Dr. Bob -- sharpshooter fallacy?

[Rococo]

[ QUOTE ]
I think it's virtually a given that the're some sort of data mining/snooping bias in his models.

[/ QUOTE ]

I think that you are being charitable. I suspect that it is 95% datamining. Now sometimes the data collects around the bullseye, but not always, especially when the sample size is small, and the cutoffs are arbitrary.

I particularly love it when he says something like, "there is a strong angle for Week 3 bounceback home favorites laying 3 pts or less, but be careful if the line moves to 4, because the angle for bounceback home favorites laying 4 points or less is not nearly as strong."

[SunOfBeach]

I wonder how much of this "Team X covered at home in 12 of their last 14 following road losses of 7+" is marketing lingo for his squareish clients, who may not understand some of the more technical aspects of his model. If he's using some sort of regression model, I don't see why he'd reveal what's really driving his picks. It's reasonable that this just could be marketing blahblah, to give him something to say other than "my model, the contents of which I can't disclose, says team x should be favored by 6 while the line is only 2.5..." yada yada yada.

[NajdorfDefense]

[ QUOTE ]
I wonder how much of this "Team X covered at home in 12 of their last 14 following road losses of 7+" is marketing lingo for his squareish clients, who may not understand some of the more technical aspects of his model. If he's using some sort of regression model, I don't see why he'd reveal what's really driving his picks.

[/ QUOTE ]

I agree 100%. I don't think he's datamining, I think he re-does that stats and puts them in his model. I just don't think he'll continue to run >58% forever.

One concern would be, if something in the game has changed since he started doing these picks - rules changes, scholarship changes, clock rule modifications -- and he has not adjusted his model for factors that affect the underlying stats and game.

[MyTurn2Raise]

IMO, you're on the right track Najdorf. I think scholarship changes are really making a difference. I'm used to hitting a pretty good clip myself as it seems it was easier to find a matchup disadvantage and go with it. Now, there are not the glaring differences and games are coming down to, more often than not, who wins the turnover and big play battle. It's who can create the big turnover. It's who has the guys that can break a punt return or turn a screen pass into 60 yards. Very few teams are the traditional line them up and smack the other team around (Ohio St comes to mind as one who does do it the old way). Lots of my models are based on stuff similar to Dr Bob. I look at ypp a lot. Anymore, I'm not sure it has as much statistical impact as it once did.

anyway, rambling response...but I think the more even teams have created differences in the game where it mainly comes down to play makers.

[NajdorfDefense]

I think if you/anyone just modeled TO margin, red zone %, 3rd and 4th down %, you'd be 90% of the way to a successful model. Obvs in the short-run things like punt TD returns and guys who don't throw INTs throwing 3 screw up a pick, but that's gambling.

In short, I don't think ypp is nearly as important as it used to be.

[MyTurn2Raise]

the think is I've down models like that
once ypp is thrown in, all the other things lose their statistical relevance