Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Formula for most probable horse winner

  1. Jun 18, 2013 #1

    I have a potential job opportunity where I will be given horse racing data i.e. average speed, max speed, weight of jockey etc etc
    I have been asked to create a formula that finds the most probable horse out of these statistics.

    For one.. Is it even possible to get to a high accuracy?

    and two.. what form do you think my employer wants the formula in?

    three... Any advice or ideas?

    Thank you
  2. jcsd
  3. Jun 18, 2013 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    This sounds like something you should ask your employer.
  4. Jun 18, 2013 #3


    User Avatar
    2017 Award

    Staff: Mentor

    With a formula significantly better than the existing ones*, you would not need an employer. You could just make money with bets on horses.

    *I have no idea how they look like, but as horse bets exist, those formulas have to exist in some way.

    In a mathematical form?
  5. Jun 18, 2013 #4
    I actually know a little about this subject.

    First, buy and read this book, Hold Your Horses by James Selvidge. It's by far the best layman's introduction to horse racing, betting, and getting a handle on figuring out which horse is going to win a race. It's out of print but you can find used copies around. Very highly recommended.


    Then, do some research on pari-mutual betting. That's the system of betting used at horse tracks. You are not betting against the track; you are betting against all the other bettors. it's essential to understand how pari-mutual betting works.

    Remember, it's not enough to just pick the winner. You have to get a good price too. If you just bet the favorite all the time, you'll have a lot of winners but you won't make much money. What you're trying to do is find an "overlay" -- a horse whose actual chance of winning is a lot better than the track odds. That's how you make money ... by finding horses that you know are likely to win, but that nobody else thinks are likely to win.

    Then, contact the headquarters of the Daily Racing Form and ask them how to get ahold of their historical data.


    They have data for every horse race run in the country going back decades. Condition of track, length of race, purse, weather conditions, horses, jockeys, all of it. This is the Bible of horse racing. You don't go near a race track without a copy of the racing form.

    If I were tasked (with sufficient financial backing) to figure out how to predict the outcome of horse races, I would take the collected historical data of the Daily Racing Form, put it into a computer, and data-mine the hell out of it.

    Of course ... this has been tried before. There's a phrase for a certain type of person ... "degenerate horse player." But hey, go for it. It's actually a fascinating area of study with both theoretical and practical implications.

    Another thing you need to do -- assuming you are being given serious time/money resources for this project -- is go down to your local race track and start talking to people. Not the degenerate gamblers. The owners, jockeys, grooms, and track employees. One thing you'll learn is that not every jockey is trying to win that day. Horses are often "held back" in order to get more favorable odds in a subsequent race; or to manipulate the price paid for the horse in a claiming race. [What's a claiming race? It's a race in which anyone can "claim," or buy, any of the horses in the race. There's a lot of manipulation that goes on in claiming races. You need to know this kind of stuff. It's all in Selvidge's book]

    Manipulation of race results is extremely common, for a variety of reasons. It's wrong to assume that the best horse wins a given race. it's normal for a good horse to be held back for strategic reasons. This is why a serous horse player has to get down to the track and obtain some real-world knowledge.

    And remember, there's some meta-reasoning that applies here. If anyone could mathematically predict the outcome of any type of gambling event ... it would already have been done. You have a lot of serious research ahead of you.

    Hope some of this helped.
    Last edited: Jun 18, 2013
  6. Jun 19, 2013 #5
    Thanks steveL27 you were very helpful.

    The rest of you on this forum are extremely condescending and unhelpful. If you don't know, don't answer.. simple.

    Of course it's going to be of a mathematical form
  7. Jun 19, 2013 #6
    Your question was vague and nearly unanswerable, which several other posters pointed out. They weren't being rude; if you want a helpful answer, you should provide enough information for others to formulate
    a helpful answer.
  8. Jun 19, 2013 #7
    Apologies... I'm on my man period. Those on this particular question were fine I was talking about posters on this forum in general.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook