Sunday, June 14, 2009

Jackpots, My Win Rate, and Volatility

Shortly after winning over $20k in a jackpot and another $1500 table share, I won yet another table share of over $1000.

I make daily records of my results in an Excel spreadsheet. I have it set up to calculate my hourly win rate and hourly standard deviation. After I won the jackpots, I simply included them each in a row of my database, listing each as having taken one hour. Of course, each jackpot actually only required one hand, so the assignment of one-hour time period was completely arbitrary.

Needless to say, the jackpots have a had an effect on my win rate and standard deviation. My SD especially has gone haywire: it went from about $350/hr over the course of my entire career to about $600/hr. My SD for this year comes to $1218/hr after being only $338/hr last year. In other words, recording the jackpots in this way is screwing with my volatility metric. What can be done about this?

I could just leave it like it is. After all, it's true that the jackpots increase the volatility of my income. However, the arbitrary choice of 1 hour per jackpot has far too much influence on the SD for me to be comfortable. For example, if I change the entries to say that each jackpot took only 1/10th of an hour, my SD increases from $600 to $1564/hr, and my 2009 SD skyrockets to $3721/hr.

I think a better solution is to try to separate out my poker-playing results from the jackpot results. I'm reluctant to do this because, in reality, the two are entangled quite insidiously. For example, I am much more likely to call with QQ when there are three aces on the board if I know there's a good chance I'll win a $20kjackpot. Similarly, my opponents certainly play differently - a few will play anytime they have an Ace before the flop because these hands have the best chance at making a jackpot. Many players would not be willing to play the game in the first place if not for the jackpot, and most players won't play with fewer than 4 opponents (the minimum needed for the jackpot to go into effect). Also, my win rate in jackpot games has always been dampened by $2-$8/hr because $1 is taken out of each pot for the jackpot.

Still, what I'm going to try to do is separate out the jackpot results from the poker results. I'll ignore all the above complications except for the most straightforward: I'll subtract out my jackpot winnings and add back in the $4/hr that I approximate I pay into the jackpot drop.

One other thing: the Bicycle Casino is being sued by some poker players over the jackpot drop. The owner is quoted a few times in the LA Times article.