I've been looking through a ton of poker stats lately in PT3. Specifically, I've been very interested in dissecting consistent, winning players' statistics to see what they're doing that the majority of losing players aren't. For example, do they play tighter than the average player? Turns out that the answer is Not Really. At the full ring $25NL tables, the top fifty ptbb/100 winners all had VPIPs that averaged 14.1%, while the average for all other players is actually a tad tighter at 13.5%. This was surprising to me. It seems that both winners and the average losing/break-even player at these stakes understand that Tight is Right, but the winners are slightly looser.
The winners, however, are much more aggressive overall. Again, looking at the top fifty players in my database (as measured by bb/100), the average PFR is 13.4%, which is well within 1 percentage point of their VPIP. The average player, in contrast, has a PFR that is 9.5, which is a whopping 4 percentage points lower than their VPIP. Simlarly, the cold call preflop (CCPF) numbers are also very low for the sharks compared to the fish. If they're gonna play a hand that someone upstream has already opened, they're more often going to raise and not call. Said another way, the winning players wait for good opportunities, and they they hit those opportunities very hard. It's that old saw: if a hand is good enough to call with, it's probably good enough to raise with, too. Limping appears to be for losers.
Aside from the VPIP difference, none of this is too surprising, of course. We all have been taught that a tight, aggressive (TAG) approach is a winning strategy to poker. If you want to win long term at poker, you need to wait for big cards (or good opportunities) and then try to knock it out of the park. What good is a million hands of data if it's just going to tell me the obvious? The answer is simple: it's not just going to tell me the obvious. There are a number of other subtle, but potentially important facts buried in all that data. One of these facts that I may have uncovered relates to stealing. Specifically, the data shows that the very good players don't actually steal a whole helluva lot.
A few posts ago I mused about stealing percentages. Should one steal every chance they get (a'la Flyboy's experiment in which his ATS was 100%)? Or should one be a little less extreme? I had a recent conversation with a pretty good all-around poker player who said his steal percentage in cash games was 65%. Is this a "better" number to shoot for? I personally steal at around 39%, day in and day out. It's been profitable, but I always wondered if I should be stealing more than that.
Well, the answer might be that I should actually be stealing less than what I currently am. Of the top fifty players in million hands of data, the average steal percentage is 34%. Only a handful of sharks actually steal more than 40%. In comparison, however, the fish steal even less on average (25%), and some of the biggest losers of the game steal at greater than 50%. In other words, there seems to be a sweet spot somewhere in the one-out-of-three zone for optimal stealing.
Another interesting stat that merits further pursuit is the CBet Flop percentages. The continuation bet average for all fish is a relatively high 60%, but the sharks are c-betting even higher at 71%. Now, we all know that a flop hits your hand only 35% of the time, so most c-betters higher than this are full of it, but the sharks are full of it 11% more than the fish. Interesting.
There's a lot more to delve into, and I feel like a kid who's been given the keys to the candy shop. Lots of fun ahead...
... finally, a silly cartoon I saw on a statistics site that made me laugh. Not really poker related, but it does point out the danger of reading too much into small sample sizes:
All-in for now...
-Bug
No comments:
Post a Comment