Re: Monte Carlo simulations
[ QUOTE ]
Excellent points, Najdorf. Although I think it depends on how confident you want to be. To say that someone IS a long term winner, based off of 500, or even 5000, or even 65k sample sets is stupid, because you can never be sure.
500 picks IS enough to determine that your ODDS are significantly better than 50% that you are a long term winner.
[/ QUOTE ]
I think 500 picks is laughable to proclaim yourself a 'long-term winner'. Give me what you think your winning % is, and I"ll run a monte carlo based on a normal curve.
I'll save you the trouble - say you are at 55% win rate, and do a monte carlo after 500 games. The first one I did you won 266 games, or +1.56% ROI with significant variance.
Next 500 you won 282, so better than average, the next was only 271, so again below average.
And 55% is a HUGE win rate. That's a 9.09% ROI at -110 odds. Every few hundred bets or so you'd double your capital. How many handicappers do you know who have done this, year-in and year-out?
Most pro handicappers do a lot more bets than that every year. In a few years or so they'd have 64x their original bankroll [before taxes.] Again, is this something that any of us have ever seen in real life? No.
Naj
|