Predictions 2007 Week 9 and Accuracy Report

The predictions are based on yards per play on rushing and passing plays (sacks and sack yards included), sack rates, third down conversion rates, penalty yards per game, and fumble and interception rates. The models are trained on the 1996-2006 seasons. Linear regression predicts the final score margin (home team points - away team points). Logistic regression estimates the probability that the home team will win.


The predictions are based on yards per play on rushing and passing plays (sacks and sack yards included), sack rates, third down conversion rates, penalty yards per game, and fumble and interception rates. The models are trained on the 1996-2006 seasons. Linear regression predicts the final score margin (home team points - away team points). Logistic regression estimates the probability that the home team will win.

Predictions based on opponent-adjusted stats
Game	Predicted Final Score Margin	P(Home Team Wins)
CIN @ BUF	1.0045	52.7523
WAS @ NYJ	-3.2205	39.4645
SEA @ CLE	1.3537	53.5113
BAL @ PIT	7.5392	75.2282
NE @ IND	2.9047	61.2232
CAR @ TEN	5.5641	67.9738
GB @ KC	-0.9309	48.7180
HOU @ OAK	1.1558	52.2015
DAL @ PHI	-1.7509	43.1141
DEN @ DET	2.1942	55.9526
SD @ MIN	1.3128	57.3044
SF @ ATL	5.2070	66.1927
JAX @ NO	-0.4927	49.1553
ARI @ TB	7.8599	73.8157

Predictions based on unadjusted stats
Game	Predicted Final Score Margin	P(Home Team Wins)
CIN @ BUF	1.3931	55.0238
WAS @ NYJ	-2.9390	40.8522
SEA @ CLE	1.2344	53.1285
BAL @ PIT	7.5106	75.0971
NE @ IND	1.9491	58.4803
CAR @ TEN	5.0706	67.4653
GB @ KC	-0.2492	51.1181
HOU @ OAK	2.4255	54.9057
DAL @ PHI	-0.5842	46.7032
DEN @ DET	6.2572	66.6293
SD @ MIN	0.7142	53.3107
SF @ ATL	5.6014	67.5739
JAX @ NO	-0.0740	50.9851
ARI @ TB	6.0552	70.9135

Quick Summary of Accuracy
LAST WEEK
Linear regression, opponent adjusted: 6 out of 13
Logistic regression, opponent adjusted: 5 out of 13
Linear regression, unadjusted: 8 out of 13
Logistic regression, unadjusted: 8 out of 13

YEAR SO FAR
Linear regression, opponent adjusted: 66.6667%
Logistic regression, opponent adjusted: 65.4762%
Linear regression, unadjusted: 61.9048%
Logistic regression, unadjusted: 63.0952%

Full accuracy numbers after the jump...

OPPONENT ADJUSTED
Linear regression
Win prediction accuracy: 66.6667%
Mean absolute error: 11.2464 points
Correlation with result: 0.3735
% of games predicted as home team wins: 69.0476%
Logistic regression
Win prediction accuracy: 65.4762%
Correlation with result: 0.3749
% of games predicted as home team wins: 72.6190%
Win prediction accuracy when 20%≤P(home team wins)<30%: 100.0000
Win prediction accuracy when 30%≤P(home team wins)<40%: 88.8889
Win prediction accuracy when 40%≤P(home team wins)<50%: 58.3333
Win prediction accuracy when 50%≤P(home team wins)<60%: 50.0000
Win prediction accuracy when 60%≤P(home team wins)<70%: 76.4706
Win prediction accuracy when 70%≤P(home team wins)<80%: 58.8235
Win prediction accuracy when 80%≤P(home team wins)<90%: 80.0000

UNADJUSTED
Linear regression
Win prediction accuracy: 61.9048%
Mean absolute error: 11.1272 points
Correlation with result: 0.3770
% of games predicted as home team wins: 64.2857%
Logistic regression
Win prediction accuracy: 63.0952%
Correlation with result: 0.3777
% of games predicted as home team wins: 63.0952%
Win prediction accuracy when 10%≤P(home team wins)<20%: 100.0000
Win prediction accuracy when 20%≤P(home team wins)<30%: 100.0000
Win prediction accuracy when 30%≤P(home team wins)<40%: 66.6667
Win prediction accuracy when 40%≤P(home team wins)<50%: 50.0000
Win prediction accuracy when 50%≤P(home team wins)<60%: 42.8571
Win prediction accuracy when 60%≤P(home team wins)<70%: 70.5882
Win prediction accuracy when 70%≤P(home team wins)<80%: 78.5714
Win prediction accuracy when 80%≤P(home team wins)<90%: 42.8571
Win prediction accuracy when 90%≤P(home team wins)<100%: 100.0000