2010 NFL Point Spread Picks Week 9

Pick 1: Chicago -3 PUSH
Pick 2: Seattle +7 incorrect NFL point spread pick

This week only two picks stood out to my struggling statistical models. Chicago is playing an 0-7 team on the road. Chicago's D is third in the league and given their expertise at taking the ball away, they could easily blowout this game without the help of its struggling offense (should they go for Moss? can they?). Seattle has a weak passing D, which could hurt. They're playing home and supposedly its one of the loudest places to play. One or two picks and they could keep this game close, and maybe take the win and keep their first place spot. Did anyone say homedog?

Again this week, we will use 20% of the bankroll for a total usage of $342.5. This places $171/game which we'll round to $175 per game. Lets take this money home this week, we need to get out of the red and jump back into green.

Other possibly good spread picks this week are Oakland and Tampa Bay. Once again, here are the NFLpickles free point spread estimates for Week 9 of 2010.











































































































GameVegas LineEstimatePrediction-VegasConfidence
CHICAGO @ BUFFALO37466.4%
NY GIANTS @ SEATTLE7-11.0-1861.5%
TAMPA BAY @ ATLANTA-8.5-2.16.457.8%
KANSAS CITY @ OAKLAND-2.5-5.0-2.557.0%
PITTSBURGH @ CINCINNATI521.416.453.5%
DALLAS @ GREEN BAY-8-9.5-1.551.1%
INDIANAPOLIS @ PHILADELPHIA-38.011.050.5%
NEW ENGLAND @ CLEVELAND4.54.90.450.0%
NEW ORLEANS @ CAROLINA6.50.5-6.050.0%
NY JETS @ DETROIT43.0-1.046.7%
MIAMI @ BALTIMORE-5.53.28.749.2%
ARIZONA @ MINNESOTA-9-3.25.848.8%
SAN DIEGO @ HOUSTON36347.9%


How to read the table:
  • Vegas Line: A NEGATIVE number implies the spread favors the HOME team
  • Estimate: NFL pickles spread estimate
  • Pred-Vegas: Subtraction of the previous two. POSITIVE implies VISITING team will cover spread.
  • Confidence: The probability that the spread pick is on the correct side.

Comments

Jaime said…
First I saw the Seattle spread at 6.5 and by the time i wrote the post, got it at 7. What are your picks this week? I'm interested as always to see what games came out of your calculations and how your picks fare to mine.
I bounced back last week with 2 covers, raising me to 5-2-1 on the season. It's been a good start, but with so few picks, it doesn't mean much yet.

So far, my model likes the Ravens -5. Waiting for some injury status clarity before making any others.

Jaime, are you raising your confidence threshold? Or are you skeptical of the Oakland and Tampa picks for other reasons? I seem to recall you picking games (selectively) with 57% confidence previously.
Jaime said…
Yes, I'm raising the bar this week.
Anonymous said…
Interestingly, those are thw two picks which I really wanted to grab, but barely missed via the model.

I'm busily drinking in SE Asia, but I'll post my picks tomorrow.
Unknown said…
Interesting Jaime, I do favor your picks just not my strongest plays. Last week my plays 2-2. My sheet was 5-8 ATS overall. Highlighted plays, favs 2-1,dogs 1-2, o/u 0-0.

My picks for the week.
KC +3
Colts +2.5
KC-OAK Over 40.5
Other highlighted plays.
Bears -2.5
SEA +6.5
Benjamin Mumma said…
Hey guys, I'm new on here but I just started with the whole NFL game picking as of 2 weeks ago. Last two weeks I was 6-0 and 3-1 respectively, but I am still skeptical of how well it will do in the long run. Anyway, my model has 7 picks this week:
Bears win (-155)
Packers win (-400)
Falcons win (-425)
Vikings win (-415)
San Diego - Houston under 50.5
Panthers +6.5
Seahawks +6.5

Just curious, but why don't you guys mark games to bet that are straight up using the moneyline? Based on my models of the past few years, it looks as though it can be fairly profitable.

Ben
Anonymous said…
As promised, my picks for the week, including (shockingly) 3 dogs and an UNDER.

413 Bucs +9
425 Colts +3
416 Lions +4
410 Panthers/Saints UNDER 41
Mike d. said…
This season has seen some high profile, historically good teams, lost to historically bad teams. It's also seen some bad teams with easy schedules put up a lot of wins. This is where I think the model might be lagging a bit. Waiting 4 weeks before making picks makes sense so that the model can get some data, but what if those 4 weeks are a misrepresentation of what is likely to happen the rest of the way? It might take 5,6,7+ weeks before the model corrects itself. I think it's clearly a year of anomlies in the NFL. With that being said, here are my thoughts:

Seattle won't have Hasselbeck. Charlie Whitehurst (who the hell is he?) will throw his first ever NFL pass despite being a 5 year veteran. Mike Williams is banged up but will probably play. The Giants are coming off the bye week. I'm going against the model here and taking the Giants.

Tampa has wins against Cleveland, Cincinati, St. Louis, Carolina and Arizona by a total of 23 points. They lost (got blown out) to Pittsburg and New Orleans by a total of 50 points. I put Atlanta in about the same category as Pitt and NO. The spread is a little large for me to feel comfortable taking ATL, so I'll stay away from this one.
Benjamin,

There probably are money line strategies that are profitable. Typically though, if you are picking favorites (as you are this week), you are risking more to win less. And this is why claiming a record that includes money lines is not comparable to one that is simply against the spread or over/unders (vigs can vary, but typically not by much before the line moves). If you are picking favorites against the money line, you can win many more bets than you lose, yet still lose money as a result. I'm sure we'd all be interested in hearing about your model's results, but you would have to specify more about your strategy, like how much you are wagering (hypothetically, perhaps) on each game, to be able to evaluate it.
Jaime said…
Holy sh&$ Hasselbeck is out, thank you for pointing that out mike. If a QB is out, I definitely stay away from the game. Because I didn't look into that before posting my plays, it's too late (at least for me) so I'll take the hit for being lazy and not looking into the injury report I usually look at.

I agree with you Mike in many ways. Especially on the strength of schedule. Although the model does account for it indirectly, I need to improve upon that. Thanks for your post, very insightful.
Added Vikings -8 now that it looks like Favre is probable for the game.
Benjamin Mumma said…
Probable,

Thanks for the comment. I evaluated my strategies using NFL results since 2003, and use a general betting strategy of 4% of the bankroll on each game, regardless of how many are played in a week. Initially, I was just looking at the spread, but then I found that when looking at games that meet specific parameters with regard to the spread of the game and my analysis, the formula picked 79% of selected home favorites (174 out of 220) and 65 out of 88 road favorites. I converted the spread for each game into a straight moneyline (estimated with a slight downward bias to give a safe estimate) and used that to estimate profit. The (again, estimated) yearly returns using this method are slightly less profitable than my formulas against the spread, but also slightly less risky.

In any event, I do agree with your analysis re: betting more to win less. Having just started to utilize my model, placing the picks (particularly the Falcons) made me a little worried.

Also, the 9-1 record so far is split into: 3-1 ATS, 2-0 O/U, and 4-0 Straight. Anyway, I'm going to stick with it for this year - I'll start posting my picks on here as well.

Ben
Anonymous said…
I don't know how you can make money betting 79% probable winners at -400, but maybe that was some other kind of average.

Anyway, best of luck, we'll be watching!
Ben,

I would recommend logging your picks at a site like covers.com or provenplays.com if you want to build a public track record, particularly if you'll be picking money lines.