Weekly game probabilities are available now at the nytimes.com Fifth Down. This week I also lead-in with a discussion of the kinds of things I'm looking at this time of the year.
Game Probabilities - Week 4
published on 9/30/2010
Subscribe to:
Post Comments (Atom)
Brian, have you thought about a weighted average of last season stats and this season since the season is still young?
I've been waiting for these all season!
Brian, you did a great job on Sirius XM NFL radio yesterday. Peter King is totally smitten...
Thanks, Sam. Doing radio is fun. I've finally lost my nervous "shaky voice."
Wizard-I have tested using a weighted average with last season, and there are just as many teams that drastically change in strength from yr-to-yr as there are teams that stay in last year's form. Bottom line is that it's somewhere between slightly negative information and a wash.
There are always exceptions--There are always a handful of teams you can count on to stay good or stay bad from yr-to-yr.
Brian, has your HFA variable changed at all this year?
Brian, can you give any stats on how your model did and how it compared to Vegas straight up and against the spread?
Do you know why the Giants are favored over Chicago? Week to Week momentum in the NFL is huge. When a team wins at home as an underdog, or against a division rival in a close game (The Bears accomplished both), this team loses vs the spread 62% of the time the following week when playing on the road. That statistic covers all games from 1993-2010. Vegas has finally caught on, and they are adjusting accordingly.
Last year some guy used your model to give betting advices : I can't remember his website, which was not all that great looking but incredibly interesting. Any chance you might remind me/us which it was ?
It looks like you didn't include opponent adjustments. Was that intentional?
Quote directly from the article "The model also adjusts for each team’s previous opponent strength and factors in home-field advantage."
"Do you know why the Giants are favored over Chicago? Week to Week momentum in the NFL is huge. When a team wins at home as an underdog, or against a division rival in a close game (The Bears accomplished both), this team loses vs the spread 62% of the time the following week when playing on the road."
I love these posts because they also bring out the moronic gamblers who can't be bothered with all the interesting analysis that you do.
Ok then how does Atlanta have a 64% chance of beating sf when its gwp is .08 lower? Home field advantage can't be that big
Anon makes a good point. HFA is a factor, but not all of it. I'd guess it's because while both teams have a poor DPASS, Atlanta's OPASS is significantly greater than San Fransisco's. I'd also think that the INT% heavily in Atlanta's favor contributes as well (SF throws lots of ints and Atlanta's caught a lot this season).
Anonymous, I believe the website was http://www.bestofblog.net/
Which is now dead....wondering if it exists under something else, but haven't been able to find anything.
Hey Brian, do you have stats on your picks? I'm wondering if 55% winners win close to 55% and if 80% winners win 80% etc....
If you have them, are they posted on the web somewhere?
Someone just sent them to me. I can post them sometime soon.
Did you ever get a chance to post them?