DR. BOB
I have a Strong Opinion on the Under (47 1/2) in the BCS Championship game and no opinion on the side.
BCS Championship Game
LSU (-3.5) 21 Ohio St. 17
05:00 PM Pacific, 07-Jan-08
It?s pretty well known that Ohio State is 0-9 lifetime in bowl games against SEC teams, including last year?s humiliating loss to Florida in the BCS Championship game, but that really doesn?t mean much to me. The SEC is perceived to be a much better conference than the Big 10 in recent years, but Big 10 teams are actually 6-4 straight up and 7-3 ATS against SEC teams in bowl games the last 4 years (1-1 this year with Wisconsin losing to Tennessee and Michigan beating up on Florida). The fact that Michigan beat Florida despite being -4 in turnover margin is an indication that the difference between the quality of the Big 10 and the SEC isn?t as much as it is perceived to be. So, forget about the hype of the SEC being a dominant conference and focus on the teams. My math model adjusts for strength of opponents, which is affected by the strength of the conference a team plays in and the math on this game reveals that Ohio State was no worse than LSU this season.
The question is whether Ohio State can compete with LSU in this game, as the Tigers are healthier on defense and at the receiver position than they were for parts of the regular season. LSU?s offense was very inconsistent this season, playing 4 games in which they averaged less than 5.0 yards per play and having 4 games in which they averaged 7.0 yppl or more. Overall the Tigers averaged 6.1 yppl against teams that would combine to allow 5.2 yppl to an average team. The rushing attack was outstanding, averaging 236 yards at 5.9 yards per rushing play (against teams that would allow 4.6 yprp), but quarterback Matt Flynn was mediocre. Flynn averaged only 5.8 yards per pass play against teams that would allow 5.8 yppp to an average quarterback and the Tigers appear to be a better team when Ryan Perrilloux is at quarterback. Perrilloux averaged 8.4 yppp on 79 pass plays this season (against teams that would allow 6.3 yppp) and that includes 7.6 yppp in the SEC Championship game against Tennessee when Flynn was injured (Tennessee would allow 6.3 yppp to an average team). Flynn was hindered a bit by early season injuries to the receivers Demetrius Byrd and Early Doucet, as Byrd missed games 1 and 4 and Doucet was out for 5 games from game 3 through 7. Flynn?s numbers were better from game 8 on when his entire receiving corps was intact, as Flynn averaged 6.5 yppp his final 5 games against teams that would allow 5.8 yppp to an average QB. I?ll assume that?s the level Flynn will play in this game and that Perrilloux probably won?t see much time, unless the Tigers need him near the goal line for a running option. Overall the Tigers rate at 1.0 yppl better than average on offense heading into this game.
Ohio State?s defense rates at 1.9 yppl better than average, yielding just 3.6 yppl for the season against teams that would average 5.5 yppl against an average stop unit. Ohio State completely shut down every better than average pass attack that they faced this season, and only Illinois averaged more than 5.0 yppp against the Buckeyes. The reason that the Illini managed to average 6.0 yppp was because they couldn?t sack the mobile Juice Williams (Williams averaged only 6.4 yards per pass attempt but was sacked just once). Matt Flynn is not that mobile, so he won?t be able to buy time to find open receivers and there won?t be too many open receivers for Flynn to find. Flynn had mixed results against the good pass defenses he?s faced this year, averaging 8.8 yppp against Virginia Tech and 8.1 yppp against Auburn, but managing just 3.0 yppp against South Carolina and 3.2 yppp against Kentucky. Ohio State?s pass defense is so much better than any team that LSU faced this year (the Buckeyes allowed just 3.6 yppp to teams that would average 6.2 yppp), but Flynn is capable of posting decent numbers and is also potentially going to be horrible. LSU?s offensive production will probably depend on their ability to run against an Ohio State defense that gave up just 3.6 yards per rushing play (to teams that would average 5.1 yprp against an average defense). Ohio State played two teams that are very good running teams and the Buckeyes gave up a combined 5.1 yprp to Washington and Illinois (who would combine to average 6.1 yprp), so they were only 1.0 yprp better than average against good running teams. Those numbers are misleading, however, as Washington and Illinois both have running quarterbacks that got a lot of those yards. Ohio State held All-Big 10 back Rashard Mendenhall to just 88 yards on 26 carries (3.4 ypr) and held Washington star Louis Rankin to 42 yards on 14 carries as quarterback Jake Locker scrambled for 107 yards on 13 rushes. Flynn is a decent runner, but certainly not in the class of Williams and Locker, so don?t imply that Ohio State?s relative troubles with Washington and Illinois apply to LSU?s good rushing attack. LSU may actually run for less than expected in this game given that the Tigers, who are 1.3 yprp better than average overall, were only 0.9 yprp better than average against the 7 better than average run defenses that they faced (5.2 yprp against Miss State, Virginia Tech, Florida, Auburn, Alabama, Arkansas, and Tennessee ? who would combine to allow 4.3 yprp to an average team). Ohio State?s defense is 0.9 yppl better than LSU?s offense and my math model projects just 303 yards at 4.5 yppl for the Tigers in this game.
The key to this game is how well LSU?s defense plays. The Tigers were a dominating unit for most of the season, but star DT Glenn Dorsey (the college defensive player of the year) was banged up late in the season and LB Barry Beckwith missed a couple of games late in the season. LSU?s defense was noticeably weaker in their final 3 games, giving up a combined 6.4 yppl to Mississippi, Arkansas and Tennessee. Reports are that the Tigers are fully healthy on defense and will be back to their dominating form for this game. LSU was 1.2 yppl better than average defensively for the season (4.3 yppl allowed to teams that would average 5.5 yppl against an average team), but the Tigers were 1.6 yppl better than average in their first 10 games and I?ll assume that they will be that good in this game. If that is the case then the Tigers have a 1.3 yppl advantage over an Ohio State offense that was just 0.3 yppl better than average with Todd Boeckman at quarterback. Ohio State was 0.6 yppl better than average against the 3 good defensive teams that they faced (5.3 yppl against Penn State, Illinois, and Michigan, who would allow 4.7 yppl to an average team). However, the Buckeyes averaged only 4.0 yppl against Michigan, who was the best defensive team that they faced this season. Ohio State should have an even tougher time moving the ball against LSU?s defense and my math model projects just 274 yards at 4.1 yppl is the Tigers defense is indeed as good as it was in their first 10 games (which I?m assuming it is). Boeckman is also a bit turnover prone (12 interceptions in just 273 passes, 4.4%) and the math model projects a 5.3% interception rate for Boeckman in this game, so the Buckeyes could be in trouble if they get down by a couple of scores and are forced to throw more than normal against LSU?s attacking defense.
Overall, my math model favored LSU by just ? a point using stats for each team for the entire season (with 2 points being give to LSU for playing in Tiger-friendly New Orleans), but that projection changes after adjusting Flynn?s stats for games in which he had all his receivers and using only LSU?s first 10 games on defense. I now have LSU by 4 points with a total of just 38 ? points, as it is pretty clear that the defenses have big advantages in this game and scoring will not be easy (unless there are some defensive touchdowns). Underdogs of 3 points or more are 6-2 straight up in BCS Championship games, including Ohio State?s 31-24 upset of Miami-Florida as a 12 point dog in 2003. The Buckeyes will certainly be motivated after hearing how they can?t compete with an SEC team for the past 6 weeks and Ohio State is better equipped to compete this season given that they?re defense is the best in the nation this year after being a very young group last season. Remember, the Buckeyes? defense was 0.3 yppl better than LSU?s defense was when it was healthy. I can see both teams winning this game and my math model suggests that there is no value in betting the side in this game. The better play is on the under given that both defenses have significant advantages and I will consider the UNDER a Strong Opinion in this game at 45 points or higher and I have no opinion on the side.