Poker stove question

[Derek in NYC]

This is a neat tool.

Forgive the dumb question for all you math/CS guys out there, but I was wondering about the "enumerate all" and "monte carlo" options for modelling results. Can somebody tell me what the difference between them is, other than Monte Carlo comes to a very similar answer much slower.

(Interestingly, the only exception I have found to this is when you run a RH against another RH. In that preflop setup, monte carlo concludes almost instantaneously that the results should approach 50/50... which is what intuition also tells you. Enumerate all, by contrast, starts with a severe imbalance 65/35, then as more simulations get run, it approaches 50/50 but does not appear to come to an equilibrium.)

Anyhow, for my purposes, is this all just noise in the system, and can I just use enumerate all when messing around with Pokerstove?

[John Paul]

I have never heard of Poker Stove - what is it? However, I can take a guess at your question. "Monte Carlo" is a statistical term used when you simulate something a whole bunch of times. So for examples, say you wanted to know how likely you are to win heads up if you hold AA and nobody folds. The Monte Carlo techniques would be to simulate a large number of random opponent hands and community cards and find out how often AA wins.

Enumerate all, I am thinking, is where you do not simualte but figure out every combination. Imagine if you wanted to find out how good AA is heads up. You would figure out every holding that your opponent could have and then every possible set of community cards and use that to figure out your question.

[Robk]

enumerate all computes the results of all possible boards, and gives you an exact answer. the downside is that for more complex requests (ie hand range 1 vs hand range 2) it takes a few seconds. the intermediate percentages are essentially meaningless, wait until the program finishes.

monte carlo just throws down a bunch of boards at random and tells you the results (it won't stop running). the results are always quite close to the enumeration.