PDA

View Full Version : A Fossilman Post from 1997

05-29-2004, 02:17 PM
I remember this post in particular. At that time my sense that Greg was a tad frustrated. Hope Greg doesn't mind, to me at least it's very interesting in light of what Greg has done.

Short term estimate of tourney EV
Posted by: Greg Raymer (graymer@gcwf.com)
Posted on: Monday, 17 November 1997, at 8:56 p.m.

I am only able to play about once per week, and I enjoy tournaments. Thus, I try to play in them every time I can, and I usually am playing in the medium buy-in events (i.e., \$50-300). Because I can only play in a few of these events per year (about 35), and because I have only been playing seriously and keeping records for 2 years, it is really not possible for me to obtain an accurate estimate of EV based solely on a statistical analysis of my monetary results. In fact, I would probably have to play for a few more years to get a reasonably accurate figure. The thing is, most of these events have fields of 75-300 players, and so even if I'm a favorite (let's say double the chance of winning of an average player), I still would only expect to win about once every other year!

Anyway, I have been exploring an alternative method so that I can estimate my long term EV using short-term results. What I do is, for each event, I record the number of entrants other than myself, and how many competitors remained when I busted out. By comparing these figures I can get a numerical ranking of my play for that tournament, with a score of 0 when I win (0 competitors remaining out of whatever number started) up to 1 if I bust out first. My hope is that by averaging out these numbers, I can come up with an estimate of whether I am a winning or losing tournament player.

For example, if you were to enter 10 tournaments at \$100./event, placed 4th in one of them for a \$600. payday, and did not win any money back in the others, you look to be a losing player. However, what if you used my system to average out your result versus the field in these events, and obtained a score of 0.25. This would indicate that you are usually outlasting 75% of the field, but are not quite making the payoff (usually about 10% of the field is paid). This suggests that you are a good tournament player with one of two problems, either you're not getting lucky near the end, or your end game is weak.

I would appreciate your input as to the usefulness/validity of this method.

Thanks, Greg Raymer (FossilMan)