This One Exploit Drastically Increases Your Winrate - Exploit population on Axx

3bet pots are some of the most important spots to master. They include big pots and a lot of money, but also occur often. So getting a slight edge in any 3bet spot can drastically increase your overall winrate.

Today, we're going to do that.

How many more bb/100 are possible to achieve in a spot? 5bb/100? 10bb/100? 100bb/100?

Today we'll answer that question by looking at a 3bet spot on an Axx flop.

How can we exploit population when they 3bet from the BB, we're on the BTN and we see an Axx flop? (x = 9 or lower)

The first thing to do to get a meaningful answer is to make some simplifications. Otherwise, the amount of possible outcomes is way too big to fit everything into digestable learnings.

So instead of looking at all possible Axx type flops, we'll work with an example flop.

For that, we'll use A♠️6♥️3♠️.
Of course, exact hands will vary on different flops (6♠️7♠️ will play a different role on this flop than on A♥️4♥️2♠️), but the concepts as a whole will be applicable to all flops of this type.

Secondly, we will use only one bet size for each player. If we used multiple cbet sizes, adjustments that are made would become too complex to get useful takeaways. So for the BB, we'll work with a 30% pot size cbet, as that is the most used one in GTO and by population.
For the BTN, we'll use a 25% stab size and a 40% pot raise size against the BB's cbet. As those are the highest EV sizes if we only use one.

How often does the BB cbet in GTO and in reality?

GTO would have the BB cbet 42% of the time on this flop.

In reality, we do see population cbet way more often on Axx flops, however. 72% of the time.

So how should we adjust against this? How can we exploit the average opponent in the BB?

Well, first we'll have to look at the cbetting composition. By upping the frequency to 72% in Piosolver, it automatically creates a cbetting range that is balanced. And to simulate population's play that is good enough for now. Because population neither deviates to an extremely value-heavy nor to an extremely bluff-heavy imbalance.

(If we also wanted to know what to do against a heavily unbalanced player (whether extremely tight/value-heavy or extremely aggressively/bluff-heavy), we would have to adjust the range composition. That is not the topic of this newsletter however, so let me know if you're interested in that and I can consider it for another one.)

BB GTO cbet

BB Population's cbet

Looking at what we should do against a cbet, we see almost no difference. We still call around 78% of our range. The only difference is that we could start raising a small amount of the time and fold a few hands less. But again, the difference is rather small.

BTN vs. cbet GTO

BTN vs. population's 72% cbet range

However, if we look at what to do against a check, the difference is much bigger.

Against a GTO check, we would bet ~46% of our range as the BTN. However, against a player that only checks 28% of the time, we could now bet much more often - 81% of our range!

BTN vs. check GTO

BTN vs. population's check

Here, the BB would check-fold 47% of the time after cbetting 72% of their range - compared to only 32% when checking a GTO range.

BB check-fold GTO

BB check-fold after cbetting 72%

Why? Because when the BB cbets more often, even if we have the solver construct a balanced 72% cbetting range, it still leaves more of the lowest equity hands in the checking range. Simply because we want to put less money in the pot with them. So when facing a bet, the BB cannot continue as often in that scenario.

Now, we could just end this analysis here and move on. Because we've node locked the spot and found out what we need to do differently than in GTO. That's all we can do, right?

Well, that's what you're usually being told. But what if I told you there's more?

If you look at the 81% stabbing range, you'll see that it's a well balanced range. It includes value hands, but also bluffs. Of course, because otherwise we could get re-exploited.

...Right?

Well my friend, this where we'll go deeper.
Technically, yes, we could get re-exploited. In theory, if the BB knew exactly what we're doing, they could exploit every little imbalance in our betting range. But... that's not how real life works.

One thing that would need to happen for the BB to prevent us from exploiting them to the bone, is that they'd have to adjust their check-calling frequencies of each hand they check.
What does that mean?

Well, if you check 100% of your range in a spot, you still have all hands in your range when you face a bet. All the nuts hands, the good hands, the medium hands and the bad hands.

Your checking range if you check 100% of hands

But if you only check 20% of your hands for example, you won't have all of these hands in your checking range with the same frequency. You might bet nut hands and bad hands more often than good hands and medium hands. Or you might bet nut and good hands more often than medium and bad hands.
Whatever might be the case, the distribution of hands when you check 20% of your range is virtually never the same as when you check 100% of your range.

Your checking range if you don't check 100% of hands (example)

And if that is the case, you will have to check-call those hands with a different frequency than when you check your entire range.

To give an extreme example: If you only end up with medium hands in your checking range, you will have to call each of those medium hands with a different frequency against a bet than if you still had all the nut, good and bad hands in your range. You'd need to call them more often. If you didn't, and still folded your medium hands as often as before, you would fold a lot more often in total when facing a bet. You would over-fold with your range and your opponent would make an instant profit.

Check-calls when you have all hands in your range (example)

Check-calls when you only have medium hands in your range (example)

(The above numbers are arbitrary and only serve as an example)

To put it simply, each hand will have to be played differently depending on which other hands are still in your range.

However... that rarely happens in practice. Most, if not all people that don't play high stakes (and also likely many of those guys), simply play their hands as they always would. "Against a bet, I do this, against a check, I do this."
So we can assume that after checking the flop, your average opponent in the BB will not adjust each hand's frequency according to the amount of their range that they checked. Instead, they'd rather play their hand based on what they learned was "GTO", so the equilibrium frequencies (and even that is the best case scenario).

/

The Maximum Exploit Strategy (MES)

To simulate that behaviour, we'll use the Maximum Exploit Strategy, or MES, approach. The MES is a powerful function in Piosolver that does not allow the BB to readjust any of their strategies, no matter how we deviate from ours. While calculating the maximum exploit strategy for us.
So in this approach, every hand would play the same frequencies as they would in the original GTO simulation, not adjusted to how differently their range looks as a result of their higher cbetting frequency. It also does not readjust to any exploit we're making as the BTN.
(I explain a lot more about how the MES works in my Mini-Course on Flop Exploits - but for now, just know that it allows one player, IP in this example, to find the best exploitative strategy against an opponent, OOP, who cannot readjust their strategy. Which allows IP to play a strategy that would be highly exploitable itself, but achieves a way higher EV than an unexploitable, balanced strategy, which a conventional node lock calculates, ever could.)

For most use cases, this is much closer to what actually happens in practice. Simply because our opponents never perfectly readjust. Not according to what their range looks like and not according to our exact strategy - because they just don't know it.
Still, I haven't seen anyone talk about this approach in the poker world. And I watch a lot of poker content. (If you have seen any content on this anywhere, please let me know as I'm genuinely interested.)
So by using the MES function, we can get a much more realistic, but also way more powerful exploitative strategy that brings us an even higher EV.

If we solve for the MES for us on the BTN after node locking the BB to cbet 72% of the time, we can see that our stabbing strategy looks much different. We stab less often, but also completely unbalanced: Instead of having nut, good, medium and bad hands in our betting range, we never bet anything that is top pair and better!

BTN vs. population's check according to MES

Instead, our betting range is extremely bluff heavy. We bet K high and worse 75% of the time and 2nd and 3rd pair 67% of the time! And again - we never bet any top pair, 2 pair or set.

The best thing about this approach is that it shows us how much EV is possible to achieve if our opponent does not readjust.
In the GTO simulation, the BTN's EV would be at 13.3bb.

The node locked simulation in which the BB cbets 72% instead of 42% increases our EV on the BTN to 13.4bb. So 0.1bb more, which is 10bb/100 in this spot - not too bad. (Yes, if we don't round numbers, it's 9bb/100)

But the MES strategy... Well, that one has the BTN's EV increase to more than 15bb!

So 1.6bb more than the conventional node lock approach, which is 160bb/100 (not a typo). Insane!

Now, I'm not suggesting to actually play this exact strategy in reality. Because at least opponents who you play against on a regular basis and whoare not braindead zombies will at some point realize what you're doing.
BUT. It does tell us a lot about how we can adjust our strategy in general. And how much more EV we could gain from it.

Because this strategy is the solver's way of speaking to us. Unfortunately, solvers are still not developped enough to speak English, so they have to use this cryptic form of showing us ranges. But if it could speak, it would tell us - well actually it would yell at us - something like this:
"THE BB IS OVER-FOLDING AGAINST A STAB! SO YOU CAN BLUFF WAY MORE OFTEN THAN IN GTO. WHICH ALSO MEANS YOU DON'T WANT TO VALUE BET AS OFTEN BECAUSE IT WOULD GO TO WASTE. SO SLOWPLAY MORE OFTEN AND BLUFF MORE OFTEN. WHAT YOU'LL GET IS A HELL LOT MORE EV. WHICH MEANS MORE MONEY BTW. THANK ME LATER."

Why exactly is the BB over-folding? Well, because of the effect we described earlier. If they don't adjust their check-calling frequency based on their checking range, and they check more bad hands to begin with, they will check-call less often overall. Which makes bluffing instantly profitable and the best exploit.

(If you look closely at the conventional node lock, it's not only the check-calls that would have to be adjusted, but also the check-raises. To compensate for a lower check-calling frequency, another adjustment would be to check-raise more often, because that also deincentivizes the BTN to stab more often. I only focused on the check-calls for simplicity and to make the concept clear.)

BB's check-folds after cbetting 72% but not readjusting (MES)

In reality, you might want to mix the 2 approaches based on your opponent. In general, against population that cbets more often on Axx flops after 3betting from the BB, you can stab more often against their checks and bet more bluff-heavy.
And the less you think your opponent is aware of your adjustment, the more bluff-heavy you can stab. If you think they are somewhat aware, just mix in a few value hands as well so it doesn't become too obvious.

That's it for today - I hope you got a lot out of this and you can apply this exploit at your tables.

If you liked it and want to get more exploits like this to improve your winrate, GTO2Exploit will be exactly for you:

GTO2Exploit

In this membership, you receive ongoing exploits against population in text form, as you did in this email. By giving you data-driven breakdowns for all kinds of positions and spots, I want you to become a master at exploiting your opponents.
Click here for more info:

GTO2Exploit

Thanks a lot for reading and I'll see you soon!

Victor
2 Card Confidence