Quote:
Originally Posted by philnewall
I'm too busy with school to go looking into bot databases to investigate this issue further, but I'm happy to guide you if this is something you wanted to pursue.
Phil
That's cool Phil. I hope school is going well. If I was to investigate the bluffing frequencies of bots, I'd probably start by looking at "might be the best hand draws" such as King high or Ace high flushdraws, King high or Ace high straight draws, and those that are a combination of both. My hypothesis is that with these hands, we'll see high pot share bluffs on the flop and turn and lots of giving up on the river. I think this will be somewhat evident that making better hands fold on the early streets is not a condition of bluffing range construction.
Then I think I'd look at pure bluffs, which I think will be given up on the early streets unless the board is such that it dictates a high folding frequency for the opponent.
For example:
huhu
small blind raises 5
4
, big blind calls
flop: 8
T
J
big blind checks, small blind?
Here I think that there is not enough fold and draw equity to result in a high enough pot share for the small blind to bluff with, resulting in a check down unimproved to the river, where this hand will be turned into a bluff if checked to.
another example:
huhu
small blind raises, 5
4
, big blind calls
flop: A
A
T
big blind checks, small blind?
Here I think that the board will dictate a high folding frequency for the big blind, resulting in a bluff with 5
4
and other hands with a similar pot share, as well as the high equity broadway draws and flushdraws. This may seem unbalanced towards bluffs on the surface, but I think that the big blind will have no choice but to continue folding a lot on such a board, otherwise the big blind would have to pay off on the flop with very weak hands, which would in turn benefit the small blinds strategy as a whole.
Of course, it's tough to find the fine line between a static board and a dynamic one if we're using the opponent's folding frequency as a qualifier. I'm not sure, but perhaps this could provide reassurance to your HLM and DQM scores.
Changing the subject a bit:
A friend of mine is currently employing a mixed strategy, but in a way that I either don't understand or in a way that I think is wrong. He's donk check raising "sometimes" with the very top of his range on the flop, which I think is a clear bet 3 bet spot. I think mixing should or will occur only on the margins where both decisions have the same expected value because to do otherwise would be missing value somewhere and it would therefore not be gto.