Forum: Classic Threads From The Past - Visit here for a great "blast from the past" and view some terrific, informative and interesting "Classic Threads" posted by thousands of savvy sportsbettors over the years here at The Rx.com.....

Thread: Win rate, Standard Deviation, Bankroll, and Kelly's Criterion

Page 1 of 3 123 LastLast
  1. #1 Win rate, Standard Deviation, Bankroll, and Kelly's Criterion 
    RX Local Fishhead's Avatar
    Join Date
    Oct 1999
    Posts
    75,207
    Feel free to discuss any and/or all of these factors and their importance they garnish to the highly successful gambler.

    STANDARD DEVIATION

    BANKROLL

    WIN RATE

    KELLY'S CRITERION

    In due time, this thread should make most everybody here a better gambler.
    Reply With Quote  
     

  2. #2  
    Rx. Senior winbet's Avatar
    Join Date
    Sep 2004
    Location
    EMERALD ISLE
    Posts
    7,744
    Understanding the value of the odds is most important, without it nothing will help you. Its incredible how many people say what a good price 1/2 is in a match but when asked if they would lay 2/1 the other team, look at you as if you were mad.
    Reply With Quote  
     

  3. #3  
    Rx Wizard Iceman's Avatar
    Join Date
    Oct 2005
    Location
    LasVegas...okay?
    Posts
    11,732
    I would like to talk about WIN RATE.

    After the college hoops season(don't want to jinx it now) is complete I will share with everyone how I have FINALLY turned my game around and have done VERY well (with a lot of help form you guys).

    My point is I am on the hottest run I have ever had and I am at 55.9% win rate at over 600 plays. I am completely convinced that you can win(which I wan't sure about 3 months ago) but I am 100% convinced that NO one is going to hit at a rate close to 60% over 500 plays. I have worked my ass off and I am have caught a few more breaks (beccasue of the hard work) and i don't see how it was possibile to up my win rate much higher. ANYONE that says that can hit at 60% over 500 plays at -110 lines, is in for a RUDE awakening.
    Reply With Quote  
     

  4. #4  
    Rx Wizard Iceman's Avatar
    Join Date
    Oct 2005
    Location
    LasVegas...okay?
    Posts
    11,732
    not sure if this fit's under bankroll but I'm not sure if people actually sat down and figured out that betting 1%(which I beleive in) of a 10k bankroll and hitting at an EXTREMLEY difficult 55% over 1,000 plays is only a 5k profiit.

    I really wonder if alot of the posters/gamblers actually have a bankroll and realize the difficulty of making alot of money at this.
    Reply With Quote  
     

  5. #5  
    RX Senior
    Join Date
    Oct 2005
    Location
    ponyland
    Posts
    1,038
    ice man,....well one day,...we probally should meet up,....ooooh boy forget hazel??!! well one day?
    Reply With Quote  
     

  6. #6  
    Oh boy! quantumleap's Avatar
    Join Date
    Mar 2004
    Location
    The Quantum Imaging Chamber
    Posts
    38,234
    Quote Originally Posted by ice man
    not sure if this fit's under bankroll but I'm not sure if people actually sat down and figured out that betting 1%(which I beleive in) of a 10k bankroll and hitting at an EXTREMLEY difficult 55% over 1,000 plays is only a 5k profiit.

    I really wonder if alot of the posters/gamblers actually have a bankroll and realize the difficulty of making alot of money at this.
    Good post ice man. I have recently been thinking about doing some modeling by tracking the net +/- after a series of bets using differing percentages. In other words, I've been thinking about using flat bets and tracking how much I am up or down given a percentage of wins.

    Which leads me to Standard Deviation. This is how much your wins or losses deviate from the average. If there is a large Standard Deviation in a system of bets you are likely to have a string of bets that don't do very well. Therefore you have to be confident that your system of bets is a valid one. This is so you can keep betting the system since your bets will likely have a series of wins that makes up for the bad streak.
    Reply With Quote  
     

  7. #7  
    Rx Wizard Iceman's Avatar
    Join Date
    Oct 2005
    Location
    LasVegas...okay?
    Posts
    11,732
    this is exactly why I beleive that you should be making a TON of plays daily/weekly. If not ,good luck with a small bankroll showing much of a profit.
    CHURN THE BANKROLL.
    Reply With Quote  
     

  8. #8  
    Banned
    Join Date
    Sep 2005
    Location
    los angeles
    Posts
    350
    Ice,
    You should check out Mason Malmuth's book called "gambling theory." Its a little out of date but the advice on bankroll size, std deviation and such would be right up your alley. The book mainly applies to poker, (he may have other books that relate to sports gambling . . check out twoplustwo.com) but the tools could be easily applied to sports wagering.
    Reply With Quote  
     

  9. #9  
    Rx Wizard Iceman's Avatar
    Join Date
    Oct 2005
    Location
    LasVegas...okay?
    Posts
    11,732
    thanks Jet,
    I love to read gambling books (have a few dozen) and I have heard of the author maybe I will check this out. Ordered Mike Lee's baseball book yesterday.
    Reply With Quote  
     

  10. #10  
    RX Local Fishhead's Avatar
    Join Date
    Oct 1999
    Posts
    75,207
    Thanks thus far!
    Reply With Quote  
     

  11. #11  
    RX Senior
    Join Date
    Oct 2005
    Location
    ponyland
    Posts
    1,038
    oooh boyy
    Reply With Quote  
     

  12. #12  
    RX Member
    Join Date
    Jan 2006
    Location
    Sportsbook
    Posts
    310
    On Fishhead recommendation, I'm here. Thanks
    Reply With Quote  
     

  13. #13  
    RX Local Fishhead's Avatar
    Join Date
    Oct 1999
    Posts
    75,207
    Quote Originally Posted by InncoLink
    On Fishhead recommendation, I'm here. Thanks
    Put 100 marbles in a can........60 black and 40 red(representing 60% winners).

    Remove a marble, take note by writing down what color it was and REPLACE, shake-up, and repeat the process.

    Do this many, many times.

    You may be amazed at how many streaks of RED you will have greater than four.............and this is picking 60% winners..........something even the best of the best professional sportsbettors cannot achieve longterm.

    Think about this experiment and stay focused in your sportsbetting..........NEVER panic, bet over your head, chase, double up, etc., etc.
    Reply With Quote  
     

  14. #14  
    RX Wizard Woody0's Avatar
    Join Date
    Jul 2002
    Location
    Richmond, British Columbia
    Posts
    6,480
    Ah, marbles in a can. A streak of 6 is significant at the 5% level.

    Oops, significance wasn't a topic.
    Reply With Quote  
     

  15. #15  
    Home of the Cincinnati Criminals. bigbet1234's Avatar
    Join Date
    Sep 2004
    Location
    You draft them, they book'em
    Posts
    17,032
    Well, I liked a girl in middle school names kelly, but that's beside the point.

    Me, I am balls to the wall. So I bet FULL Kelly.

    and for Standard Dev., well, let's just say I have trouble thinking long term, and that is exactly what SD is.

    Fish you forgot on important factor. DISAPLINE!

    Without Disapline, don't even consider the other 4 factors above.

    BB
    Reply With Quote  
     

  16. #16  
    Oh boy! quantumleap's Avatar
    Join Date
    Mar 2004
    Location
    The Quantum Imaging Chamber
    Posts
    38,234
    Quote Originally Posted by Fishhead
    Put 100 marbles in a can........60 black and 40 red(representing 60% winners).

    Remove a marble, take note by writing down what color it was and REPLACE, shake-up, and repeat the process.

    Do this many, many times.

    You may be amazed at how many streaks of RED you will have greater than four.............and this is picking 60% winners..........something even the best of the best professional sportsbettors cannot achieve longterm.

    Think about this experiment and stay focused in your sportsbetting..........NEVER panic, bet over your head, chase, double up, etc., etc.
    I was going to do this with an Excel spreadsheet and W's and L's in one column put in randomly at a certain % to see how many win streaks and how many losing streaks I would get.

    It's important to have faith in the system (or handicapper or yourself) once you get the losing streaks. Otherwise you may abandon a good betting system.

    The problem I have is that I get over-confident when I am doing well and I increase my bets even on events that I don't have good betting strategies on. I've gone from 100% of my bankroll to 200% back down to 100% 3 times this year already. I'm going to step back and take a closer look.
    Reply With Quote  
     

  17. #17  
    Oh boy! quantumleap's Avatar
    Join Date
    Mar 2004
    Location
    The Quantum Imaging Chamber
    Posts
    38,234
    Quote Originally Posted by Woody0
    Ah, marbles in a can. A streak of 6 is significant at the 5% level.

    Oops, significance wasn't a topic.
    So? Would you care to elaborate even though it may not be directly on topic?

    Thanks in advance.
    Reply With Quote  
     

  18. #18  
    Simply the best Kiss1's Avatar
    Join Date
    Sep 2001
    Location
    West Palm Beach
    Posts
    4,165
    Quote Originally Posted by Fishhead
    Feel free to discuss any and/or all of these factors and their importance they garnish to the highly successful gambler.

    STANDARD DEVIATION

    BANKROLL

    WIN RATE

    KELLY'S CRITERION

    In due time, this thread should make most everybody here a better gambler.
    1) Eventually will regress to the mean

    2) 2-8% per play, shark and molly agree

    3) VS Standard juice ATS = WP 53%

    4) Kelly % = W – [(1 – W) / R]

    Did I pass ?

    Reply With Quote  
     

  19. #19  
    RX Wizard cincy_'s Avatar
    Join Date
    Nov 2000
    Posts
    8,834
    Quote Originally Posted by quantumleap
    So? Would you care to elaborate even though it may not be directly on topic?

    Thanks in advance.
    It means that the chance of you getting 6 straight is less than 5%.
    Reply With Quote  
     

  20. #20  
    Oh boy! quantumleap's Avatar
    Join Date
    Mar 2004
    Location
    The Quantum Imaging Chamber
    Posts
    38,234
    Quote Originally Posted by cincy_
    It means that the chance of you getting 6 straight is less than 5%.
    Thanks much cincy_. I caught that part. I had meant to ask if he had any additional insight on significance that could help us out.

    I owe you one though.
    Reply With Quote  
     

  21. #21  
    RX Wizard WildBill's Avatar
    Join Date
    Sep 2004
    Posts
    8,780
    Beware of overbetting. Sure way to eventual ruin and it actually can cost you more money in the long run. Most people greatly overrate their expected win percentage, the reason half-Kelly is one of the most important concepts in sports betting. I personally think Kelly is a bad idea, but that is a long argument for another thread. If you are going to use Kelly, be sure you use half your suggested unit, if not less.
    Reply With Quote  
     

  22. #22  
    RX Wizard Woody0's Avatar
    Join Date
    Jul 2002
    Location
    Richmond, British Columbia
    Posts
    6,480
    Quote Originally Posted by quantumleap
    So? Would you care to elaborate even though it may not be directly on topic?

    Thanks in advance.
    Sure, 0.6^6 < 0.05.
    Reply With Quote  
     

  23. #23  
    Oh boy! quantumleap's Avatar
    Join Date
    Mar 2004
    Location
    The Quantum Imaging Chamber
    Posts
    38,234
    Quote Originally Posted by Woody0
    Sure, 0.6^6 < 0.05.
    Thanks woody0.

    Reply With Quote  
     

  24. #24  
    RX Wizard Woody0's Avatar
    Join Date
    Jul 2002
    Location
    Richmond, British Columbia
    Posts
    6,480
    Quote Originally Posted by quantumleap
    Thanks much cincy_. I caught that part. I had meant to ask if he had any additional insight on significance that could help us out.
    Well you can use tests of significance to determine whether you are a winning player, although I guess checking your bankroll is good enough for most.

    Lets say you want to determine whether you are winning at a 53% rate after 1000 games. You expect to win 530 and lose 470 and the square of difference between both results (observed and expected) can be used in a chi square test with one degree of freedom.
    Reply With Quote  
     

  25. #25  
    Oh boy! quantumleap's Avatar
    Join Date
    Mar 2004
    Location
    The Quantum Imaging Chamber
    Posts
    38,234
    Quote Originally Posted by Woody0
    Well you can use tests of significance to determine whether you are a winning player, although I guess checking your bankroll is good enough for most.

    Lets say you want to determine whether you are winning at a 53% rate after 1000 games. You expect to win 530 and lose 470 and the square of difference between both results (observed and expected) can be used in a chi square test with one degree of freedom.
    Woody0:

    I was thinking more short-term. What if you were picking 60% winners after 50 picks, or you were following someone with the same rate, or you were trying out a system. How would you know if it is a bad choice? Is it a bad system/capper if 6 losers are picked in a row? That's where you could test it using the numbers you mentioned.
    Reply With Quote  
     

Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •