1. #1
    jokerjoe
    jokerjoe's Avatar Become A Pro!
    Join Date: 05-29-08
    Posts: 10

    Multiway Kelly Betting - Back multiple vs Lay single

    Hi all,

    I'm backtesting a horse-racing model. There is a big difference in results between backing multiway (using Ganchrow's formula from ages ago) compared to simply laying the single biggest overlay. Intuitively it seems to be the former should be no worse as one can lay using multiple back bets, and probably should be better as one can fine-tune the expected utility better.

    However I'm finding that, with some simple filtering on odds ranges and overlays, that the opposite seems to be true and simply laying the biggest overlay is more profitable. Any thoughts on the matter?

  2. #2
    FourLengthsClear
    King of the Idiots
    FourLengthsClear's Avatar Become A Pro!
    Join Date: 12-29-10
    Posts: 3,808
    Betpoints: 508

    If I am reading your post correctly, they should be identical.

    Backing horses 1, 2 and 3 should produce the same recommended net wager size as simply laying horse number 4.

    If we take a 4 horse race where runner is available to back or lay at 4.00/+300 (not practical I know) and you figure horse 4 only has a 22% chance of winning the race then fully Kelly would mean:

    1) Backing horses 1, 2 and 3 each for 4% of BR or
    2) Laying horse 4 for 12% of BR (in terms of exposure/risk).

  3. #3
    RickySteve
    SBR is a criminal organization
    RickySteve's Avatar Become A Pro!
    Join Date: 01-31-06
    Posts: 3,415
    Betpoints: 187

    That's not what he means. Even though you feel you've been anointed king of the idiots in this forum, you really don't need to post in every single thread.

  4. #4
    MonkeyF0cker
    Update your status
    MonkeyF0cker's Avatar Become A Pro!
    Join Date: 06-12-07
    Posts: 12,144
    Betpoints: 1127

    Quote Originally Posted by jokerjoe View Post
    Hi all,

    I'm backtesting a horse-racing model. There is a big difference in results between backing multiway (using Ganchrow's formula from ages ago) compared to simply laying the single biggest overlay. Intuitively it seems to be the former should be no worse as one can lay using multiple back bets, and probably should be better as one can fine-tune the expected utility better.

    However I'm finding that, with some simple filtering on odds ranges and overlays, that the opposite seems to be true and simply laying the biggest overlay is more profitable. Any thoughts on the matter?
    Over what sized sample are you seeing these results? Are the overlays exceeding their expected winrate?

  5. #5
    FourLengthsClear
    King of the Idiots
    FourLengthsClear's Avatar Become A Pro!
    Join Date: 12-29-10
    Posts: 3,808
    Betpoints: 508

    Quote Originally Posted by RickySteve View Post
    That's not what he means. Even though you feel you've been anointed king of the idiots in this forum, you really don't need to post in every single thread.
    Ahh. So you do know what he means but rather than address that and post something constructive, you choose to slam me for misunderstanding, overvaluing myself and posting too much.

    Since we are bordering on psycho-analysis though, why are you so bitter and twisted, champ? I get that you feel 99.9% of the members of this forum are not worthy of engaging you in sensible discussion but there must be more to it than that.

Top