Game Predictions Week 8

Game probabilities for week 8 NFL games are listed below. The probabilities are based on an efficiency win model explained here and here. The model considers offensive and defensive efficiency stats including running, passing, sacks, turnover rates, and penalty rates. Team stats are adjusted for previous opponent strength.


Edit: I was just reminded here that the NYG - MIA game is in London this week. I have revised the game probability to reflect this. The result was a swing of 0.06, so MIA's probability of winning went from 0.24 to 0.18. The Giants have benefited from another favorable scheduling aberration recently, when they hosted NO for one of their post-Katrina home games.

Also, if the SD-HOU game is played at an alternate site, the new probabilities would still have SD as the favorites at 0.64 to 0.36.

  • Spread The Love
  • Digg This Post
  • Tweet This Post
  • Stumble This Post
  • Submit This Post To Delicious
  • Submit This Post To Reddit
  • Submit This Post To Mixx

5 Responses to “Game Predictions Week 8”

  1. Derek says:

    Our predictions are very far apart this week it seems, though I'd rather go with yours for the 2nd week in a row. How are you adjusting for opponent strength for your predictions? It's not the GWP is it?

  2. Brian Burke says:

    I'm surprised your model likes MIA at home over the Giants. I'm guessing its due to the intraconference HFA factor you include. Otherwise, we're pretty close. For some reason my model really hates CHI. That's the biggest doubt I have this week.

    Regarding opponent strength, here's what I do: I calculate generic win probability for each team. That's simply an application of the model for each team vs. a notional league-average team. I also set the home field variable to 0.5, so it's theoretically at a neutral site.

    I then average the GWP for all the opponents of each team. Then I calculate what the log odds ratio would need to be given that average opponent win probability. (I basically go backwards through the logistic probability math). I add the resulting log odds from the log odds for each team in a given game. Then I recalculate the odds ratio (going forwards in the logistic math again).

    For some teams like BAL it made a big difference. Without the adjustment, BAL would have been a big favorite against BUF. But the opponent adjustment put BUF as the solid favorite.

    The opponent adjustments diminish as the season progresses because each team's schedule tends to even out.

  3. Tarr says:

    You should probably do the same "neutral site" adjustment for the HOU-SD game, which is likely to be in Phoenix or Dallas.

  4. Derek says:

    Actually, the weird part is that I don't include the interconference aspect of it at all in my prediction system. I keep track of interconference results as a supplement to the predictions. Miami ranks #2 in rushing offense, which we both measure by yards per carry, and is only slightly below average in terms of passing offense. In fact, Miami is not far behind the Giants in most of the inputs I use, but they are behind. I guess it's not enough to overcome HFA. All of the relevant stats tables are here. I'm not surprised your system hates Chicago. Hester's saved their asses in both wins probably. They rank 31, 26, 23, and 23 in yards per carry gained, ypc allowed, yards per pass gained, ypp allowed, respectively. It'll take a few more weeks of Brian Griese performing at a high level for that to start evening out.

    The biggest difference in our opponent adjustments seems to be that my opponent adjustments are on an individual stat level and yours seem to be on a team level.

  5. Brian Burke says:

    Tarr-Good point, I'll add another edit.

    Derek-The weird thing about the 'Fins is that their offense was fine with Trent Green. It's their defense that has collapsed.

Leave a Reply