What is GTO strategy?
A Game Theory Optimal (abbreviated as GTO) strategy has the highest expected value, versus the complete gamut of strategies taken as a whole available to opponents. Importantly, a GTO strategy does NOT necessarily have the highest expected value versus any particular given strategy that an opponent is choosing to use. When all players in the game are playing a GTO strategy, we can say the game is at equilibrium (that is to say we have reached a Nash Equilibrium). When all players have reached a Nash Equilibrium, one player cannot unilaterally change their strategy to the detriment of the any other.
A GTO strategy is unexploitable, however not all unexploitable strategies are GTO (as there might be a different unexploitable strategy for different possible bet sizes, the best of all these is the GTO strategy).
What are poker solvers?
Just like you would use an equity calculator, to explore equities of hands/ranges, you can use a solver to think about GTO. There are various companies who produce this type of solver software, and we won’t recommend any one over another. Most have some free limited functionality, so we suggest you try this out before committing on purchasing any of these (as they are certainly not cheap).
Solvers make Nash equilibrium calculations, to find unexploitable strategies. If there are multiple unexploitable strategies, it compares these to find the best one. Solvers work by iteration. For a given set of assumptions, its keeps constantly changing each players strategy, until no player has any incentive to deviate (i.e. we have reached equilibrium).
You can also use a solver to lock one player’s strategy, and find the way to best exploit that particular strategy.
Why no one is playing true GTO in NLH
They can’t be, as no one knows what this is for no limit Texas Holdem. Because of complexity of the game tree in No Limit Holdem solvers need to work on assumptions which simplify the game tree (for example what the preflop/postflop ranges are, what the available bet sizes are). So a solver only finds the best unexploitable strategy, based on the assumptions it is using. So, infact the solver has only found the GTO strategy for the starting assumptions we give it. Solving for the GTO in deep stacked situations is especially difficult, as the possible range of bet sizes is massive.
Even if one day a computer program, such a solver, did solve No Limit Holdem, and gave you the perfect set of GTO strategies – i.e. you would know exactly what to do at every decision point (regardless of bet sizing etc.), it would be impossible for you to memorize. Live poker could still exist, as long as such tools were banned at the table.
If a computer program existed, where you could enter the decision you face in a hand, and get an answer instantly as to what the GTO play would be, that would be the end of online NLH.
So, why study GTO strategy?
Despite all this if you want to become truly good at poker, we suggest you study GTO, even if you plan on playing an exploitative game.
For the time being, solvers are useful to finding the GTO solution, for a given a set of assumptions which are often limiting. The better you get at providing the solver with accurate assumptions, the closer to the full GTO solution you will get.
Regular use of solvers will help you build up an idea of what a GTO solution might look like, which you can then use in your games to play a quasi-GTO strategy when needed.
By being able to think in terms of GTO, you will be able to:
- construct ranges that are balanced, for all lines
- make your opponent not care whether they call or fold etc.
Learning to approximate GTO, is useful even if you only plan on playing exploitable poker. Once you can approximate a GTO solution for a given situation, you can see how your exploitable opponents are deviating. This will help you find the best way to beat them. Equally, you can see how you are deviating from the GTO solution, and spot ways your opponents may be looking to exploit you so you can anticipate them.
When to play a quasi-GTO strategy
Using solvers, you will start to get an idea of what a quasi-GTO strategy looks like.
When you have absolutely no information (and don’t think player pool information, or live observations) can help you find an exploitable strategy on your opponents, you should look to play a quasi-GTO strategy as a base, whilst always being on the lookout for opportunities to deviate if their is anything you can exploit in your opponent’s play. This prevents tough opponents exploiting you.
Whenever you are playing a quasi-GTO strategy, it is important to always be on the lookout for any weakness in your opponent’s strategy that you can take advantage of by playing in an exploitative way.
In a raked game such as poker, if one day everyone at your table was able to play a GTO strategy no one would win in the long run except for the house. That is not to say, you should not play a GTO strategy if you believed all your opponents were – infact you have to play a GTO strategy, if you want the highest expected value. You should probably quit the game and find another one, if that was the case!
When to use an exploitative strategy
If you have the opportunity to exploit your opponent you should take it (provided you are ready to react if they exploit you back).
Remember your goal in poker is to make the highest EV decisions versus the strategy any opponent is employing – this highest EV line is not necessarily the GTO strategy.
If you are playing low stakes live cash games, or microstakes online, you should almost always be looking to use exploitable strategies from the get go. At low stakes live cash games (or the lowest stakes online) your unstudied opponents are likely to be making a ton of errors, and studied players in these games are likely to be making a lot of errors too. You need to use observations about their tendencies, as well as their frequencies, to come up with the best exploits.
Warning about exploitative play
As soon as you deviate into doing anything exploitative, you face a problem – your strategy is no longer unexploitable. This means your opponents can exploit you too, and they may be able to do this more effectively than you can to them (depending on their skill levels). They might even be exploiting you totally accidentally! Of course, you can re-exploit them (and they can exploit your re-exploit, and so it goes on).