Here's a gem by jimbob: "Excuse me but I'm not going to play 1 billion hands in my life time so I could care less about a sim like this."
This is hilarious. Despite claiming to be a "statistician by trade," you have absolutely no understanding of statistics, or the uses and meaning of simulation. Simulation is a COMPUTATIONAL TOOL to calculate expectation (the "mean"), among other things. The fact that the sim required 1 billion hands has nothing to do with its applicability to human lifetimes. Once we have computed the expectation (by any means we could have chosen, including nonsimulation methods), that expectation applies to EVEN A SINGLE TRIAL OF A GAME. Get it?
I'm racking my brain trying to find an explanation of this concept that you can understand, but it's probably futile. I hope, at least, that everyone ELSE here understands the point I'm making. I will try it again.
Suppose that we have a modified baccarat game because a casino changes the hitting rules and tie rules. We suspect that betting Player may now have an edge. We use a billion-hand computer simulation to estimate that the expectation on Player is +10.7%, while Banker is -12.0%. Now, armed with this information, you go into the casino with the intention of playing one hand of baccarat. Are you going to reason that since you do not intend to play a billion hands, that your choice of Player or Banker is irrelevant. Or that the estimated Banker expectation of -12.0% doesn't apply to you? Even if you are going to play one hand of this game, isn't Player the smart bet? And if you are going to bet $100 on that one hand, isn't your expected profit $10.70?
This is hilarious. Despite claiming to be a "statistician by trade," you have absolutely no understanding of statistics, or the uses and meaning of simulation. Simulation is a COMPUTATIONAL TOOL to calculate expectation (the "mean"), among other things. The fact that the sim required 1 billion hands has nothing to do with its applicability to human lifetimes. Once we have computed the expectation (by any means we could have chosen, including nonsimulation methods), that expectation applies to EVEN A SINGLE TRIAL OF A GAME. Get it?
I'm racking my brain trying to find an explanation of this concept that you can understand, but it's probably futile. I hope, at least, that everyone ELSE here understands the point I'm making. I will try it again.
Suppose that we have a modified baccarat game because a casino changes the hitting rules and tie rules. We suspect that betting Player may now have an edge. We use a billion-hand computer simulation to estimate that the expectation on Player is +10.7%, while Banker is -12.0%. Now, armed with this information, you go into the casino with the intention of playing one hand of baccarat. Are you going to reason that since you do not intend to play a billion hands, that your choice of Player or Banker is irrelevant. Or that the estimated Banker expectation of -12.0% doesn't apply to you? Even if you are going to play one hand of this game, isn't Player the smart bet? And if you are going to bet $100 on that one hand, isn't your expected profit $10.70?
