Variance and bad players

[Zaghomat]

I see a lot of people who seems to think that if you play on a site full of "donks", your variance will be higher.

I can't see any reason why your variance should be higher, I think it would be the opposite (because of your increased winrate).

Am I right?

[Tiki]

Your variance depends on both who you play and how you play.

For instance, if you play at a table full of very loose players and wait patiently for good hands then you should win with a relatively low variance if you are good at choosing your spots.

However, you will probably win more if you loosen up yourself. This is because you will be gambling on typically smaller edges. Your variance will increase as a result of this type of play however.

So basically, if you play optimally at a table full of loose players then your variance will go up.

[pzhon]

[ QUOTE ]
I see a lot of people who seems to think that if you play on a site full of "donks", your variance will be higher.

I can't see any reason why your variance should be higher, I think it would be the opposite (because of your increased winrate).

Am I right?

[/ QUOTE ]
People mean different things by variance.

In mathematics, variance is precisely defined. Mathematical variance is a measure of how spread out a distribution is from its mean. It is the square of the standard deviation. The mathematical variance does not decrease in the sense you meant when your win rate increases. The mathematical variance may increase due to the looseness of your opponents, or it may decrease because your opponents are less aggressive. These effects mostly cancel in limit games.

When some people say "variance," they mean particular effects of the mathematical variance such as downswings, not the mathematical variance itself. Downswings tend to be shorter and less severe when your win rate improves.