Dr. Bob
FLORIDA (-5.0) 42 Oklahoma 32
National Championship
08-Jan-09 05:15 PM Pacific Time
Oklahoma averaged a record number of points this season (53.8 per game in 12 games against Division 1A opposition), but there is mounting evidence that the Big 12 South is an overrated division with worse defensive units than previously thought. My math model is very good at sorting out differences in conference strengths and my model mostly picked against the teams from the Big 12 South, as it correctly picked Mississippi to upset Texas Tech and Ohio State to upset Texas. Texas managed to beat the Buckeyes with a touchdown with 15 seconds remaining, but I easily won my bet on Ohio State and Mississippi scored at will in their win against Texas Tech. Ole Miss racked 47 points on 521 yards at 7.3 yards per play against Texas Tech, Oregon scored 42 points on 566 yards at 8.0 yppl against Big 12 South rep Oklahoma State, and Ohio State gained 379 yards at 5.9 yppl against Texas. Oklahoma’s offense is still great, but it certainly doesn’t look quite as amazing after seeing how the Big 12 South’s other teams performed defensively in the bowl games.
The Sooners averaged an incredible 7.2 yppl in 12 games against Division 1A opponents that would combine to allow 5.1 yppl to an average team and they scored 35 points against a TCU defense that I rate as the 2nd best in the nation behind USC. That 35 points was the lowest total of the year for the Sooners, who also scored 35 points against a good Texas defense, so the Sooners have proven that they can score a good number of points even against very good defensive teams. Oklahoma will be facing a very good defensive team tonight, as Florida yielded just 4.5 yppl to 12 Division 1A teams that would combine to average 5.7 yppl against an average defensive team. Florida didn’t face an offense as good as Oklahoma’s this season (because Oklahoma had my highest rated offense), but the Gators played at about the same level defensively against the good attacks of Mississippi, Georgia, and Alabama as they did overall and gave up an average of just 20.3 points in those 3 games. Oklahoma’s offense is not going to be quite as good without star RB DeMarco Murray, who will be missed in the passing game more than in the running game (#2 back Chris Brown is just as good), but I still rate the Sooners’ attack at 2.0 yppl better than average without Murray after I take out the stats of the backup quarterbacks – so Oklahoma does have a solid 0.8 yppl advantage over Florida’s good defense.
Florida’s offense is just as good as Oklahoma’s offense on a compensated yards per play basis, as the Gators averaged 7.0 yppl with Tim Tebow in the game while facing teams that would combine to allow 5.0 yppl to an average attack. Oklahoma’s defense, meanwhile, gave up 5.4 yppl to teams that would average 5.8 yppl against an average team, so the Sooners are just 0.4 yppl better than average defensively – giving Florida a 1.6 yppl advantage when they have the ball. Oklahoma’s defense has the same rating as the average of the other 12 Division 1A teams that the Gators faced this season and Florida averaged 43 points in those games and should score around that number in this game.
With both offenses rated about the same from a compensated yards per play perspective and Florida having a much better defense, the Gators are clearly the better team from the line of scrimmage. Oklahoma scored an incredible 53.8 points per game while Florida averaged 43.1 ppg in division 1A games, but the Sooners also averaged 79 plays in their hurry up offense while Florida averaged only 61 plays in the more traditional SEC. If the pace of Florida’s games, which averaged just 123.0 total plays, were as fast-paced as the Oklahoma games (150.9 total plays) then Florida would have averaged 53 points per game – so the Sooners are not really better offensively than the Gators no matter how you look at it. But, Florida is certainly better than Oklahoma defensively even if you look at points and adjust for pace. Florida is also much better in special teams, as the Sooners gave up 4 kickoff return touchdowns and are below average overall in special teams while Florida has good special teams. Oklahoma also isn’t likely to be +2 in turnover margin in this game, which is what they averaged during the season, as Florida averaged a +1.6 in turnover margin in D1A games. In fact, Florida actually has a projected edge in turnovers. Tim Tebow is less likely to get intercepted (he threw just 2 interceptions all season while Oklahoma quarterback Sam Bradford threw 6 and Florida’s defense had a 5.6% interception rate while Oklahoma’s defense had a 3.6% interception rate. A big part of Oklahoma’s turnover difference was purely luck, as the Sooners were +13 in fumble margin with just 2 lost fumbles all season compared to 15 fumbles recovered by their defense. Fumbles are 90% random in college football, so most of that +13 in fumbles is just random good luck that is not likely to project forward into this game. In other words, the fumbles should be about even in this game and Florida is less likely to throw interceptions. My math model actually projects Florida was a +0.4 turnover margin in this contest.
My math model favors Florida by 14 points in this game after putting it all together and I actually admit that is too high, as it does not take into account Oklahoma’s ability to convert 4th downs into first downs or touchdowns and to score in the redzone at a higher rate than normal. To take those factors into account I can use a compensated points model. I’ll use a simple version to give you the concept. Oklahoma out-scored their Division 1A opponents by an average of 27.4 points while facing a schedule of teams that was 7.7 points better than average, so they are +35.1 points better than average based on that. Florida, meanwhile, out-scored their 1A opponents by 30.8 points (43.1 to 12.3) while facing a schedule that was 5.8 points better than average, so the Gators are +36.6 points better than average using that simple analysis. Those numbers would favor Florida by 1 ½ points, which was the opening line on this game.
What the oddsmakers failed to take into account was pace, as Oklahoma wouldn’t have out-scored their opponents by as much as they did had they played their games at a normal pace, while Florida would have out-scored their opponents by more than they did had they not played at such a slow pace. I mentioned earlier that Oklahoma averaged 150.9 total plays per game and that Florida’s games averaged 123.0 total plays. My math model projects 140.2 plays for this game, so I want to adjust the point margins for the projected pace of this game. If Oklahoma had averaged 140.2 plays then they would have out-scored their opponents by 25.5 points rather than 27.4 points (27.4/150.9 x 140.2 = 25.5), which would make them +33.2 points when incorporating their +7.7 schedule strength. Florida, meanwhile, would have out-scored their opponents by 35.1 points had they averaged 140.2 plays per game instead of 123 total plays and that would make the Gators +40.9 after adding in their +5.8 schedule strength. So, instead of Florida being better by 1.5 points, the Gators would have been 7.7 points better than Oklahoma if each team averaged the 140 total plays that are expected in this game. That projection of Florida by 7.7 points does not take into account that Florida is expected to have a 0.4 turnover edge in this game rather than being 0.42 turnovers worse per game than Oklahoma. Oklahoma’s actual +2.0 TO margin per game would have been +1.86 had they played at a 140 play pace and Florida’s +1.58 TO margin per game would have been +1.80 had the Gators played at a 140 play pace, so part of the random turnover difference was already incorporated while adjusting the scoring margin for pace. But, the difference between my projected turnover margin of 0.4 turnovers in favor of Florida and the 0.06 pace-adjusted turnover margin in favor of Oklahoma is 0.46 turnovers, which is equivalent to 1.7 points in Florida’s favor. We can add that one to our previous pace-adjusted points margin prediction of Florida by 7.7 points, which gives us an adjusted scoring margin prediction of Florida by 9.4 points. That’s not as much as my math model prediction of Florida by 14 points, but it’s a more realistic number and is the minimum that I would favor the Gators in this game.
In addition to the line value, Florida coach Urban Meyer has a proven track record in big games and his teams at Bowling Green, Utah and Florida are an incredible 27-4 ATS against non-conference opponents, including 4-1 ATS in Bowl games. Oklahoma’s Bob Stoops, meanwhile, is 3-6 ATS in Bowl games, including 0-4 straight up and 0-4 ATS in his last 4 BCS Bowls.
I’ll consider Florida a Strong Opinion at -6 points or less.
Using a pace adjusted approach on total scoring would project 77 total points scored in this game and my math model projects 74 total points, so I’ll lean over 72 points or lower.