1. #1
    FredMadison
    FredMadison's Avatar Become A Pro!
    Join Date: 06-19-10
    Posts: 37
    Betpoints: 194

    Need advice on a book for a handicapper based sport betting system

    Hello!

    I am working on a sport betting system based exclusively on a database with the historical results of handicappers as well as their wagers in real-time.
    In short, I follow the bets of the most profitable handicappers depending on their past results.
    I have compilled all the data in spreadsheets and I am running all sorts of statistics about the handicappers.

    I have first tested my system without going very deep into maths and got promissing results, so I want to go one step further and make some serious statistics calculations and simulations in order to optimize my system.

    I have bought some statistic books but I get lost in alot of useless topics that's why I am looking for an advice on the best book that would help me to solve these particular problems mathematically.

    Here are some examples of the kind of problems I am confronted to:

    - Mr X has made 5 000 wagers with an average ROI of 4%.
    Mr X gives me a tip on an event with a 1,4 odds. What is my expectation on this bet given the informations I have about Mr X.
    Most generally how to rate mathematically the bettors based on their performance and the number of wagers I have about them in my database.

    - How likely is it that Mr X. is in fact a lucky looser and how should this impact my bets exactly?

    - How much of a drop in the odds between the price the handicapper got and the price I get is still acceptable?

    - What is the optimal amount to assign to a bet in order to lower maximally my variance and optimize my profit depending on the bettor performance, the number of historical bets I have about him, the size of my bankroll, the number of bets I am able to take everyday, etc...(I guess this would have something to do with the Kelly Criterion but how to apply it optimally in my case...?)

    I am working with Excel so if you have also a book that explain how to do these tasks in excel that would also be great.

    Thanks for your help!

  2. #2
    FredMadison
    FredMadison's Avatar Become A Pro!
    Join Date: 06-19-10
    Posts: 37
    Betpoints: 194

    Given the answers I got, and after several days of research - getting lost in tons of books about risk and money management, statistics, sport betting, excel, Kelly criterion, etc, etc...- I at least understand now that my questions were too broad to call for a single answer. By answer I mean books to read to be able to solve these problems (using excel preferably).
    So I'll go one-by-one hoping to get more help:

    1st question) How to rate a Handicapper? (or any bettor, or yourself)

    Is it better to follow the advice of a 300 wagers/15% ROI handicapper or a 8000 wagers/4% ROI?
    How likely is it that a 5000 wagers/10% ROI is just the luckiest guy in the pool?
    Which test and simulations should be run?
    How to rate your own data? Which numbers should you reach to be sure that you are a profitable bettor?
    How to know if the data you own, have a predictive value?

    Thanks in advance to anyone who will give me the title of the right book to read in order to be able to solve this.

  3. #3
    Miz
    Miz's Avatar Become A Pro!
    Join Date: 08-30-09
    Posts: 693
    Betpoints: 3132

    you should look at z-score. there have been a million posts about this. just troll this forum's posts from like 2005-2008

  4. #4
    antonyp22
    antonyp22's Avatar Become A Pro!
    Join Date: 01-12-14
    Posts: 78
    Betpoints: 2528

    Closing line value is what you should be looking at

  5. #5
    bullheaddad
    Update your status
    bullheaddad's Avatar Become A Pro!
    Join Date: 03-16-14
    Posts: 206
    Betpoints: 227

    Just pick a capper that has a winning system, record, roi, or whatever u want to rate em by ( you can find a bunch here for FREE) ,stay within your bankroll,( i suggest 1-3% a wager) and stay with whomever you choose through the ups and downs for a year.staying consistant, not cherry picking plays, and not going on tilt should get u a profitable year. Put all that other crap away. Your wasting time..IMO

  6. #6
    Buffalo Nickle
    Buffalo Nickle's Avatar Become A Pro!
    Join Date: 11-12-14
    Posts: 3,228
    Betpoints: 878

    They don't write those kind of books. You can't be Warren Buffett or Bill Gates unless you've got it in you. If you have to ask the questions, you cannot find the answers. The kind of people that can do what you want to do just do it. Plus, there are no handicappers that you can deliver the kind of return you are looking for. You'd also have to pay $1,000s of dollars to buy their picks and then you'd find they can't deliver.

    So you'd be putting out massive amounts of money and time with very little prospect for a return.

    You've got to be able to create your own model to predict games or you've got to be able to predict the future based on your own analysis.

    The best thing for you to do is get yourself a shiny quarter. Pick Team A and Team B. Assign Heads to one and Tails to the other. Flip the coin and this is key: ALWAYS CALL HEADS. Bet appropriately.

    This will save you hours of work and thousands of dollars and will allow you to beat almost every gambler out there in terms of winning percentage.

    If you want to learn to build winning math models, apply with Goldman Sachs. After a few years, you might be able to take your skills and try to build yourself a sports handicapping model. Some smart guy is doing it every day.

  7. #7
    PassTheDutchie
    PassTheDutchie's Avatar Become A Pro!
    Join Date: 02-05-13
    Posts: 66
    Betpoints: 1320

    I'm currently doing something similar, but not with touts as a source.
    There are several problems with this.
    1. If you go based on majority opinion of your selected sources, you will usually not get an optimal line
    2. If you go based on win percentage vs implied odds then you could end up following players on a down swing each time and not on the up swings
    3. If you follow an actual sharp, that really moves the market, io words beating the closing line, you might end up following pump fakes. Lot of that going on in first weeks of college football this season.

    If you really want to go this way, safest thing is to follow market movers and middle after the market has moved a couple of points. Find a book that moves slowly, like nitrogensports that seems to follow pinnacle.

    but I wouldn't quit a day job for this.

  8. #8
    statnerds
    Put me in coach
    statnerds's Avatar Become A Pro!
    Join Date: 09-23-09
    Posts: 4,047
    Betpoints: 103

    Quote Originally Posted by antonyp22 View Post
    Closing line value is what you should be looking at
    what he said

    and you are making this way more complicated then it has to be. focus on one Market at a time. CAP the Market, not the independent events. know where the number is most likely headed and be there first. bet openers as much as possible, have as many outs as possible.

  9. #9
    agendaman
    agendaman's Avatar SBR PRO
    Join Date: 12-01-11
    Posts: 3,656
    Betpoints: 14044

    1 am a mathematician trust me forget it
    Last edited by agendaman; 12-04-14 at 01:07 AM.

  10. #10
    bihon
    bihon's Avatar Become A Pro!
    Join Date: 11-03-09
    Posts: 731

    1.
    Get the P-value calculator from this guy:
    http://www.football-data.co.uk/blog/...ts_betting.php

    Modify it by adding a sort of Z-score to it using his Standard Deviation value:
    Z=(AWB-EWB)/SD/10
    where
    AWB=number of Actual (observed) Winning Bets
    EWB=number of Expected Winning Bets (TotalNumberofBets/AverageOdds)
    SD= Standard Deviation

    Z over 2.0 and p-value under 0.01 is good.

    2.a
    Forget Kelly, it is a hype. Way too agressive, no wonder there are 1/2, 1/3 and so on variations of it.

    2.b
    The optimal bet depends on the edge, Risk of Ruin and Drawdawn ratios.
    No math revealed here, do your own research (it's pretty simple), however with an edge of 0.05 and 2.0 odds, the bet should be around 1%-1.5% of the bankroll. Expected DD in this case will be no more than 35%, with very good growth.
    Last edited by bihon; 12-04-14 at 07:51 AM.

  11. #11

  12. #12
    southmaniac6
    southmaniac6's Avatar Become A Pro!
    Join Date: 12-05-14
    Posts: 109
    Betpoints: 121

    Agree with agendaman

Top