1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Predicting where a roulette wheel will stop

  1. Oct 23, 2012 #1


    User Avatar

    NOTE: I am not talking about normal roulette. Here in the midwest, some states have not approved traditional roulette, so some casinos got "clever" and made a version without the ball, just a wheel that stops on a pointer, similar to wheel-of-fortune.


    The wheel spins for about 1 minute before stopping, and you are allowed to place bets until the last ~20 secs. The first time I watched this I thought that it should be very easy to predict where the wheel would stop if you apply some basic physics. If you can simply predict with certainty which HALF of the wheel will stop under the pointer, the advantage gained is 89% over the house. Normally you have a 1/36 chance of hitting a number for a 1:35 payout (approx -5% edge for the player). Guessing which half it will land on will give you a 1/18 chance for a 1:35 payout (89% advantage for the player).

    I know there has to be an easy way to do this, but I've been trying hard for a couple months and have only seen sporadic success. I want to hear some other ideas thrown around. My method so far has been to time one complete rotation of the wheel (from 0 to 0 for example), and then map this to a result, either the 0 half or 00 half. For example, if the lap times are hitting 2.9, 3.1, or 3.3 the wheel will stop on the 0 half, but if the lap times are hitting 2.8,3.0, or 3.2, the wheel will stop on the 00 half. This method works great if I'm studying the same spin over and over (from a video), and I can get 100% accuracy, but in the real world, I've found the timings to skew every other spin, by around 0.1 sec, which is enough to often throw everything off for early predictions. I've been shredding my brain to pieces trying to figure out a pattern or the source of this shift, but haven't come up with anything useful.

    I'm going to avoid discussing how one might actually accomplish these timings in a casino setting, but I want to hear some ideas, given a hypothetical situation where you could sit at the table with a laptop and programming knowledge, how would you go about this?
  2. jcsd
  3. Oct 23, 2012 #2
  4. Oct 23, 2012 #3
    If I owned a casino, I'd make darn sure the wheel had an unpredictably varying friction/velocity response.

    Probably be enough to just sprinkle a little sand in the bearings. :D
  5. Oct 24, 2012 #4
    Is it possible for you to provide data on the angular rotation of the wheel as a function of time (total degrees vs time, or total revolutions (including fractions of a revolution) vs time)? We would like to understand the functionality better.
  6. Oct 24, 2012 #5


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member


    I think this featured in a TV documentary.

    I thought when casinos discovered it was possible they started preventing you betting after the ball was run.


    Last edited: Oct 24, 2012
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook