Tuesday, February 2, 2010

"I'm All-In..."

I made quite a discovery yesterday. In attempting to further investigate what I consider to be the validity of EV statistics (or lack thereof), I incorporated a couple of filters in my stats to determine how many times I was all-in before the river (when EV stats come into play), vs. how many times I am not all-in before the river (when EV is not affected).

Since starting to play no-limit in the first half of 2009, I’ve logged approximately 310,000 hands. Of this amount, I was all-in before the river approximately 1200 times. That’s less than ½ percent of all hands played. To put it another way, my EV statistics fail to account for more than 99.5% of the hands that I play.

But that wasn’t the surprising discovery. Here’s the discovery that nearly caused me to lose consciousness:

  1. In the 310K hands that I’ve played, my overall winrate is an admittedly modest 3.01 bb/100. (Hey, I’m still a relative no-limit neophyte, lol.)
  2. In the 1,200 hands in which I’ve been all-in before the river, my winrate is 742.75 bb/100.
  3. In the other 309K hands, I’m showing a modest monetary LOSS. (Due to playing at different blind levels, my overall winrate is actually a positive 0.11 bb/100 over this sample.)
So, essentially, I’m a breakeven player over the course off 99.5% of the hands that I’ve played, and a huge winner over the course of the remaining 0.5% of the hands that I’ve played. Virtually all of my profit has come from less than 0.5% of all of my hands.

I’m honestly stunned by this. At first, I thought it was rather damning evidence that my play is sorely lacking in some respects (and I suppose that may still be the case) - but in reviewing the stats of other winning players within my database, the pattern seems to repeat itself over and over again. Players who have material positive winrates overall seem to be breakeven players or net losers over the course of all hands in which they are not all-in before the river – which is obviously the vast majority of hands.

Are these stats ridiculously skewed, or do most players have similar patters?

(I’ll post on Stoxpoker for feedback/commentary…)

(As an aside, I'd love to know how to replace those stupid flowers up above with actual bullet points lol...)

Music by the ELS Experiment