Kill Pots and Standard Deviation

[garyc8]

I hope I've chosen the correct forum for this.

Some time ago I sat in a 5/10 game (played with a full kill). My roommate was already in the game, and in the midst of a "rush" with the $10 kill blind posted. By the end of the streak he'd won 6 in a row (a couple with rubbish he normally would not have played). Of the 6 in a row, 5 were kill pots.

Later, at home we were logging our results. My friend noted that his standard deviation was only 120. Per the 3 deviation rule, he should expect not to win any more that 360 + his win rate in any given hour. He said that he had won at least $600 in the hour of his "rush".

I thought about it for a moment and concluded that the numbers were skewed by the kill. His SD of 120 was built on a large sampling of hands, the large majority of them being standard 5/10 hands. In that particular hour, most of the hands he played were 10/20 hands.

I also speculated that the kill would be unlikely to have the same effect in the opposite direction, since he would be very unlikely to play a large number of kill pots in one hour, except under the very cicumstances under which it had occured ( a winning streak).

Any thoughts welcome.