Jazbo has values for the covariance, but the games Jazbo lists are probably not the games you play. It would be nice to have variance and covariance listed in the games database, at least for the popular N-play games.
Covariance cannot be estimated from variance. You have to have software that calculates the probabilities of each outcome for the optimum hold for each hand in order to get the covariances. This is tens of millions of values, so it cannot be determined by hand.
If you do have software that will do this calculation, you will still not have the distribution of payoffs for the game. I prefer to use a Monte Carlo calculation to estimate the total pay distribution for multi-play games. Then, I can also calculate the percentage of return that result in W2-G payoffs, as well as W2-G frequency. This is important in states with nasty tax policies.
Finally, spin poker covariance is not the same as regular N-play covariance. Spin poker draws are correlated.
Effen
--- In vpFREE@yahoogroups.com, kelso 1600 <kelso1600@...> wrote:
>
>
> You might want to take a look at the following website:
>
> http://www.jazbo.com/
>
