Understanding The Wacky World Of GTO Poker

If you take a lap around a cardroom these days, you will undoubtedly hear the term “GTO.” For some beginners, and even some veterans, this can be a little intimidating.

GTO, or Game Theory Optimal, is simply the way to play any game in which it is impossible for your opponent to have an edge verses you.

The way in which GTO works in poker is to build bet sizes and strategies for a given situation in a way that will unexploitably capture the most expected value. Which leads us to “Solvers.”

The first public GTO solvers, PioSolver and MonkerSolver, came out in 2015 and 2017. How poker solvers work is they use Nash equilibrium mathematics to “hard solve” a situation, based on the user’s input for what bet sizes and overall options they want each opponent to have for their ranges.

So, if I wanted to study a specific situation with a GTO calculator, I’d have to put in each player’s range pre-flop, then put for each player what bet sizes I wanted to give them, and what raise sizes I wanted to give them, as well as the max number of raises for a situation. This is what is known as building out the game tree.

Once I have those parameters set, I then have to set the calculator to run. To have the solver run for a typical modern game tree I’d need a computer with at least 64GB of RAM, as well as a good processor.

Your goal when running the solve is to allow it to solve to a “high degree of accuracy.” What this means is, as the solver runs it steadily builds closer and closer to a completely unexploitable solve, generally speaking your goal is to get that solve down to where it is 99.5% or greater close to GTO.

The more complex the situation, meaning the more complex the ranges, and the more options given for each player, the longer it will take in order for it to solve to that degree of accuracy and the more RAM and processing power you will need. A typical solve these days for a single flop will take anywhere from 20 minutes to over an hour to get to a proper solve accuracy.

Building a proper tree takes a lot of expertise in the game, as well as an understanding of how to use the actual solvers, a powerful computer, and the willingness to wait for the solve to fully run. These are a lot of barriers for a player to learn how GTO approaches a specific situation in order to play at a high level.

However… Now we enter the world of GTO Study tools.

The rate of change in the game, and what we know about “GTO” has shifted drastically over the last few years. This has happened for a handful of different reasons.

First of all, more GTO calculators entered the market and the programs that have been around the longest received upgrades that made them work significantly faster.

Secondly, the use of Artificial Intelligence (AI) has allowed these programs to help build solves for situations as well as solve those situations more accurately.

Finally, there have been a number of GTO study tools that have entered the market. What these tools have done is pre-build GTO solvers with easy-to-use interfaces for members to be able to look at solves quickly.

The study tool that I helped build, RangeTrainerPro, has a database of well more than 200 situations solved to an accuracy of 99.5%, with a tree that often requires 128 GB to 256 GB of RAM to run. It also requires more than a Petabyte of hard drive space to store the data.

It would take me more than an hour to set up a tree and look up one scenario on my own, but now I can simply go to the site, log-in from any device, and within 30 seconds I have the answer.

There are a lot of GTO study tools out now that have a wide range of prices, user interfaces, situations solved, solve accuracies, and other features. If you are curious about what tool might be best for you, I’d think about what games you want to study, how easy you want the user interface to be, and what type of companies you want to give your money to.

It’s fairly easy to find many different companies in the space for each format of poker, and while I personally prefer certain tools, my recommendations of which you should use would certainly be biased. That being said, I very strongly recommend finding one that works easily for you, to make studying GTO as comfortable as possible.
Why Study GTO?

You might say that GTO is only for the top pros. But the way in which to best make money in a given hand, whether the most exploitable vs. a recreational player, or the closest to theory vs. an elite pro, both require the same thing: A good understanding of what GTO is for that situation.

If you are playing vs a recreational player, how do you know what type of mistake they are making, if you don’t know what GTO even does in a given spot?

The better you understand any situation, the easier it will be to understand how they are making errors, and thus be able to figure out the best possible way to exploit the mistake they are making in order to achieve the most profitability.

If you are brand new to GTO I strongly recommend to first join a training site. This will help you understand the general ideas that go into theory decision making. Once you have that basis built, I would then focus extremely heavily on pre-flop play. Once you have that down pretty well, then you can start jumping into the incredibly complex world of post-flop theory.

You should be spending about 30% to 50% of your poker time studying, compared to playing. Even if your game improves and you feel like studying is less effective, then I’d still study 15% to 25% of the time.

Studying can be reviewing hands and discussing them with friends, getting coaching, watching training videos, using GTO study tools to train hands or to review specific situations you have played.

You also need to study every single week to stay sharp. View it the same way you do spending time in the gym.

The more serious you are about playing profitable poker, the more important it becomes to follow the above suggestions and to immerse yourself fully in the wacky world of GTO Poker.

Thank you for reading, stay safe and healthy, and best of luck at the tables! ♠

Ryan Laplante is the co-founder of LearnProPoker and the RangeTrainerPro GTO study tool. The 14-year veteran of the game has more than $8 million in career tournament earnings. The Minnesota native has wins at the Poker Masters and WSOP Circuit to go along with 14 World Series of Poker final tables and a bracelet. You can find him on Instagram RealProtential, Twitter/X @ProtentialMN, or send an email to pokerprotentialgmail.com.