Roulette

Discussion in 'Physics & Math' started by Lakon, Sep 3, 2013.

  1. Dinosaur Rational Skeptic Valued Senior Member

    Messages:
    4,885
    Tach You need to read a good book or take a course in probability/statistics.

    BTW: The following is nonsense:
    Who collected when you lost & who collected when you won?

    In a previous post you claimed to have won 30 consecutive roulette bets. For a European wheel the probability of this winning series is (18/37)[sup]30[/sup] which comes out to circa once in circa 2.44 billion attempts. The house edge for an American wheel is worse. This makes me suspect your credibility.

    Roulette is a independent trials game. Unlike blackjack, what happened in the past has no effect on future turns of the wheel.

    There are books describing haunted houses & alien abductions. I do not believe the claims made in those books & similarly do not believe the claims made in the Jarecki book.

    Having posted the above, it does not seem necessary to further discuss this issue. Since I have some time on my hands, I will continue.

    The odds, probability, house edge (however you want to express it) is the same on every spin of the wheel.

    Do you understand the concept of independent trials games? Do understand that roulette is an independent trials game?

    In case you think otherwise, you can ignore the rest of this post. You would neither understand it nor believe it without taking a course or reading a book relating to probability/statistics. For the even money payoff bets (odd/even, high/low, & red/black), the European wheel probabilities are as follows for 100 turns of the wheel:

    0.355 345: Win
    0.076 734: Break even
    0.567 916: Lose

    I think that the above implies that if 1000 players each play for 100 turns of the wheel, the following is expected:

    355 are winners
    77 Break even
    568 are losers​

    The statistics for an American wheel are worse: European win probability is 18/37; American wheel probability is 18/38.

    While some might be lucky for a small number of plays, everyone loses in the long run.

    If you notice that somebody is a loser & start betting opposite to his choices, the long term results are that both of you will be losers. You might lose more than the other player or less. In the long run, you will both be losers.

    Since roulette is an independent trails game, there is no betting strategy/algorithm which changes the above statistics. The only hope for beating the house is finding a wheel with a bias which can be discovered & exploited as claimed in the Jarecki book.

    Note that a slight bias would not overcome the above described house edges.

    If you study the layout of a wheel, it does not seem to present any possibility of an exploitable bias. Furthermore, wheel manufacturers/casinos do their best to avoid making/using biased wheels.

    It would take a large bias to beat the house edge (circa 5% for American wheel & about half that for the European wheel).

    Again I mention that shops in Vegas & Reno sell forms for use in recording roulette results in order to search for an exploitable bias. Players use those forms in full view of pit bosses & other casino personnel, who do not expect the players to find an exploitable bias. I trust the judgment of casino personnel.
     
  2. Google AdSense Guest Advertisement



    to hide all adverts.
  3. Fednis48 Registered Senior Member

    Messages:
    725
    Good question! You would be right except for one thing: the zero only pays off at 35:1, not 36:1. So the expected return for a zero bet is 35/37, which is 1/37 less than a red or black bet.
     
  4. Google AdSense Guest Advertisement



    to hide all adverts.
  5. Dinosaur Rational Skeptic Valued Senior Member

    Messages:
    4,885
    Fednis48: I am not sure what you mean by the following.
    In a game with a house edge, player expectation should be negative.

    I think the probability theory expectations of gain for a unit bet on European wheel are as follows.

    For a bet on zero or any other number: 35*(1/37) - (36/37) = -0.027

    For the even money bets (red/black, odd/even, high/low): (18/37) - (19/37) = -0.027 ​
     
  6. Google AdSense Guest Advertisement



    to hide all adverts.
  7. Lakon Valued Senior Member

    Messages:
    1,117
    (Staying with a single zero scenario) .. are you sure about this ? I thought that as single bet on zero returns exactly the same as a single bet on any other number. Are you really sure ?
     
  8. Lakon Valued Senior Member

    Messages:
    1,117
    PS; see the wiki page, here ..

    http://en.wikipedia.org/wiki/Roulette

    Look down the 'Expected value of $1 bet, French' column. It seems to be as I have been saying - and as I see, Dinosaur confims, above.
     
  9. Fednis48 Registered Senior Member

    Messages:
    725
    Edit: I just realized I wasn't wrong for quite as dumb a reason as I thought I was!

    Please Register or Log in to view the hidden image!



    When I read that a zero bet pays back 35:1, I thought that meant that if you put a chip on zero and won with it, you would get 35 chips back including the one you bet with. If this were true, each chip placed on zero would give you on average 35*1/37=35/37 chips back. A color/parity bet gives you back 2 chips for each one on the table, so each chip placed on a color/parity would give you on average 2*18/37=36/37 chips back, which is 1/37 more. Based on what you wrote above, though, it sounds like winning with zero actually gets you 35 new chips plus your original chip. This extra chip makes up the difference and puts the two types of bets at equal expected return.
     
  10. Lakon Valued Senior Member

    Messages:
    1,117
    Bet a chip on zero and you get back a total of 36 inclusive of your initial chip.

    Same with any other number. Absolutely NO difference.

    36 back for 37 possibilities - thus the house edge. It's that simple.
     
  11. Lakon Valued Senior Member

    Messages:
    1,117
    Ergo, a drunk spilling his chips on the table and letting 'em ride has the same chance of winning / losing as anyone.

    As I said earlier, a snapshot of a roulette table, with many and varied bets on it from many and varied people, IS basically identical to a drunk spilling his chips on the table.

    Moral of the story ?

    Drink !
     
  12. Dinosaur Rational Skeptic Valued Senior Member

    Messages:
    4,885
    When analyzing expectations & payoff probabilities, it is easier to view a game like a person to person game rather than a casino to player game.

    In person to person, I keep my money & the other person keeps his/hers. Later the loser pays the winner, who still the amount he/she bet.

    In a casino to player game, the casino holds all the money & pays a winner his/her profit plus the amount bet.

    My calculation of expectation a few posts back was based on person to person betting considering the winner's gain, not the gain plus the amount bet.

    It is easy to make errors due to not using person to person betting analysis.
     
  13. Lakon Valued Senior Member

    Messages:
    1,117
    Hi Dinosaur - I missed this one ..

    I don't see a problem with your explanation a few posts back.
     
  14. Dinosaur Rational Skeptic Valued Senior Member

    Messages:
    4,885
    Tach: You Posted the following
    I comprehended your Post. Your story was that you won by betting the opposite of choices made by some other players & won 800 Euros. I think you claimed that they were using a computer to determine their strategy.

    Assuming your story was true, it merely indicates that you were lucky & have little understanding of probability analysis. You also made the following remark
    Who were you betting against? Who took your money when you lost? Perhaps some leprechaun or the tooth fairy? The latter statement is some type of fallacious argument or an attempt to baffle with Bulls**t when you cannot dazzle anyone with brilliance.

    Your rational was based on the concept that the Casino was willing to accept their bets, thus indicating that casino personal understood their system & knew it would be unsuccessful.


    Casino personal know that an American Wheel has a 5% house edge, while a European wheel has about half that edge.

    Casino personnel rely on that house edge & would not waste time analyzing some system to decide whether or not it had merit.

    Your POV/analysis is utter nonsense!

    Consider analysis of even money roulette bets (red/black, odd/even, & high/low).


    Assuming that neither zero nor double zero ever occur, one of you figures to win a bit: It could be you; It could be the other player. If it happened to be you, it was due to luck rather than a clever strategy. Note that in 100 or more spins, an exact 50/50 split is highly unlikely; One of you figures to be a winner. Which is the winner would be a matter of luck.

    If zero or double zero occurs, you both lose. Over a large number of spins, you both figure to be losers, especially with an American Wheel.​

    BTW: The Jarecki book is fiction. Believing it to be a true story reminds me of some primitive cultures who believe that any printed book must be a valid account.

    How many books have been published describing alien abductions? Do you believe any of them? They are works of fiction as is the Jarecki book

    Some Poster here at Sciforums considered the Jarecki book to be a true story & used a quote from it as a citation! I believe that less than 10% of the Posters at Sciforums are morons, but sometimes wonder about that estimate.
     

Share This Page