Friday, October 23, 2009

Week 8: Predictions

Below are my predictions for week 8. You'll notice that two of the games have already happened: UTEP-Tulsa and UNC-FSU. I got one right and one wrong, which is about how the numbers said I would do. The favorites for each game are highlighted in green, and the final column is the probability that the favored team will win.

Home Visitors Odds
UTEP17 Tulsa24 78
North Carolina17 Florida St.24 53
Army17 Rutgers24 88
Alabama24 Tennessee17 92
Arizona24 UCLA17 79
Arkansas St.24 FL-International17 72
Baylor17 Oklahoma St.24 74
Bowling Green17 Central Michigan24 54
BYU17 TCU24 77
California17 Washington St.14 98
Cincinnati24 Louisville17 93
Colorado St.24 SDSU17 84
Duke24 Maryland17 76
Eastern Michigan17 Ball St.24 78
Hawaii24 Boise St.31 98
Houston38 SMU31 88
Kansas17 Oklahoma24 84
Kansas St.31 Colorado24 56
Kentucky24 LA-Monroe17 91
LA-Lafayette24 FL-Atlantic31 50
LSU24 Auburn17 72
Marshall24 UAB17 74
Miami-FL17 Clemson24 55
Miami-OH17 Northern Ill.24 95
Michigan17 Penn State24 85
Michigan St.17 Iowa24 73
Middle Tennessee24 Western Kentucky17 89
Mississippi31 Arkansas24 72
Mississippi St.24 Florida31 97
Missouri17 Texas24 83
Navy24 Wake Forest17 81
Nebraska24 Iowa St.17 95
Nevada31 Idaho24 82
New Mexico24 UNLV31 56
New Mexico St.24 Fresno St.31 95
Northwestern24 Indiana17 70
Notre Dame24 Boston College17 76
Ohio24 Kent St.17 84
Ohio St.24 Minnesota17 95
Pittsburgh24 South Florida17 72
Purdue24 Illinois17 87
Rice17 UCF24 66
South Carolina17 Vanderbilt14 91
USC24 Oregon St.17 95
Southern Miss.31 Tulane24 97
Stanford24 Arizona St.17 63
Syracuse24 Akron17 71
Texas Tech47 Texas A&M28 95
Toledo17 Temple24 59
Troy31 North Texas24 94
Utah17 Air Force14 77
Utah St.17 LA Tech24 53
Virginia24 Georgia Tech17 60
Washington17 Oregon24 86
Western Michigan31 Buffalo24 60
West Virginia24 Connecticut17 79

You'll notice that the majority of the games have a projected final score of 24-17. This is due to some expectation maximization I perform in order to make the predicted outcomes more realistic. For example, a final score of 25-18 may be what the raw formula projects, but in reality is uncommon. The algorithm looks for "nearby" scores that are more likely and settles on those. I'll probably revisit this as the season goes on since it masks some of the differences between teams.