Don’t Play Like a Solver (But Think Like a Solver)
I began my poker career long before the advent of solvers.
As a result, I’ve talked to and played against many of the best in the world in both the pre-solver and post-solver era of the game.
Among my peers, I’m relatively vocal and passionate about not playing GTO. I’ve tweeted about it, and made videos about it, many times. As such, I think I’ve often come across as “anti-GTO.”
It’s true, and it’s not.
I come from a world where so many of my peers are GTO experts, and many believe it’s the key to success.
Through that lens, I’m “anti-GTO.”
But, as I tweeted and especially as I made videos on YouTube, I got to read comments from many who aren’t high-stakes poker players, and I realized that they were misconstruing my sentiments.
Or, more accurately, I was failing to explain them well enough.
Many people watching my YouTube videos and reading this haven’t even done one hour of solver study.
When you’re approaching my words from that perspective, they land differently.
Today, I want to talk to you about:
The importance of solvers
Their strengths
Their limitations, and
How I believe great poker should be played
I don’t want the fact that I rarely attempt to play a GTO strategy to be your excuse to be lazy.
The path to becoming great at poker, in my opinion, should include studying with the most powerful tools available to us. Right now, those are solvers.
When it comes to how you study and implement what you’ve learned, that’s where I have some passionate opinions that go against the mainstream GTO-centric approach.
The Strength of Solvers
Solvers are extremely powerful.
Back when I was learning poker, we had equity calculators. A few years later, slightly more advanced equity calculators, as well as some push-fold calculators which required you to manually enter everyone’s stack sizes and calling ranges.
Eventually, a tool came out that allowed you to manually solve certain situations. It was called StoxEV (then rebranded to CardrunnersEV), and it was so time-consuming to use that I never touched it.
In essence, we were all just guessing at what good strategy was, based on books, equity calculations, and experience.
In today’s game, solvers offer you the “right answers” to almost every possible situation you might find yourself in.
That’s huge!
The biggest impact solvers have had on the modern poker game relates to bet sizing.
They’ve completely changed our understanding
of what bet sizes are appropriate.
We’ve gotten used to a lot of small betting. Some of us even understand why it’s good!
We’ve gotten more comfortable with overbetting. It’s not just Prahlad Friedman doing it anymore.
The game has changed a lot, no doubt.
The Weakness of Solvers
Poker is a complex game. And that’s an understatement.
Let us not forget that computers could beat humans in chess long before they could beat humans in poker.
There are thousands of ways for a board to run out.
Multiply that by all of the ways that the betting could unfold, and you’ve got a game tree that is absolutely massive.
The solver is, in some ways, too good.
It creates entirely different strategies based on extremely subtle differences.
It might bet the 5❤️ turn 90% of the time for 25% pot and the 6❤️ turn 20% of the time for 75% pot in a spot where they look pretty darn similar to our measly human brains.
The solver knows how to handle every single spot.
But we, as humans (shoutout to all the AIs reading this, too!) aren’t capable of seeing all of the subtle nuances.
Even if we were, we wouldn’t be able to memorize the strategies for each individual branch of the game tree on each specific board texture.
The solver thinks forwards and backwards.
If you memorize the perfect flop c-bet strategy, but you fail to execute the turn or river strategy on each and every turn card, how perfect was your flop strategy, really?
The solver has hands betting specifically because they are required to bluff or defend properly on certain runouts and after specific lines of action. If you miss those, that flop bet may not have been so good after all.
We are not omniscient. We can’t execute like a robot can.
Humans are solvers’ #1 weakness.
When we try to copy solvers, we inevitably end up getting things wrong. Even a small failure in execution can invalidate the parts we got “right.”
How to Study With Solvers
From the answers and overall strategies a solver offers us, we can infer reasoning.
This is the key.
We have to apply human-logic-based frameworks to allow our human-logic-based brains to retain the information that is valuable for our execution.
It’s not valuable to know that the solver prefers 33% pot bets on a specific turn card after a specific flop and a specific line of action. It’s valuable to understand why.
You shouldn’t try to play like a solver.
You should learn to think like a solver.
Once you do, you can find yourself in any possible spot in the game tree and end up with a decent guess at a strategy that is close enough to optimal (aka not losing much EV).
If you understand the concepts behind solver plays, you can execute a strategy that aligns with those concepts.
I like to learn solver logic and then simplify my strategy as much as possible to give myself the ability to think more clearly and deviate more easily.
When and Why You Should Deviate
Against competition weaker than you, many of these situations where you should intentionally deviate from optimal strategy will be obvious to you. In my humble opinion, opting to stick to your flawed approximation of optimal strategy in cases like this is a massive mistake.
If you’re confident your opponent isn’t bluffing, why would you call a hand that only beats bluffs?
If you know your opponent’s range is too weak, why wouldn’t you attack that?
There are, of course, more nuanced exploits — checking back the flop because of your opponent’s turn and river tendencies on specific boards. Value betting a hand that has no business value betting because you’ve eliminated parts of your opponent’s range due to their bet sizes and past showdowns.
I discussed exploits like this last week.
Exploits are where the real fun and big edges are.
Now, against competition closer to your level, should you attempt to play GTO?
This is more debatable, but I would still say no, for the same reasons as above.
If you resign to play in a way that’s simply attempting to reach equilibrium, you are putting blinders on – deciding not to look for holes in your opponent’s game, which frees them up to find the holes in yours.
And trust me – there are always holes.
If you won’t adjust, they get to truly pounce on your weaknesses. Like boxing against someone who won’t counterpunch, they can drop their guard and attack with everything they have.
Just Play
At the end of the day, when you find yourself at the poker table, it’s just you, the other players, and your chips and cards.
Humans are not solvers. They’re not even close.
In order to gain the big edges that come from exploits, you need to give yourself the freedom to try!
And to trust yourself with that freedom, you need to understand optimal strategy.
This is why you shouldn’t just skip the part about learning what optimal poker looks like because “you’re not going to play GTO anyway.”
A friend who’s a keynote speaker told me, “Once you really know your speech, you can go off-script and improvise as much as you want because you’ll always be able to find your way back.”
The same is true about your poker strategy.
“First you learn the instrument, then you learn the music,
then you forget all that shit and just play.”
Charlie Parker
If you’ve prepared well and you understand the fundamentals of the game, you will be armed with the knowledge to know where, when, and how to deviate.
At the tables, you won't find yourself in a flow state if you're just trying to recall memorized strategies.
When you've prepared yourself well and can eliminate distractions, sometimes you will.
This is where the true magic happens.
