Quote:
Originally Posted by Shark Junge
I am confused now. You said both are standard deviations.
Of what? They can't represent the same standard deviations.
I want to know so that i can workout confidence intervals for the ev of a play. Which one should I use then?
Thanks
When the Monte Carlo engine computes results, it will base these results on a large number of simulations. Because of this method, all results will only be an estimate. Since I felt, given that all numbers had an error, that it should be made clear to the user what that error was, I've added a field (std: ...) behind the most relevant numbers.
For example, see the pic below. Here we have a popup that says (amongst other things) that, based on the simulations, the equity is estimated as being 33.9%. However, the software expects to be off by approximately 0.27%. So reasonably, you can expect the actual equity to be within the range [33.6%,34.2%].
So, basically, whenever the software says something like:
Equity: 56% (std: 1%), it's basically saying "my best estimate of this number is 56%, however, it's perfectly reasonable to expect the real value to be between 55% and 57%".
This applies to any number computed by the monte carlo engine. The std number is simply the expected error in that number.
And the Variance is just a measurement for how much you can expect your stack to fluctuate by performing a certain action.
So if the popup says:
EV: 8 (std: 1) Variance: 40
It means that
- the expected value of the move is estimated as $8.
- the error in the EV is $1, meaning that it's most likely between $7 and $9.
- you can
roughly expect your winnings/losses to fluctuate between -$32 and $48