-
Theory question: 10d8h5c board
I’m starting to play with subsets in the solver, which basically allows you to run spots across various representational flop types. Here are the basics of the scenario. I’ve been running a fairly standard MP open (2.25bb) and big blind defend spot. We go to a flop with 6bb pot (includes bb ante). I can share ranges if curious, but they’re fairly standard. I’ve then given the big blind the option of donk leading or checking and then given the MP player three options for betting when checked to (I did so because I was curious about overbets on flops): 1.8bb (30%), 4.8bb (80%), 7.5bb (125%). I then ran the simulation at 20bb, 40bb, and 100bb effective.
To my surprise, the deeper we got, the more the solver leaned toward that flop overbet sizing. This is not some trivial, never-really-usable finding. When we’re deep the solver likes this overbet flop cbet about 25% of the time. In particular, it liked using it on the 10d8h5c board. On this board when 40bb effective, it still chose this sizing, but only 18% of the time. On this board at 100bb effective, it chose this sizing nearly 76% across our entire range. Just by adding a flush draw (10d8h5d), this plummets to 51% of our range, but it still one of the top flops we elect to go with this overbet.
Some of the other findings make a ton of sense to me. Our most common small cbet is on paired boards with a broadway card (JdJh6d or QsQc10c); our most common check backs are on boards that give the big blind lots of possible advantages (As3c3h, 7d6d5h) or boards that we absolutely dominate (AsAcKh).
But I’m struggling to think about this board and others like it where the solver leans toward an overbet cbet. Other boards where the solver really liked this sizing across range include:
Td8h5c; Td8h5d; 9c7h4d; 9c6s2s; 8d4d2h
Since these are mostly disadvantage boards, I currently play a lot of boards like these as a large range check back against competent opponents. What are your thoughts about this spot as the MP player? Do you ever employ overbet cbets when deep? Why do you think the solver prefers boards like these to employ them?
-
2 Replies
-
This is very interesting….While I don’t use the flop overbet myself much, I have studied a couple spots where it makes sense from a GTO perspective.
I think the range vs range advantage is key here, we are opening 20% of hands or so and BB should defend a lot more due to the good pot odds and closing action…so in theory over 60% of hands at 40bb. So a bigger percentage of our range has a strong top pair and we also have overpairs, and all the sets which have great equity vs most of BB range. I think this is why the solver can use a polarized strategy for the c bet to attack a condensed range, lets it include more bluffs, and build a big pot at deeper stacks with our value. But it also has to be coupled with a checking range of hands that want to get to show down cheap and some traps.
My understanding is that for c betting we can either:
1) cbet small 100% with our equity and nuts advantage
2) Polarized large bet (w more overbets the deeper we are) This strategy lets us have a lot of bluffs (e.g. K high, Q high, backdoor draws) to couple with our big value hands, and we have a checking range that is trying to get to show down cheaply (e.g. weak pairs, some weak backdoor draws). Given that we have the nuts advantage and can include more bluffs, then the BB will have a hard time defending to a large sizing. We also inflate the pot when deeper with our value hands that have great equity against the BB’s capped range (they didn’t raise pre).
I think in practice the cbet small is much easier to implement (we don’t have to come up with a check back range) and we can assume the field overfolds to small bets. I also suspect the field doesn’t defend the BB as wide as a GTO approach would, so we actually arrive with more similar equities on these T high flops and maybe our actual range advantage isn’t as strong. But there has to be a V type where the overbet does better….maybe if they defend wide preflop and are prone to XR or float a small bet…and stacks are deeper…then we can punish with our nuts advantage and overbet (block XR, build a pot with value, can easily fold the weak hands to a raise and get it in with value)
-
Will you try a few more flops for me with this scenario?
T85 has many straight draws and more importantly, the T and the 5, the only cards that are required to build any five card straight in poker.
Try flops like (T84) and (J85) and (985) if you would and let me know what you get.
I predict that a lot of flops containing the T and 5 will on average be higher than flops with only one T or 5 and definitely higher than flops with neither a T or 5. Without a T or 5, the chance of a straight is much lower because it needs to hit the turn or river or be in your hand.
Log in to reply.