Thursday, November 17, 2005

Bad Beat Jackpot

Last time I checked, Party's bad beat jackpot had climbed above $170,000. I am curious as to how to calculate the EV for playing the jackpot tables.

To avoid having lazy me delve into the math, I read this article by Brian Alspach that concludes $158,300 is the magic breakeven number when making reasonable assumptions of how real people play. I'm not certain if this number needs to be scaled up to account for the 20% of the jackpot that is used to seed the next jackpot and the 10% skim by Party.

According to bonuswhores (see the Bad Beat Jackpot section), the magic number is between $130k- $190k, depending on the stakes played. They provide absolutely no details on the assumptions used for these calculations.

For the sake of argument, let's use the $158,300 number. Assuming a 10 seated table, then each players' contribution to the jackpot is $0.50/10 or $0.05 per hand. I guess that means if the jackpot was $316,600 that the jackpot EV from each raked hand at the jackpot tables is +$0.05? That seems like a lot of free money. In a low stakes game, $0.05/h is on the order of magnitude of 1BB/100h.

How am I supposed to know who's magic number to go by? I guess to make things simpler for me, I will wait until the jackpot exceeds $250k before considering to bring my business there. The +EV below this point is pretty small anyways, and god forbid that I actually play on these tables when there is negative Jackpot EV in case the assumptions on bonuswhores are too optimistic.

Another factor to consider related to jackpot tables is the quality of the opposition. From the threads on two plus two, the general consensus is that at smaller jackpot sizes and lower limit tables, the average player is worse than average compared to non-jackpot tables. The reasoning is that i) tougher +EV seeking players will avoid playing in the -EV jackpot environment resulting in better games, ii) the minimum stakes of $2/$4 is higher than the minimum stakes at non-jackpot tables forcing some jackpot seeking players to play higher limits that their norm.
However, as the jackpot size grows very large, the consensus is that the tables start becoming highly populated with tougher multitabling TAGs. I suppose you could combine the pluses and the minuses to be a wash. (yes, a great analysis cop out...)

A final point to consider is which stakes are most ideal to play at in a high jackpot environment? It stands to reason that on a per capita basis a lower stakes table is likelier to hit the jackpot than a higher stakes table since the supposedly looser play should result in a higher percentage of potential qualifying hands to be played.

I recall reading on Scurvy's blog that he multitabled the $2/$4 jackpot tables the last time the jackpot was large, although he normally seems to play between $15/$30 and $30/$60. I wonder if his decision to play at the $2/$4 stakes were influenced by the higher likihood of the jackpot being hit.


My latest initial foray at the Party $10/$20 tables is off to a decent start, but only after a very small number of hands. After ~700 hands, I'm running at ~+11.5BB/100h. I hope the regression to the mean does not come in a particularly painful fashion.

It is going to take some time to get a decent number of hands in. I've probably been averaging slightly less than 2 tables at a time. I only went up to 3 tables for a brief 30 minute period when a 3rd overwhelming good table was available.

I've also noticed that games at $10/$20 are more heavily centered on the providers. Once those players go broke and/or leave, the table quickly breaks down, often as quickly as within 1-2 hands.

No comments: