Found something else interesting from my PT4 analysis. Hopefully I didn't screw up the filters this time
I decided to take a look at river calls to see how I'm doing there and was happy to discover that I'm quite profitable, winning at 369 bb/100 (admittedly a small sample of just under 400 hands). I'm also profitable calling rivers from all positions.
But here's where things get interesting. I decided to break it down by bet size (as a % of the pot) and the pot size itself. Here's what I found:
1) I am unprofitable when calling river bets in small pots (<20bb) but profitable when calling bets in large pots. In fact, my win-rate increases as the pot size increases. This seems to indicate that when pots are small, I'm more likely to make a "what the hell, let's see it" call, but my instincts are good in large pots.
2) I broke down the bet sizes into 4 groups: 20-35% of pot, 36-50%, 51-66%, and 67%+. It turns out that I'm profitable at all sizes except the 51-66% size. What I find interesting about this is that the 51-66% pot-sized bet on the river is classic value sizing. While the sample size is way too small to draw population based conclusions, I wonder if there's something there in terms of a population-based exploit (I play buyins ranging from $3-$22). It could be that most players at this level use a 51-66% bet size on the river as a value bet, while opting for either a small or large bet when bluffing (or incorrectly betting weak value).