The Most Important Skill In Poker

The concept I want to teach you in today's post can be visualized through the hand below.

Let's review it together.

Action folds to my opponent who opens from the SB to 3x, and I defend my big blind with 8 ♣ 7 ♠.

We see a flop of Q ♠ T ♣ 8 ♦, and my opponent checks to me.

This is in fact a type of board texture where he should be careful with his c-bet strategy - at least in theory. High-Low-Low boards should be checked around half of the time at equilibrium, as you can see in the print below.

Faced with this check, I have a very easy check behind with 3rd pair. I check and we see a turn.

The turn is the 7 ♣. What a nice card! If it wasn't for this card, we wouldn't be here talking about this hand.

We pick up 2pair and our opponent checks to us. The decision at this point is only one - how much we are going to bet, because betting is obvious.

The way I like to think about bet sizing is by first picturing how strong my range is relative to my opponent's. Typically, in spots where we have nut advantage, we want to overbet. Because we have a lot of nuts that our opponent doesn't have, he will be forced to calldown with bluffcatchers even against big sizes, if he doesn't want to get run over. That being said, if the distribution of nutted hands is similar, our opponent has enough 'auto calldowns' in his range that makes the need to calldown with bluffcatchers much lower. At the same time, if he can also have the nuts, we can't afford to overbet with our nuts and then bet medium with the other hands that want to get value, like top pairs. So, in scenarios of even nutted distribution, a medium size is the way to go.

In our hand, the 7 completes 96s, but both of us have that in our ranges. 77 also improves to a set, but both of us have it. The 2 pairs we also both have at similar frequency - Q7, T7 and 87 are all in both of our ranges. What this quick analysis reveals is that this card doesn't hit me particularly better than my opponent, and he still has more sets than me. For this reason, I completely discard an overbet size (at least for now - later you'll see me picking an overbet size that will definitely surprise you).

We bet 3/4 pot and our opponent calls. The river is dreamy.

The river is the blankest of blanks, the 2♦. This is the best runout we can hope for, and now our job is again to figure out how to get the most value out of our hand. Before continuing, ask yourself and write down: how much would you actually bet there with 87? That answer will tell you a lot about yourself.

At this point, we can split the population of poker players into 3 groups.

The first group consists of the weak regulars and lowstakes players. These players apparently have only one sizing in their poker clients: 75% pot.

Holy shit, I've been coaching pros and semi-pros since 2018 and the amount of missed value I've seen from people who are seemingly incapable of betting another bet sizing is insane.

I guess we can blame risk aversion and lack of knowledge as the main drivers of this behaviour. Some people know they could bet bigger, but the thought of doing so scares them. What if the pot gets too big? The idea of playing for a huge pot is uncomfortable, unless they have the stone cold nuts. 3/4 is good enough - it's more than half pot, and it is safe.

Then there is lack of knowledge, which can be split into 2 parts: 1st, the people who simply have never seen anyone overbetting in their lives, so betting bigger than the pot is not even part of their world - it's a completely foreign concept. Those are a minority. Then 2nd, we have the people who subscribe to the idea that 'people will fold too much to a bigger bet, therefore I must size down'. This idea gets thrown around on the internet very often, and a lot of players are super confident about it. For example, this is one of the comments I got in one of my recent shorts on YouTube, where I showed a hand I shoved all in for 2x pot with my value hand:

This viewer here demonstrates why this idea is so sticky in people's heads: it's hard to understand that even if your bet gets called less often, it can still make more money. People hyperfocus on the frequency they are getting called (which absolutely decreases as you increase bet sizing), but they forget about the reward they get. The viewer wants to bet half pot because that gets more calls than an overbet that gets called only 'once in a blue moon'.

The thing is that once in a blue moon can be just what you need, if the bet is big enough. And we'll see that next.

The second group of people are the people who got past the mistakes from the previous group, most likely by studying theory. People in this group know that you're supposed to overbet when your range contains more nutted hands than your opponent's range. They know that the game is called No Limit Hold'em for a reason.

I believe this group of players would bet the sizing the solver recommends here:

On this runout, overbets are a thing in theory, and this is because the previous actions - flop check-check + turn check-bet-call - lead to a situation where the BB (me, in the case of the hand) can show up with a bunch of 2pairs by the river, whereas OOP has a very limited amount of those hands in its range (those would have fast played before). This matchup leads to the (pretty cool) equity distribution graph on the river:

Notice how the SB almost doesn't have air in its range (which makes sense, from check-calling turn), and how 80% of its range are middling strength hands. At the same time, BB has 40% of its range being complete trash, about 25% of its range with 80-90% equity, and 20% with more than 90% equity. This is why the BB splits its range between a 3/4 size and a 1.5x size (see print above) - 2 different regions of value betting hands that want to get different amounts of value, with 87 belonging to the obvious overbet group.

So now you might be wondering... 

"Saulo mentioned 3 groups of people, the first bets 75%, the second bets 150%. What about the 3rd? How much did he bet there?"

Yeah, there is one group left.

This group is composed of players who understood that this game we love is all about playing the opponent. While having a solid theoretical foundation is important, it's also important not to limit yourself within the confines of theory. Money and edge are maximized when we understand and exploit our opponents.

I decided to go all-in for 662% of the pot.

And here is the key concept I want you to take away from this post: if you want to make the most money in poker, your #1 job, the most important thing you should be doing at all times, is anticipating your opponents' reactions.

I'm not trying to replicate the solver here. And I also wouldn't bet this sizing against every opponent. But I did it because I knew that this particular opponent was not only capable of bluffcatching against a massive overbet, but also to perceive me as overbluffing with this massive overbet.

To anticipate your opponents' reactions, you must know them and know yourself. You must understand their psychology and understand how they perceive you as a player.

I know my image is one of a very aggressive player, capable of running huge bluffs. I know that my range contains lots and lots of missed draws and random trash. I also know that my opponent knows that his range looks capped. This is the perfect combo.

What is a player who is capable of calling down going to do when they see a very aggressive player taking a line that doesn't seem to make sense, applying maximum pressure against what everyone knows is a capped range, while having tons of potential bluff candidates?

It doesn't take a genius to figure out. They are going to call. And they do it because in their minds they are about to crush me.

The thing is that, I can anticipate all of that rationalization. By knowing myself and knowing my opponent, I can precisely estimate how he is going to react against my actions, and if I can predict how he is going to react against my actions, I can crush him.

My opponent SNAP calls my 7x pot all-in bet with top pair. No joke, he called within 2 seconds.

I won 99 big blinds in a spot where most people would win 10 big blinds (3/4 pot size) or 22 big blinds (1.5x size). And I was able to do it because I anticipated my opponent's reaction. I knew that a player like him, on a board like this, against a player like me, would be eager to make the call. And so it happened.

I do this all the time, and I have posted several hands about it. Here is the 2x pot jam criticized by our friend above. And here is another massive overbet all-in that got called by a mediocre hand - exactly as I wanted.

While studying with solvers and exploiting population tendencies through mass data are vital parts of the modern poker era, none of these paths would have led me to win my opponent's whole stack. Solver bets 1.5x pot, and MDA will say population overfolds too much vs all-in for this to be a thing for value. In the modern era of the game, psychology and metagame are still extremely important. It's only through the understanding of your opponent's psychology and specific game dynamics that you can develop huge exploits like this. Don't neglect this part of the game, otherwise you'll certainly cap your potential. Take this concept seriously, and you'll see big winrates are far more achievable than you imagined.

Thank you for reading, my friend. See you next Monday!

Saulo