play: Point Spread: -4.0/100 Houston Rockets Pick Title: | |
| Houston is coming off a loss yesterday at San Antonio but I expect the Rockets to bounce back tonight as they head home to take on the Wizards. The Rockets blew a double-digit lead on Sunday and lost despite outshooting San Antonio 49.4 percent to 47 percent as the difference was their putrid 58.1 percent mark from the free throw line and 24 turnovers. Houston is now a half-game behind Memphis in the Southwest Division and going back, it is 6-1 ATS in its last seven games following a non-cover and 30-11 ATS in its last 41 games after committing 23 or more turnovers in its previous game. Washington is off to a great start thanks to a very easy schedule. The Wizards have played the easiest slate in the NBA and have played 18 home games compared to just 11 road games. They are a solid 6-2 against the Western Conference but this is the first road game of the season against a team from the west. Washington is just 1-4 against top ten power ranked teams and 3-8 ATS in their last 11 games against teams with a winning record. Additionally, the Wizards are 1-5 ATS in their last six games after scoring 100 points or more in their previous game. Play (710) Houston Rockets Matt is coming off PERFECT 3-0 SWEEP in the bowls Saturday and is a on a PERFECT 7-0 Bowl run! Overall, he is an AWESOME 8-2 (80%) with his bowl plays YTD! He has posted a TERRIFIC +$9,262 in profits and he is far from done! It continues on Monday with winners in ALL THREE BOWL GAMES which extends his AWESOME +$50,324 Football Run! | |
