what has the least variance

[greatwhite]

I posted this in the variance forum, but they are too busy whinning about bad beats there. O8 is often regarded as a game with very low variance. However, I was wondering if it is lower than stud/8's. I know that both game's variance are lower then non-split games. I also know that omaha high as a high variance than stud high.

[beset]

I have no data or anything but my own experience to rely on, but I would think the extra betting round in stud/8 would have to make it higher variance.

[simplyred]

I don't recall he specifics but Poker Essays 1 has some comparison of variance for different games. Based on my recollection of that, I would think Stud/8 is higher variance (perhaps significantly).

[januarymute]

SR, I'm unfamiliar with Poker Essays 1. Can you provide a link or something? I'm particularly curious to have a look at the author's method for determining variance--it seems to me that it would be very difficult to analyze this without tracking software like Pokertracker. Not impossible, but difficult... and so I think an "authority" might be prone to taking the easy route (guessing) instead of committing to mathematical rigor.

Also, FWIW I don't believe that an extra betting round seals the deal for S8 having more variance.

[Raul Wong]

Ive played a lot of stud8 and O8 and while I have no evidence to support my opinion that stud8 has more variance I am almost positive that I am correct. While I havent done a ton of sims for O8 I have done a million for stud8 and most people would be astonished about how minor an edge is really being pushed in that game.

[simplyred]

Poker Essays is a collection of Mason Malmuth's columns from various magazines. I read a friend's copy so I can't eleaborate on his methods. I think the rigor is certainly there but it is obviously only as good as the assumptions.

http://www.twoplustwo.com/books.html#Poker%20Essays

[Jettz]

ack I agree completely that stud8 has a much higher variance based on personal experience. I know I can play absolutely horrible stud8 players that will play any 3 and not be shocked after losing 30bb in 300 hands. However, I will hate the world. Against decent opponents, there is a ton of check/call/pray they don't have the low straight. While I haven't run the sims, I agree completely that the edge in stud8 isn't all that large and that the variance isn't all that small.

[Chamonyx]

Although it is a split pot game like O8, S8 sees far fewer 1/4 pots at show down due to the lack of community cards. This means that the distribution of results is concentrated on +/- pot, +/- half pot, so the E(x^2)is much higher than for O8.

Consequently unless a given player has a much higher edge - i.e. E(x) - at S8 than at O8, his variance will be higher [=E(x^2)-(E(x)^2)]

This is also partly why Omaha high has a higher variance.