Poker is inherently a game of numbers - card values, chip amounts, frequencies, combinatorics - and as the multiwayness of preflop is too complicated for existing abstractions to approximate it remotely well, we are left with qualitative assessments that attempt to reduce the math to statements of 'relatively more' or 'relatively less' in 'some spots'. The issue is when these reductions grow so detached from mathematical reality that all arguments for anything are assigned equal viability and force, thus allowing conclusions to be cherry-picked to suit the analyst's disposition or convenience.
Quote:
It is fairly trivial to not be capped on many boards. Being deeper makes 3betting better because villain can't reach a leverage point with a reasonable 4bet, which means we can 3bet a really large range and then have a 5bet and a flatting range.
Past a certain point of depth, it would follow from this argument that Villain will be able to reach a leverage point with a reasonable 6-bet, making 4-betting once again hot and 5-betting not - and so 3-betting must tighten up again. One would imagine this occurring for 2.5x BTN opens at 600 BB effective stacks, for instance. Past another threshold, 3-betting can rewiden, because the last person to put in the stack-threatening amount of money is OOP, with the 7-bet.
What this argument fails to consider is that the probability of reaching each leverage point shrinks at a devastating rate. This is due in large part to the Ace blocker preflop. With AA, it will eventually become so likely that someone has a bluff as opposed to the single remaining combination of AA that flatting becomes a highly competitive option. As a result, I do not expect the increased EV-significance of these lines to nearly catch up with their rapidly diminishing frequency.
In this particular spot, I would only 3-bet SB vs BTN, and never flat, but for vastly different reasons: I expect both EV(3-bet) and EV(flat) to be monotonically decreasing with stack depth, but EV(flat) to be hit much harder because of BB's incentives, especially considering the open size.