"Who was that? Is that the blind girl? Helen Keller ... I don't know who the fuck Anne Frank is. I'm mad right now. Fuck it. I'm not as swift as I usually am.
-Channing Crowder
OK, I know I’m two weeks late with the final revisions for a model predicting total wins in a season so in fairness to this project I’m going to use the previously posted projects to judge my success. However, I was excited to discover that by deleting 7 seasons that were acting as outliers I was able to bump the adjusted R2 up to .293 from .19. If you’ll recall the goal of this project was to get above 56% of my pick correct and I was hoping to accomplish that by having as high of an adjusted R2 as possible. The observations deleted were all extreme jumps or drops in wins such as the Dolphins 1 win 2007 season or the Patriots 16 win 2007 season. Below is the new and hopefully better model.
| Team | Revised | Lines 9/5 |
| Dolphins | 10 | 7.5 |
| Bills | 7 | 5.5 |
| Patriots | 11 | 11.5 |
| Jets | 8 | 10 |
| Ravens | 8 | 10 |
| Bengals | 6 | 5.5 |
| Browns | 9 | 7 |
| Steelers | 5 | 10.5 |
| Texans | 12 | 9 |
| Colts | 11 | N/A |
| Jags | 8 | 6.5 |
| Titans | 7 | 6.5 |
| Broncos | 7 | 6 |
| Chiefs | 6 | N/A |
| Raiders | 10 | 6.5 |
| Chargers | 13 | 10 |
| Cardinal | 6 | 7.5 |
| Falcons | 8 | 10 |
| Panthers | 4 | 4.5 |
| Bears | 5 | 8 |
| Cowboys | 6 | 9 |
| Lions | 9 | 8 |
| Packers | 8 | 11.5 |
| Vikings | 5 | 7 |
| Saints | 11 | 10 |
| Giants | 9 | 9 |
| Eagles | 11 | 10.5 |
| 49ers | 9 | 7.5 |
| Seahawks | 8 | 6 |
| Rams | 3 | 7.5 |
| Buccanee | 5 | 8 |
| Redskins | 4 | 6 |
As you can see there are still some goofy predictions. This model really didn’t like the Steelers or actually the NFC in general. And based on the first 2 weeks it really seems to over value my Dolphins. But if my thinking is correct this model should be more accurate in predicting the over/unders for total wins.
Anyways, below is the predictive model I’ll judge my success by. I’ll keep track of both how the model did and how I would’ve betted. Keep in mind the Colts and Chiefs would not have been betted on because there was no line.
| Division | Prediction | Lines 9/5* | My Bet |
| AFC East | |||
| Dolphins | 8 | 7.5 | Under |
| Bills | 6 | 5.5 | Over |
| Patriots | 11 | 11.5 | Under |
| Jets | 10 | 10 | Over |
| AFC North | |||
| Ravens | 10 | 10 | Under |
| Bengals | 6 | 5.5 | Over |
| Browns | 6 | 7 | Under |
| Steelers | 9 | 10.5 | Under |
| AFC South | |||
| Texans | 11 | 9 | Over |
| Colts | 12 | N/A | N/A |
| Jags | 9 | 6.5 | Over |
| Titans | 6 | 6.5 | Under |
| AFC West | |||
| Broncos | 6 | 6 | Over |
| Chiefs | 10 | N/A | N/A |
| Raiders | 7 | 6.5 | Over |
| Chargers | 12 | 10 | Over |
| NFC West | |||
| 49ers | 6 | 7.5 | Under |
| Seahawks | 9 | 6 | Over |
| Rams | 4 | 7.5 | Over |
| Cardinals | 8 | 7.5 | Over |
| NFC South | |||
| Falcons | 9 | 10 | Under |
| Panthers | 4 | 4.5 | Under |
| Buccaneers | 5 | 8 | Under |
| Saints | 11 | 10 | Under |
| NFC North | |||
| Lions | 6 | 8 | Under |
| Packers | 9 | 11.5 | Under |
| Vikings | 5 | 7 | Under |
| Bears | 8 | 8 | Over |
| NFC East | |||
| Giants | 9 | 9 | Over |
| Eagles | 11 | 10.5 | Over |
| Redskins | 5 | 6 | Under |
| Cowboys | 6 | 9 | Under |
As of week 2, I remain optimistic that me or the model will get above 56% of the bets correct.
In future updates of this model (probably not until summer), I hope to control for multicollinearity among variables, individual years nested within teams, and more extensive outlier analysis.
Finally next week, I hope to unveil weekly predictions of whether teams cover the like or not. However, this seems like its going to be a particularly intricate model and multiple regression probably won’t cut it. So if anyone has any experience with multi-level or nested modeling and would like to get on board with this please let me know.
No comments:
Post a Comment