Win probabilities for week 8 NFL games are listed below. The probabilities are based on an efficiency win model explained here and here with some modifications. 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.Pwin GAME Pwin 0.47 
BUF at 
MIA0.53 0.43 
OAK at 
BAL0.57 0.31 
ARI at 
CAR0.69 0.46 
TB at 
DAL0.54 0.59 
SD at 
NO*0.41 0.28 
ATL at 
PHI0.72 0.17 
KC at 
NYJ0.83 0.44 
STL at 
NE0.56 0.91 
WAS at 
DET0.09 0.31 
CIN at 
HOU0.69 0.30 
CLE at 
JAX0.70 0.39 
NYG at 
PIT0.61 0.35 
SEA at 
SF0.65 0.33 
IND at 
TEN0.67
*See the comment below by Jeremiah regarding the London game. The probability above has been updated to reflect the neutral site location of the SD-NO game.
Week 8 Game Probabilities
published on 10/23/2008
Subscribe to:
Post Comments (Atom)
I find it interesting that you and Vegas agree on the Cowboys' prospects (The Greek's true prob. for them is about 44%). Vegas accounts for the fact that Brad Johnson is playing instead of Tony Romo; you - I presume - don't, save that Johnson played in place of Romo last week. Three explanations that I can think of:
1) Vegas (or the betting population, whoever) overvalued Dallas compared to your metrics in the past, and now, with a Brad Johnson discount, they've caught up to you
2) Romo isn't that good, so they're expected to play as well this week with Johnson as they have in the past with Romo (mostly), which wasn't THAT good (0.59 GWP)
3) The QB doesn't matter that much, regardless of the gap between Johnson and Romo
Thoughts? I assumed that a significant change in personnel is one factor that could undermine the predictive ability of your model.
Since the Chargers-Saints game is being played in London, shouldn't the Chargers have the higher probability of winning the game on a neutral field (based on the team rating tables)?
Jeremiah-Good catch. Thanks. I had forgotten all about the London game. You're right, the Chargers would now likely have the small edge. I'll update the numbers this evening. But it's basically a toss up either way.
DP-Good point. My feel is that the QB does matter. I think explanation #1 is very true. I think the QB matters a lot, and Johnson is a significant downgrade from Romo (but the Cowboys have other problems too). Dallas was generally overvalued, and still are to some degree. I suppose I could manually de-tune the off pass eff stat for them by half a YPA or so to account for the QB change, but then I'd be just guessing.
Yeah, I was leaning on my first explanation too. Romo might be a bit overrated, but he's definitely an upgrade over Johnson, and the QB matters a lot, if not quite as much as is often supposed in the media. A significant difference between your ratings and the public's (even the betting public's) perception early in the year has got to be last year's performance. So I suspect that Dallas was getting too much credit for staying healthy last year and not enough for having a poor secondary this year.
Wow Brian - I owe a rise from 22nd to 15th in my pick'em pool to you. I used your probabilities here in a straight up pick'em pool this week (with the exception of the sd-no game) and ended up going 10-3 with tennessee playing tonight.
Muchas gracias!
I count a 36-33 record ATS for this season, around 52.2%, due to an inordinate amount of favorites not covering the spread (something around 75% of winning favorites have covered, down from the typical 82.7%).