Game Predictions Week 7

Game probabilities for week 7 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.



















VprobVisitorHomeHprob
0.19ARIWAS0.81
0.54ATLNO0.46
0.38BALBUF0.62
0.15MINDAL0.85
0.92NEMIA0.08
0.09SFNYG0.91
0.86TBDET0.14
0.51TENHOU0.49
0.64KCOAK0.36
0.26NYJCIN0.74
0.11CHIPHI0.89
0.12STLSS0.88
0.48PITDEN0.52
0.73INDJAX0.27

  • 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

4 Responses to “Game Predictions Week 7”

  1. Derek says:

    Interesting how many disagreements our models have this week, but I'm more confident in your picks because my model is overconfident in the home field advantage for intradivision games, which has been an actual disadvantage so far this season. Maybe the trend will start to reverse this weekend?

  2. Anonymous says:

    Last week’s upset specials (my phrasing, not yours), WAS and MIA, were 0-1-1 ATS. To be fair, you did suggest staying away from the Lemon-flavored Fish.

    This week's game predictions point to five possible upsets:
    #25 BAL (-3) at #19 BUF
    #3 TB at #27 DET (-2)
    #20 ATL at #28 NO (-9)
    #18 KC at #29 OAK (-3)
    #6 PIT (-3.5) at #9 DEN

    The first four correlate with predictions by the USA Today computer, which has the Bills (by 2.3) and Bucs (by 1.22) winning straight up and the Falcons (-.01) and Chiefs (-.56) covering. Your longest shot appears to be Sunday night’s game, which Sagarin expects to be a blowout (-9.78).

  3. Brian Burke says:

    Derek-That's funny because I like your model's picks this week.

    I think you might be slicing up the data set too finely by estimating HFA for intradvision/intraconference games. There just aren't that many of them, at least so far. I'm not sure if there is a good reason to believe HFA is stronger intradivision or not.

  4. Derek says:

    Splitting up the model that way didn't help accuracy at all when I tried it last spring, so I don't do it. But that's what makes it strange.

Leave a Reply

Note: Only a member of this blog may post a comment.