[swiftrhett]

Maybe the swing metric you brought up is something like (profit / #ofgames) - (standard deviation / #of games). So, if a player has an ROI of 20% and a standard deviation / tourney of 1 buyin, then the tourney buyin scales as the square root of the number of games he plays. So, for instance, if this player does 100 games / week, that's a "swing metric" of 20% * 100 / 1 * sqrt(100) = 20 - 10 = 10. Basically, this means, even one standard deviation from your average, you expect to be up at least 10. You could divide them if you're looking for relative swings I guess. The problem with quantifying this is that players who play wildly different amounts of tournuments see this very differently, and in general your standard deviation goes down relative to other things if you play more.

[ QUOTE ]
[ QUOTE ]
And I believe that was what OP was asking.

[/ QUOTE ]

Fair enough. I mean, I know that's what the OP was asking. That's what everyone is asking when they ask about variance. Trouble is they keep calling it variance, and that's only a part of it.

...then the math inclined get frustrated when nobody gets why variance is pretty constant.

Believe it or not, I am not actually that formally trained in math myself, so I don't know it the metric I describe has a technical name. If it doesn't I nominate 'fluctuation metric' or maybe 'swing metric'. Even 'relative variance' seems ok, as that is what we are really talking about - variance relative to average profit.

I am pretty sure it must already have a name

Regards
Brad S

[/ QUOTE ]