Roulette Wheel Game

Discussion in 'Trading' started by JB3, Feb 12, 2016.

  1. Humpy

    Humpy

    The way I see it is
    100 (total slots)- 2 (0 and 00)= 98/2 ( half are red, half black) = 49%

    A 49% chance of winning is a loser eventually.
    The 2+2 = 5 people will say I'm wrong but its just their take on it.
     
    #21     Feb 14, 2016
  2. Cswim63

    Cswim63

    People confuse risk with expectancy. If I put two bullets in a gun with a hundred chambers, would you play Russian roulette with me to make three percent? OK you want me to go first now?
     
    #22     Feb 14, 2016
    kut2k2 and IAS_LLC like this.
  3. IAS_LLC

    IAS_LLC

    I like the positive expectancy of the game, but the positive expectancy relies on repeated trials....if u get hit by the 2% tail on the first roll...than what? In no circumstance would I ever risk 100% to make 3%. I'd need some sort of hedge to bring my risk down to 10-20% for me to want to play
     
    #23     Feb 14, 2016
    kut2k2 likes this.
  4. kut2k2

    kut2k2

    Bingo. As I said before, even taking winnings out of the game requires surviving a minimum of 34 spins in a row just to get to the point where losing the original account doesn't leave you a net loss. And surviving 34 spins is just barely above 50% likely. Risking a lot for a gain of a little in basically a 50/50 bet. Not my idea of a good investment.
     
    #24     Feb 14, 2016
    IAS_LLC likes this.
  5. JB3

    JB3

    I'm saying you win whether you land on red or black.
     
    #25     Feb 14, 2016
  6. How did you come up with 15% gain after 5 spins? A very simple simulation shows that on average, after 5 spins, assuming no compounding, you'd make about 4%, not 15%.

    And after 50 spins, the average is now a loss of about 9%.

    Here is the code (Java):
    Code:
    import java.util.Random;
    
    public class Simulation {
    
        public static void main(String[] args) {
    
            Random rnd = new Random();
            int spins = 5;
            long trials = 1000000;
            double averageGain = 0;
    
            for (long trial = 1; trial <= trials; trial++) {
                double gain = 0;
                for (int spin = 1; spin <= spins; spin++) {
                    int outcome = rnd.nextInt(100) + 1;  // random number between 1 and 100, inclusive
                    if (outcome <= 2) {
                        gain = -100; // got "0" or "00" slot
                        break;
                    } else {
                        gain += 3;
                    }
                }
                averageGain += gain;
            }
    
            averageGain /= trials;
            System.out.println("Average gain or loss after " + spins + " spins : " + averageGain);
        }
    }
    
     
    Last edited: Feb 15, 2016
    #26     Feb 15, 2016
  7. JB3

    JB3

    Every successful spin nets you 3% gain to your account balance.

    Let's say you had $1000 in your starting account balance. And we are not compounding to make the math easier. So we put $1000 on the line on every spin, and you get +$30 for every spin.

    Spin 1) $1030
    Spin 2) $1060
    Spin 3) $1090
    Spin 4) $1120
    Spin 5) $1150

    $1150/$1000 = 1.15 or +15%.
     
    #27     Feb 15, 2016
    kut2k2 likes this.
  8. JB3

    JB3

    For all of those people that say they will not play this game because you can hit 0 or 00 on your first spin, and lose your whole account. That is technically true, so can a black swan financial event.

    Asked another way: At what odds would you play? What if the odds were 99% winning instead 98%, and the payout if +10% per winning spin?

    I'm not trying to change the rules now, but I'm seeing if people's minds change at what risk. At some point, you have to look at it as, I'm going to take this risk.

    It is interesting because this is trading, where you have to risk something to get something.
     
    #28     Feb 15, 2016
  9. kut2k2

    kut2k2

    The problem is in your game, the player has to risk everything to get something. That's not trading, that's gambling. The Kelly fraction for this system is considerably less than one. I never bet more than the Kelly fraction.
     
    #29     Feb 15, 2016
    IAS_LLC likes this.
  10. I don' understand. Why are you not accounting for the probability of a loss?
     
    #30     Feb 15, 2016