1. 01-11-11 11:57 AM #1
    simplydusty
    simplydusty's Avatar Become A Pro!
    Join Date: 12-18-08
    Posts: 229
    Betpoints: 24

    Creating NFL model

    So I have my database now filled with around 20-25 stats for each team, each game from 2002-current. I want to create a predictive model using regression analysis by the start of next season.

    My plan is to use all 9 seasons when running my regression analysis. Then use 6 of the 9 seasons to create my model, and the other 3 to test the model.

    Does that seem like the best plan of attack? Granted more data is better, but 9 full seasons worth of data should work, right?

    Thanks for any help
    Reply With Quote Reply With Quote
  2. 01-11-11 12:26 PM #2
    Indecent
    Indecent's Avatar Become A Pro!
    Join Date: 09-08-09
    Posts: 758
    Betpoints: 1156

    Quote Originally Posted by simplydusty View Post
    So I have my database now filled with around 20-25 stats for each team, each game from 2002-current. I want to create a predictive model using regression analysis by the start of next season.

    My plan is to use all 9 seasons when running my regression analysis. Then use 6 of the 9 seasons to create my model, and the other 3 to test the model.

    Does that seem like the best plan of attack? Granted more data is better, but 9 full seasons worth of data should work, right?

    Thanks for any help
    Only run the regression analysis using the first 6 of the 9 seasons.
    Reply With Quote Reply With Quote
  3. 01-11-11 12:34 PM #3
    simplydusty
    simplydusty's Avatar Become A Pro!
    Join Date: 12-18-08
    Posts: 229
    Betpoints: 24

    So basically keep 3 season out of anything involved with creating the model?

    Do you think 3 is about right for the number of 'testing' seasons?
    Reply With Quote Reply With Quote
  4. 01-11-11 01:24 PM #4
    Justin7
    Justin7's Avatar Become A Pro!
    Join Date: 07-31-06
    Posts: 8,577
    Betpoints: 1506

    6 seasons is a good base. Adding 3 might give you a little more power, but not a lot.

    It's not a huge datamining issue to use the last 3 seasons for the regression though... as long as none of your regressions look at a game spread, total, or any other market you want to wager.
    Reply With Quote Reply With Quote
  5. 01-11-11 01:29 PM #5
    MadTiger
    Wait 'til next year!
    MadTiger's Avatar Become A Pro!
    Join Date: 04-19-09
    Posts: 2,724
    Betpoints: 47

    Quote Originally Posted by simplydusty View Post
    So basically keep 3 season out of anything involved with creating the model?
    Yes. I was trying to look for a link that talks about the statistical analysis principle behind this warning.

    But yeah, you don't want to use data that helped create the model as a test data sets.
    Reply With Quote Reply With Quote
  6. 01-11-11 02:18 PM #6
    Peregrine Stoop
    Peregrine Stoop's Avatar Become A Pro!
    Join Date: 10-23-09
    Posts: 869
    Betpoints: 779

    lol @ 25 stats. You're going to find a lot of auto correlation in your model.
    Reply With Quote Reply With Quote
  7. 01-11-11 02:51 PM #7
    greg66
    greg66's Avatar Become A Pro!
    Join Date: 09-04-06
    Posts: 151
    Betpoints: 0

    Quote Originally Posted by Peregrine Stoop View Post
    lol @ 25 stats. You're going to find a lot of auto correlation in your model.

    Hetereoscedasticity as well.
    Reply With Quote Reply With Quote
  8. 01-11-11 02:51 PM #8
    rfr3sh
    Blind Knucklehead
    rfr3sh's Avatar Become A Pro!
    Join Date: 11-07-09
    Posts: 10,229
    Betpoints: 604

    I would say start with something besides NFL
    since the limits are like 70k on pinny
    Reply With Quote Reply With Quote
Top
« Previous Thread | Handicapper Think Tank | Next Thread »