1. 05-02-09 02:46 PM #1
    Wrecktangle
    Wrecktangle's Avatar Become A Pro!
    Join Date: 03-01-09
    Posts: 1,524
    Betpoints: 3209

    Wreck's Commandments for the new sports modeler

    I’m sure most of you think I’m pretty twisted; i.e. a modeling fanatic who rarely ventures out in the light of day....

    …this thread will convince you…

    If you are a newbie to sports modeling, these are Wreck’s Commandments to live by when you build a model:

    (imagine James Earl Jones reading this to you)

    1. Thou shalt (TS): Pick one sport to model. Actually one variant of a sport like NBA sides. Pick a sport you like and want to research because you will be spending a lot of time on it if you are to do even moderately well. Until you can do one sport variant well, you should not do others, as you’ll simply spread yourself out too thinly and accomplish little.

    2. (TS): Examine the data available to pick which type of modeling to use. In the NFL, with Gamebooks, you can build Monte Carlo Simulations (Sims) to do both Expected Value (EV) modeling and simulations. NBA is too difficult to model play-by-play so most mortals do EV, as the data to do play-by-play doesn’t exist online.

    3. (TS): Divide your data set into two sections, one to build your model from, the second to test against. This helps in preventing overfitting, but is no guarantee. I have a woeful story about this in regards to the NFL and the old USFL.

    4. (TS): Spend time and money, if you have to, getting quality data. Most of the data in on-line dbs are full of errors. Sharing data with other modelers works well, but most modelers are like herding cats, and won’t trade their precious data. Not only that, but a great portion of them think they are God’s gift to the modeling world so why should they help anyone else out? …come to think of it, my sh*t doesn’t stink either…

    5. (TS): Paper trade your system until you get it working, and use a very small bankroll (BR) in your first season when you fist make wagers. If you exhaust your BR, don’t refill it, work on the model to prepare for next season.

    6. (TS): Keep strict records of how your model is doing against the line (ATS) and straight up (SU). Annotate any changes in modeling against your win% both ATS and SU.

    7. (TS): Not use Kelly Criterion to bet your first year. You might want to seriously consider NOT using Kelly in other years too. But if you do, plan on spending some quality time on Bayesian Priors. BTW if you use it, plan on Kelly kicking your ass…a lot…

    8. (TS): Plan on spending a lot of time rebuilding your model during the season, especially when your model breaks down, i.e. can’t hold a winning percentage over at least 30 games.

    9. (TS): Plan on your model breaking down, because it will. When it does, stop everything (yes folks, this does mean: no betting) and examine everything. Look at your data set first because: garbage in with get you sewage out.

    10. (BTW): This list doesn’t cover good betting practices (don’t chase, don’t bet more than you can lose, yadda, yadda). Plenty of advice on this elsewhere.

    When you conquer a single sport variant, PM me with a copy of your code, and go back to step 1 for the next variant.

    Have fun boys and girls…

    …and girls, PM me anyway, cause any chick who can code has got to be another alien like me…
    Points Awarded:

    pdx107 gave Wrecktangle 1 SBR Point(s) for this post.

    Melloweitsj gave Wrecktangle 1 SBR Point(s) for this post.
    Reply With Quote Reply With Quote
  2. 05-02-09 03:42 PM #2
    TomG
    TomG's Avatar Become A Pro!
    Join Date: 10-29-07
    Posts: 500
    Betpoints: 1

    Watch the opening (overnight lines) and note which sides and at what price your model would have as a play. Then watch the line. Every few hours, check and see how the line moves compared to your prediction. From this information it should be pretty clear soon enough how well your model will perform.
    Reply With Quote Reply With Quote
  3. 05-02-09 05:59 PM #3
    Wrecktangle
    Wrecktangle's Avatar Become A Pro!
    Join Date: 03-01-09
    Posts: 1,524
    Betpoints: 3209

    Tom,

    Thanks, but that's advanced stuff. This thread is for Newbies.
    Reply With Quote Reply With Quote
  4. 05-02-09 07:26 PM #4
    smitch124
    smitch124's Avatar SBR PRO
    Join Date: 05-19-08
    Posts: 12,564
    Betpoints: 1547

    I only trust Trapezoid's that model.
    Reply With Quote Reply With Quote
  5. 05-05-09 09:41 PM #5
    marcoforte
    marcoforte's Avatar Become A Pro!
    Join Date: 08-10-08
    Posts: 140
    Betpoints: 396

    Amen brother. As a 20+ year veteran, my mistakes have proven the value of all of them.
    Reply With Quote Reply With Quote
  6. 05-17-09 12:39 PM #6
    Flying Dutchman
    Floggings continue until morale improves
    Flying Dutchman's Avatar Become A Pro!
    Join Date: 05-17-09
    Posts: 2,467
    Betpoints: 759

    So, If I'm just starting out, which sport do you think is best to win against?

    I can watch any that gets me $$$.
    Reply With Quote Reply With Quote
  7. 05-17-09 03:45 PM #7
    Wrecktangle
    Wrecktangle's Avatar Become A Pro!
    Join Date: 03-01-09
    Posts: 1,524
    Betpoints: 3209

    I'm biased as I like the NFL, and since I have so much time and dbs built against it, I see it as the easiest.

    But starting up, I'd probably concentrate on one of the college sports: football or basketball because I think if you build a Predictor type of model (similar to Jeff Sagarin's of USA Today) and concentrate on one or two leagues, I think you can do well starting off. I've posted on how you do that elsewhere here at SBR, but if you can't find it, PM me and I'll send the info.
    Reply With Quote Reply With Quote
  8. 05-17-09 05:29 PM #8
    durito
    escarabajo negro
    durito's Avatar Become A Pro!
    Join Date: 07-03-06
    Posts: 13,173
    Betpoints: 438

    NFL is the hardest.
    Reply With Quote Reply With Quote
  9. 05-17-09 06:08 PM #9
    Wrecktangle
    Wrecktangle's Avatar Become A Pro!
    Join Date: 03-01-09
    Posts: 1,524
    Betpoints: 3209

    Quote Originally Posted by durito View Post
    NFL is the hardest.
    I certainly hope so...
    Reply With Quote Reply With Quote
  10. 08-11-09 08:47 PM #10
    Flying Dutchman
    Floggings continue until morale improves
    Flying Dutchman's Avatar Become A Pro!
    Join Date: 05-17-09
    Posts: 2,467
    Betpoints: 759

    bump.

    Give this guy some points for this post.
    Reply With Quote Reply With Quote
  11. 08-11-09 08:50 PM #11
    pdx107
    pdx107's Avatar Become A Pro!
    Join Date: 06-20-09
    Posts: 923
    Betpoints: 1816

    I concur, he knows his shit
    Reply With Quote Reply With Quote
  12. 08-12-09 03:44 AM #12
    Leverage
    HURR DERP DEEP DURP
    Leverage's Avatar Become A Pro!
    Join Date: 07-30-09
    Posts: 253
    Betpoints: 0

    I find baseball to be the easiest.
    Reply With Quote Reply With Quote
  13. 08-12-09 03:49 AM #13
    head_strong
    head_strong's Avatar Become A Pro!
    Join Date: 07-02-08
    Posts: 4,318
    Betpoints: 500

    Quote Originally Posted by Leverage View Post
    I find baseball to be the easiest.
    Agree, baseball works best for me as well.....
    Reply With Quote Reply With Quote
  14. 12-06-09 11:42 AM #14
    Flying Dutchman
    Floggings continue until morale improves
    Flying Dutchman's Avatar Become A Pro!
    Join Date: 05-17-09
    Posts: 2,467
    Betpoints: 759

    bump again for the newbies...
    Reply With Quote Reply With Quote
  15. 12-06-09 12:05 PM #15
    Dokholiday
    Dokholiday's Avatar Become A Pro!
    Join Date: 05-04-09
    Posts: 111
    Betpoints: 50

    good info
    thanks
    Reply With Quote Reply With Quote
  16. 12-06-09 12:22 PM #16
    RickySteve
    SBR is a criminal organization
    RickySteve's Avatar Become A Pro!
    Join Date: 01-31-06
    Posts: 3,415
    Betpoints: 187

    o/u on winning sports bettors in this thread: 1½
    Reply With Quote Reply With Quote
  17. 12-06-09 12:23 PM #17
    durito
    escarabajo negro
    durito's Avatar Become A Pro!
    Join Date: 07-03-06
    Posts: 13,173
    Betpoints: 438

    Quote Originally Posted by RickySteve View Post
    o/u on winning sports bettors in this thread: 1½
    Give me over for the max.
    Reply With Quote Reply With Quote
  18. 12-06-09 12:33 PM #18
    RickySteve
    SBR is a criminal organization
    RickySteve's Avatar Become A Pro!
    Join Date: 01-31-06
    Posts: 3,415
    Betpoints: 187

    What's your line?
    Reply With Quote Reply With Quote
  19. 12-06-09 12:39 PM #19
    durito
    escarabajo negro
    durito's Avatar Become A Pro!
    Join Date: 07-03-06
    Posts: 13,173
    Betpoints: 438

    Well, I know 2 for sure

    Sorry, I'm being obnoxious today.
    Reply With Quote Reply With Quote
  20. 12-06-09 01:10 PM #20
    roasthawg
    roasthawg's Avatar Become A Pro!
    Join Date: 11-09-07
    Posts: 2,990
    Betpoints: 1

    I use Kelly... at least my own modified variant of it. Assume a 53% winning percentage on your bets and you'll be ok. Understand that if your model tells you you have a 60%+ chance on any one game that it is lying to you... this is how Kelly can crush you.
    Reply With Quote Reply With Quote
  21. 12-06-09 01:45 PM #21
    20Four7
    Timmy T = Failure
    20Four7's Avatar Become A Pro!
    Join Date: 04-08-07
    Posts: 6,703
    Betpoints: 4120

    Quote Originally Posted by roasthawg View Post
    I use Kelly... at least my own modified variant of it. Assume a 53% winning percentage on your bets and you'll be ok. Understand that if your model tells you you have a 60%+ chance on any one game that it is lying to you... this is how Kelly can crush you.
    your model will tend to give you over estimates (at least all of mine have tended to). I have kept track and found that my model gives over 70% I win those games at a rate of about 56%.... when it gives 60% I win those games at a rate of about 54%, when it's under 60 it's about 51%..... this gives me an indication of how to utilize kelly.
    Reply With Quote Reply With Quote
  22. 12-06-09 05:36 PM #22
    RickySteve
    SBR is a criminal organization
    RickySteve's Avatar Become A Pro!
    Join Date: 01-31-06
    Posts: 3,415
    Betpoints: 187

    Quote Originally Posted by durito View Post
    Well, I know 2 for sure

    Sorry, I'm being obnoxious today.
    You got me on a grading technicality. That's why I can't open my own shop.
    Reply With Quote Reply With Quote
  23. 12-07-09 01:19 AM #23
    Indecent
    Indecent's Avatar Become A Pro!
    Join Date: 09-08-09
    Posts: 758
    Betpoints: 1156

    Adding to rule number three....

    To prevent overfitting, it's important that you use a random sample of games from your entire database for the testing and training sets (the two data sets you keep separate). A good ratio of training to testing is 2:1, or 66% of games go into training the model and 33% of games are used to test for overfitting issues.

    I'll add another rule.. Be creative with your data representation. The only bad ideas are one's you have tried and proven to be useless.
    Reply With Quote Reply With Quote
Top
« Previous Thread | Handicapper Think Tank | Next Thread »