Wednesday, October 10, 2007

Win Probabilities and Expected Wins 2007 - Week 5

Edit: I decided to remove the Hail Mary interception that Favre threw at the end of the Bears-Packers game from the box score. I have posted the resulting changes for this article, but as it does not significantly affect this week's predictions, I am not going back and fixing this week's other articles. Starting with Week 6, however, the stats will reflect the removed play. I have also done this with the bizarre Hail Mary play at the end of regulation in the week 1 MIA@WAS matchup.

Well, so far I'm 11-1 in predicting this week's game (with adjusted stats) and got some of the margins close to dead on. I don't know how many people saw the Chargers killing Denver either. Denver had a 0.0134% chance of winning that game. Damn! And to end the weekend, the predictions based on adjusted stats go 12-2.

The following table lists the probabilities that the home team would win each game given the rushing averages, passing averages (includes sack yards lost and sacks as pass attempts), sack rates, third down conversion rates, penalty yards, and interception and fumble rates (by number of completions and rushing plays) for both teams. A logistic regression model trained on 1996-2006 data computes the probabilities. The final score margin is Home team points - Away team points. The sum of these probabilities of victory form the expected win totals in the second table to be added after the Monday night game. The first table will be updated as more box scores come in.

Monday night update In spite of Romo's 6 turnovers, a preliminary look at the stats shows that the Cowboys still had a 54% chance of winning that game. I'll update sometime tomorrow with the final numbers. In every other way, the Cowboys were more efficient, especially in pass efficiency (surprisingly enough).

Home TeamAway TeamP(Home Team Won) (%)Final Score Margin

Teams that "should have" won but didn't
St. Louis Rams (vs. Cardinals), Win Prob = 74.3333% Arizona's 2 fumbles, both of which they recovered, were the difference between a P(Rams win)=74.333% and a P(Rams win)=49.99% In other words, this game was very evenly matched. The Cardinals had better numbers but just good enough to neutralize homefield advantage.

New Orleans Saints (vs. Panthers), Win Prob=68.7236% Same story here. Carolina fumbled twice but recovered both fumbles. Excluding the 2 fumbles, P(Saints win)=34.92%. David Carr somehow managed to average 7 yards a pass.

Green Bay Packers (vs. Bears), Win Prob=53.3206% They were about equal in yards per pass, but the Packers had a huge advantage in rushing averages (5.5 vs. 2.5) and third down conversions (6-14, 3-14). But the Bears had short field position on their TD drive in the 3rd quarter off an interception and were more efficient in the red zone. It speaks poorly on the Bears that they couldn't do better on 4 turnovers. That said, the Bears did manage 7.5 yards per pass, which is pretty good even if against a mediocre pass defense.

1 comment:

Anonymous said...

As a professional handicapper, I am curious as to whether your analysis takes current "vegas lines" into account when making predictions. Predicting the winner of a game is a great ability, but predicting the winner against the posted line is what turns that ability into profit!