Defending at Disequilibrium

Against an opponent using an intelligent mix of bluffs and value bets, defending is always a hard chore. If you read a bunch of modern books on poker, you’re likely to come across topics like pot odds, minimum defense frequency, considering the effect of blockers (or the lack of them) on the size of your calling range, and discussions about what kinds of hands are thus best used for calling in various spots.

With pot odds, you compare the ratio of the bet you are facing to the size of the pot to see how often you need to be ahead to call profitably; with minimum defense frequency, you discover how often you need to call to keep bluffs from being profitable; with blockers, you’re able to make a more informed decision about the ratio of bluffs to value in your opponent’s range. This all can then be especially useful when all you have is a bluff catcher, a hand that loses to pretty much all of your opponent’s value hands, but beats basically all of their bluffs.

But what happens when your opponent is not a powerful AI, a solver, or even a top pro? This is presumably the situation most of us find ourselves in at Replay, and against more ordinary players, it’s unreasonable to expect a perfectly balanced ratio of bluffs to value. How do you change calling frequencies in such an environment?

If you remove even a few bluffs from what had been an optimally balanced range, the new solver solution will be to basically completely eliminate calls with “bluff catcher” type hands (and even hands somewhat better than that). Conversely, if you add even a few extra bluffs, rather than calling at GTO frequencies with these hands, you’ll shift to calling with nearly all of your bluff catchers close to 100% of the time (a maximally exploitative adjustment). Another way of saying this, is that against someone that doesn’t bluff enough, you can basically stop making hard calls, while against someone that bluffs too much, you should be calling much, much more often.

But making large adjustments like this is easier for your opponent to notice, exposing you to counter adjustments, and so if you think you have a thinking opponent that is paying attention to your play, a more measured modification that is less likely to result in your opponent adjusting their bluffing frequencies can often allow you to maintain a smaller but more consistent advantage over your opponent over a longer time frame.


As a tournament player, I’m facing a little different dynamic. I don’t think much about smaller advantages over longer periods of time.

Calling to bluff catch is OK, but raising is also a valid option sometimes. I know, never raise into a polarized range, and blah blah, but raising, especially when you unblock V’s likely bluffs, does the same thing without giving anything away.

I don’t really see the point of wanting to keep V’s bluff frequency high if you are just going to mostly fold.

I want them to make adjustments because most players I meet tend to drastically over-correct. You should anticipate, then see their adjustments if you pay attention to their betting patterns. Since you have basically forced them to adjust, you know how they will adjust, usually before they know, and you will be ready with the counter.

Getting them to change their game can also put them out of their comfort zone, which is a good thing.

All of this is much easier said than done, of course.


If someone bluffs at a really high frequency, you want to ask yourself what their range really looks like. Is it polarized? Or is it just weak? Further, I think the oft quoted wisdom about attacking polarized ranges is really just about how to play the middle of your range, and does not apply to the top or bottom of your range. You mostly don’t want to attack with a hand that has show down value, but that will lose to most of the calling range, but if your hand has no showdown value, and you think your opponent’s range is very weak… need I say more?

Yes, I think this is one of the benefits to raising and 3 betting pre-flop at higher frequencies than most people are used to. In a limped pot, most people don’t feel overly committed, and they are in their comfort zone, calm and relatively rational. In a raised pot, many will have a hard time really thinking normally at all, with heart pointing, and their poker decisions often being more emotional than rational.


Yeah, I definitely agree with this.

Still, if raising will get them to fold all of their bluffs and some of their value hands, it’s worth a shot sometimes when you have little or no showdown value.

Not all who over bluff are the same though. Most that I see will insta-fold to any raise, and a bluff 3-bet is rare in the games I play. This makes me much more likely to attack.

Occasionally, I do see maniacs who are willing to bluff-shove if raised. Of course, they will take the same line with premium hands too, so calling is a better option sometimes.

I don’t have the luxury of seeing thousands of hands against any specific player, which makes it so much harder to know how they are approaching the game. Add to this the “sudden death” aspect of tournament play and it just gets harder.

As a mostly exploitative player, I don’t worry too much about “optimal” frequencies against specific players. Against some, I’ll do something more, against a different player, I might do that thing less. If my frequencies are overall optimal, it’s just by chance, not design.

One more reason to raise instead of call is something I think about as “interlinked tendencies.”

There are a few examples of this, but the one that comes to mind here is that those who bluff too much also tend to overplay bottom pair type hands and small to mid pairs. Basically, if they’re willing to bet air, they’re willing to bet any pair.

To me, this suggests that it’s rarely a good idea to try to bluff catch with ace high type hands against this specific kind of player.

At the same time though, something as strong as second pair or even top pair (and more rarely, even hands like sets), are reduced to just being bluff catchers.

At the end of the day, though, for both raising and calling, it comes down to a similar question. For calling, how much of your opponent’s range are you ahead of; for value betting, how much of your opponent’s calling range are you ahead of; and for bluffing, how much of your opponent’s range that is ahead of you will fold (or re-raise with hands behind).

I have a big problem that is as bad as yours, but, maybe worse, I would like to discuss with you if you would like to see if you want to hear???