1. #1
    Spartak
    Update your status
    Spartak's Avatar Become A Pro!
    Join Date: 08-29-15
    Posts: 57
    Betpoints: 622

    [Q] Actual yield vs Expected Yield

    Hi folks,
    I am getting some weird numbers here and decided to ask you for opinion. So there is the case:

    I have set with 195 bets from the begging of 18/19 season. Flat betting at specific local bookie. All bets are in 2-5% value compared to the model which give me 3.40% Expected Yield. Not bad but not a big deal.
    However the actual results far away from that number. At the bottom I see +31.90% Actual yield with +62.2 units profit. 75 wins out from 195 bets with average odds 3.73 (+273)

    I know that sample is too small and some difference is expected, but should I be worried. Should I expect some huge downtrend?
    Last edited by Spartak; 01-29-19 at 10:52 AM.

  2. #2
    gui_m_p
    gui_m_p's Avatar Become A Pro!
    Join Date: 09-18-13
    Posts: 123
    Betpoints: 1531

    You should expect a reversion to the mean in the future, but many people don't get the concept right. It does not mean that a huge increase in your bankroll will be followed by a significant downtrend (that is a common bias).

  3. #3
    danshan11
    I am good at coin flips, I really am!
    danshan11's Avatar SBR PRO
    Join Date: 07-08-17
    Posts: 4,101
    Betpoints: 2887

    interesting post
    how do you determine an expected yield?

  4. #4
    Spartak
    Update your status
    Spartak's Avatar Become A Pro!
    Join Date: 08-29-15
    Posts: 57
    Betpoints: 622

    I do consider the average value as expected yield because the model Is odds based. I bet only if get better price compared to odds from Pini and other major bookies. Also I am deducting the margin to get the true odds.

  5. #5
    danshan11
    I am good at coin flips, I really am!
    danshan11's Avatar SBR PRO
    Join Date: 07-08-17
    Posts: 4,101
    Betpoints: 2887

    oh ok so you are just line shopping
    book A has +120 and pinnacle has +100 you bet at book A to get the juice from the line, that is how you base the value?

  6. #6
    Spartak
    Update your status
    Spartak's Avatar Become A Pro!
    Join Date: 08-29-15
    Posts: 57
    Betpoints: 622

    Pretty much

  7. #7
    Spartak
    Update your status
    Spartak's Avatar Become A Pro!
    Join Date: 08-29-15
    Posts: 57
    Betpoints: 622

    That is why I suppose that some regression should occur.

  8. #8
    danshan11
    I am good at coin flips, I really am!
    danshan11's Avatar SBR PRO
    Join Date: 07-08-17
    Posts: 4,101
    Betpoints: 2887

    makes sense, that is probably a real grind, best of luck!

  9. #9
    HeeeHAWWWW
    HeeeHAWWWW's Avatar Become A Pro!
    Join Date: 06-13-08
    Posts: 5,338
    Betpoints: 177

    Not worth looking at subsets <1000 bets, tbh. It's tempting of course, I do it too.

  10. #10
    Miz
    Miz's Avatar Become A Pro!
    Join Date: 08-30-09
    Posts: 558
    Betpoints: 794

    i'd expect future performance to perform in the 2-5% range, however that estimate may be incorrect... more data probably needed. Can you calculate the likeliihood you would see this type of extreme performance? Easy to do with point spreads... more complicated with moneylines. Any time you have big underdogs as the typical play, you have big variance. If 6 games go the other way, how does it look? I think exploring those kinds of questions is worth while.

  11. #11
    tsty
    tsty's Avatar Become A Pro!
    Join Date: 04-27-16
    Posts: 499
    Betpoints: 4291

    195 bets with average odds 3.73

  12. #12
    danshan11
    I am good at coin flips, I really am!
    danshan11's Avatar SBR PRO
    Join Date: 07-08-17
    Posts: 4,101
    Betpoints: 2887

    Quote Originally Posted by tsty View Post
    195 bets with average odds 3.73
    yeah 195 is nothing to even test with
    and 3.73 odds will be all over the place in small samples!

Top