Aggression is one of the most important parts in poker. When you're aggressive, you can win the pot without a good hand. And when you're known to be aggressive, you will get paid off more often when you have a good hand.
But how much aggression can you apply profitably? And. how much is too much?
Linus vs Kevin Paque
In this hand, one of the most aggressive players in the game, LLinusLLove, opens suited from the hijack and gets called by Kevin Paque in the big blind. On the rainbow flop, Linus c-bets full pot.
A big bet size is indeed preferred on this structure and formation, but as is often the case, the solver uses all sizes to some extent.
If you want to make your life easier in-game, simplifying to only using one bet size is a good option. And since all options are used in the solver's mixed strategy, all sizes would be fine options if only using one. Even though they all have different benefits, for small stakes, a small size is probably the best option. It's easy to play because it could be simplified even more by using a range bet and not lose a lot of EV by doing so.
And even better, the population at lower stakes doesn't check-raise often enough against the small size on average. So betting small and often is even more profitable in practice than in GTO. At these stakes, that mistake doesn't happen as much. Instead, betting this big has the benefit that it makes Kevin's life harder than a small bet. Because against the small bet, he could always continue all third pairs and better, and all gutshots—which, apart from finding the right check-raising frequency, is relatively an easy strategy.
Against a pot-size bet, this looks a lot different.
He would have to mix calling and folding with every hand class that is an underpair or worse, including gutshots. Those hands are indifferent in their action, which in theory means calling and folding—or even raising—have the same EV. However, in practice, that means it's just very hard to get the mixing frequencies even remotely right, and not continue too often or too rarely. So from Linus' point of view, betting this big likely forces a bigger mistake from his opponent than betting small.
On the turn ( ), Linus goes for a bet of around 75% of the pot.
The solver would also use that size or an overbet with our range.
Sets and straights would always bet, and the best two-pair and top pair would as well. The best bluffs to bet would be those that have the best equity, like good straight and/or flush draws, or a straight draw and a pair.
Having a pair is valuable because not only do we have additional outs against the big blind’s calling range, we also block their best hands in two-pair and sets. And even though a pair is ahead of some of our opponent's hands, they mostly have 10 or more outs or can easily bluff us on the river. So winning at showdown doesn't happen often enough to make checking the better option. Additionally, betting would even fold out a lot of better hands, like third and second pair, and even some of the worst top pairs.
Linus continues with his aggression on the river ( ), and he is confirmed by GTO.
In fact, the solver would bet even bigger and jam here.
In this line, the big blind should be very capped in the amount of nuts he has with , especially after a pot-size c-bet and another big turn barrel. We, however, can have a lot of them when barreling through. Other nut-like hands— , , or other sets—likely check-raised on a previous street, so those combos are very rare too.
is a hand that still blocks two-pair and sets, so using it as a bluff is a good choice. Why does Linus bet 150% instead of an all-in size? Well, of course, I can't read his mind. But if we look at the big blind's GTO responses against each size, we will find a similar pattern as on the flop.
Against the jam, finding the correct GTO frequencies to call with isn't easy, but it's also not insanely difficult:
- Top pairs fold almost always, which is not difficult to imagine most players would get right here against the jam.
- Sets and straights always call—also very understandable.
- Only two-pair are hard to get completely right, as they call about one-third of the time, and the combos are heavily dependent on the hijack's jamming range.
However, if we look at the GTO play against the 150% pot size, we'll see that not only do top pairs have to call a bit more often and two-pair are allowed to call two-thirds of the time, we also see that the big blind would have to implement a complete additional strategy for check-raising. Of course, straights check-raise all the time, but then sets sometimes raise and mostly call. And he would have to construct a bluff check-raising range—not only with the correct frequency, but also the correct hands—which is undoubtedly more difficult than just having to think about your calling range.
This is aggression according to GTO, with maybe a little exploitative spin on it.
And it worked: Kevin folded.
Linus vs PR0DIGY
A more exploitative approach of aggression can be seen in this next hand.
Linus opens the button and PR0DIGY calls in the big blind. On rainbow, Linus checks behind, which is a valid play according to theory, even if it's rare with this exact hand.
And while simplifying to range betting small might again be a good option at lower stakes, it does take away a huge chance to exploit the average regular in the checking line. Now, PR0DIGY is everything but average, and we'll see later how that shows. But on mistakes, checking behind instead of betting opens up some great opportunities to punish some of the population's big mistakes and make more money.
On the turn , Owen checked again. When the big blind checks in this spot, data clearly shows that they overfold against the delayed c-bet, which makes one thing especially profitable: bluffing.
Finding spots where the average opponent folds or calls more often than they should is the easiest way to multiply your win rate. If you know where you can bluff a lot and in which spot value betting is more profitable, the EV will naturally come your way.
For example, knowing the population overfolds by seven percentage points doesn't mean we can bluff every single hand in our range. So, how much would our opponent need to overfold to make a bluff bet? In this example—well, if we note an overfold of seven percentage points for this exact board—we'll see that we as the button can start bluffing a whole lot of hands.
For example, the lower aces— to —instantly go from never bluffing in GTO to always bluffing as an exploit. However, is still never betting and always checking. That's because still has showdown value against a lot of A-x hands that the big blind would check down as well. So it's one of the last bluffs to add to our betting range.
Only when we increase the big blind's overfold to 12 percentage points does bluff betting become similarly profitable as checking.
Twelve percentage points of an overfold does, however, mean that PR0DIGY would need to fold quite a lot of hands. It would mean every second pair and better would never fold, but third and fourth pair would already fold 70% of the time against this bet right away.
Owen called on the river. And we can only estimate whether this is the case in this high-stakes battle. But if I'd have to guess, I'd probably say PR0DIGY would not overfold by that much.
On the river ( ), we see more aggression. Linus overbet $2,500 to $1,650.
But for the solver, is again just never bet. Checking back is simply the clearly better option, as would be any here.
The only way A♠10♠ could become a profitable bet is if the big blind overfolded on this river again, which—for what it's worth—is not the case for a midstakes population in this line. Or, if Linus thought PR0DIGY might call too often on the turn and then heavily overfold the river.
But if he doesn't deviate heavily from GTO, this is an example of too much aggression, which—instead of exploiting your opponent and increasing your win rate—can open yourself up to get re-exploited and lose you a lot of money.
In GTO, almost no , , or A-x would ever check-call on the river. But even just having Linus bet all 25% of the time here—which can't be too far off, seeing he's bluffing —he could print money check-calling down any hand. You only really run into that risk when you're playing an opponent who is very aware and capable of quickly adjusting to your deviations.
However, unfortunately for Linus, that is the case.
