2014 NFL Betting Picks - Week 3

Pick 1: San Diego +2.5 correct NFL point spread pick
Pick 2: Carolina -3.5 incorrect NFL point spread pick
Pick 3: Philadelphia -6.5 incorrect NFL point spread pick
Pick 4: Indianapolis -6.5 correct NFL point spread pick

Not a bad start! Had I played with emotions rather than data, I would have picked my Bears and gone 4-0. Ahh, in hindsight everything seems so easy. This week we have another set of exciting data-driven NFL point spread predictions. I'm liking how the models continue to hate on Jacksonville and give them no chance even at home. Carolina is playing great D and with Newton back I'm surprised they are seen almost as even against Pittsburgh. Philadelphia started slow on MNF but their offense is showing that many players can pick up the team when needed, it was Sproles' turn yesterday. San Diego had a huge upset last week and I see them giving the Bills a hard time going to 3-0.

We continue this week with 12% of bankroll split evenly across games giving us a bet of $315 per game. Here is the full list of NFL betting spread predictions:

GameVegas LineEstimatePrediction-VegasConfidence
SAN DIEGO @ BUFFALO -2.5 4.3 6.8 62.9%
PITTSBURGH @ CAROLINA -3.5 -14.3 -10.8 60.7%
WASHINGTON @ PHILADELPHIA -6.5 -13.1 -6.6 59.1%
INDIANAPOLIS @ JACKSONVILLE 6.5 9.6 3.1 57.0%
DALLAS @ ST LOUIS 2.5 -2.2 -4.7 56.6%
TAMPA BAY @ ATLANTA -6.5 -5.2 1.3 55.8%
BALTIMORE @ CLEVELAND 2 3.1 1.1 55.1%
DENVER @ SEATTLE -4.5 -2.8 1.7 55.6%
GREEN BAY @ DETROIT -1 0.4 1.4 52.9%
KANSAS CITY @ MIAMI -5 3.8 8.8 50.0%
OAKLAND @ NEW ENGLAND -14.5 -15.0 -0.5 50.0%
SAN FRANCISCO @ ARIZONA 3 4.2 1.2 50.0%
TENNESSEE @ CINCINNATI -7 -8.2 -1.2 49.9%
CHICAGO @ NY JETS -2.5 -6.1 -3.6 49.7%
MINNESOTA @ NEW ORLEANS -9.5 -12.5 -3.0 48.9%
HOUSTON @ NY GIANTS 2.5 -3.0 -5.5 48.3%

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

Comments

Alan Hibino said…
Jamie, a couple of questions, what site do you use to get your lines? I've been looking for that SD +2.5 line.

Also, does your algorithm account for teams playing on a short week? (Noticing that your picks from Week 2 of teams playing on a short week went 3-1, wins for Arizona (over Giants on short week), Carolina (picking against Detroit), and San Diego...with the loss being Pittsburgh in the Thursday night game)

I see that your picks for this week include both Philadelphia and Indianapolis...previously I have tended to fade teams playing on a short week...wanted to see your thoughts? thanks, keep up the great work!
Alan Hibino said…
Follow up from my previous post..."picks" being what your predictions would be regardless of confidence level...
Jaime said…
I get lines from footballlocks, sportsbook, and bovada.

No, the algorithm does not account for short weeks. I would not conclude anything regarding short weeks based on on week of data.

Picks should be based on confidence level. Games outside of high confidence level perform weak over time.
Unknown said…
I am also curious as to adjustments prior to sundays games in regards to injuries.. Do you adjust your picks in between those right now and lets say sunday noon? (Right off the bat im thinking about the fact that peterson got deactivated for sunday for example..)

Thanks!
Ken said…
First time to this site and I'm still trying to understand the information here. I'm a little confused on the KC-MIA games. point spread has MIA -5 yet your algorithm(?) has it MIA+3.8. That's a pretty big swing yet it only has a 50% confidence level.
Unknown said…
I won 5 bucks on week 2! Nothing this week. I convert your analysis into confidence picks from a Pool I belong to.
Unknown said…
Stupid Panthers! Cost me my parlay!