Predictions 2007 Week 15 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.
Quick Summary of Accuracy
LAST WEEK
Linear regression, opponent adjusted: 11 out of 16
Logistic regression, opponent adjusted: 11 out of 16
Linear regression, unadjusted: 11 out of 16
Logistic regression, unadjusted: 11 out of 16
YEAR SO FAR
Linear regression, opponent adjusted: 67.6136%
Logistic regression, opponent adjusted: 66.4773%
Linear regression, unadjusted: 65.3409%
Logistic regression, unadjusted: 63.6364%
Full accuracy numbers after the jump...
Predictions based on opponent-adjusted stats | ||
Game | Predicted Final Score Margin | P(Home Team Wins) |
BAL @ MIA | 0.9523 | 54.7323 |
NYJ @ NE | 12.9239 | 85.0081 |
BUF @ CLE | 2.8910 | 59.0441 |
JAX @ PIT | 4.8960 | 65.9684 |
DEN @ HOU | 1.3260 | 51.3723 |
TEN @ KC | -0.5546 | 47.5495 |
IND @ OAK | -4.5726 | 32.4282 |
DET @ SD | 5.6125 | 65.6524 |
PHI @ DAL | 7.6678 | 73.9214 |
WAS @ NYG | 2.7065 | 59.4838 |
CHI @ MIN | 7.0065 | 74.7751 |
SEA @ CAR | -1.2256 | 47.2636 |
ARI @ NO | 3.7373 | 60.2355 |
ATL @ TB | 7.4745 | 75.9095 |
GB @ STL | -3.7296 | 37.5372 |
CIN @ SF | -1.5455 | 46.8054 |
Predictions based on unadjusted stats | ||
Game | Predicted Final Score Margin | P(Home Team Wins) |
BAL @ MIA | 0.3806 | 52.8320 |
NYJ @ NE | 12.4678 | 84.4442 |
BUF @ CLE | 4.4071 | 63.8429 |
JAX @ PIT | 5.7254 | 68.9211 |
DEN @ HOU | 1.5179 | 51.4334 |
TEN @ KC | -0.4387 | 47.0516 |
IND @ OAK | -3.9297 | 34.0468 |
DET @ SD | 5.7060 | 65.5140 |
PHI @ DAL | 8.1466 | 75.0314 |
WAS @ NYG | 3.1209 | 60.5767 |
CHI @ MIN | 9.0740 | 79.0734 |
SEA @ CAR | -1.8746 | 45.7222 |
ARI @ NO | 2.6761 | 57.8857 |
ATL @ TB | 7.6254 | 77.1451 |
GB @ STL | -3.4341 | 38.8902 |
CIN @ SF | -1.9900 | 46.0685 |
OPPONENT-ADJUSTED
Linear regression
Win prediction accuracy: 67.6136%
Mean absolute error: 11.5177 points
Correlation with result: 0.4033
% of games predicted as home team wins: 72.7273%
Logistic regression
Win prediction accuracy: 66.4773%
Correlation with result: 0.3946
% of games predicted as home team wins: 75.0000%
Win prediction accuracy when 20%≤P(home team wins)<30%: 100.0000
Win prediction accuracy when 30%≤P(home team wins)<40%: 81.2500
Win prediction accuracy when 40%≤P(home team wins)<50%: 68.0000
Win prediction accuracy when 50%≤P(home team wins)<60%: 47.7273
Win prediction accuracy when 60%≤P(home team wins)<70%: 72.9167
Win prediction accuracy when 70%≤P(home team wins)<80%: 65.6250
Win prediction accuracy when 80%≤P(home team wins)<90%: 87.5000
UNADJUSTED
Linear regression
Win prediction accuracy: 65.3409%
Mean absolute error: 11.5080 points
Correlation with result: 0.3907
% of games predicted as home team wins: 71.5909%
Logistic regression
Win prediction accuracy: 63.6364%
Correlation with result: 0.3830
% of games predicted as home team wins: 72.1591%
Win prediction accuracy when 10%≤P(home team wins)<20%: 100.0000
Win prediction accuracy when 20%≤P(home team wins)<30%: 75.0000
Win prediction accuracy when 30%≤P(home team wins)<40%: 68.7500
Win prediction accuracy when 40%≤P(home team wins)<50%: 61.5385
Win prediction accuracy when 50%≤P(home team wins)<60%: 44.4444
Win prediction accuracy when 60%≤P(home team wins)<70%: 64.5833
Win prediction accuracy when 70%≤P(home team wins)<80%: 77.4194
Win prediction accuracy when 80%≤P(home team wins)<90%: 63.6364
Win prediction accuracy when 90%≤P(home team wins)<100%: 100.0000
No comments:
Post a Comment