I see yet another GTO conversation going down on Twitter, so here it is, THE definitive thread on how to not sound like an idiot when you are talking about Game Theory as it applies to poker.
1) If someone is playing GTO, it means they are playing 100% perfect, unbeatable poker. Not they are playing well and so they play GTO, it means there is 0 mistakes in their strategy and they are completely unbeatable by any counter strategy.
2) No one in the world plays GTO currently. All players are essentially on a sliding scale in terms of how close to GTO they are playing. The bets bots are becoming much closer to that line. The best humans are getting closer as well, but no one is there until the game is solved.
3) GTO and the result of a solver are different things. Solvers have a constraint, they can only use a certain number of bet sizes before the required amount of memory is too large to produce results in a meaningful time frame (or it simply takes too much computing power)
Because of this, solvers solve for an abridged version of the game where players must choose only a few bet size options.

A true GTO player would destroy a solver in an actual game, as it is not going to be limited in number of bet sizes in any spot (Again, doesnt exist yet)
4) Solvers/bots can start to approximate gto based on their programming/power. Solvers will give you a perfect solution given its constraints, but that is the caveat, those constraints prevent it from playing GTO.

However, you can learn very valuable concepts from solvers.
5) You can also calculate the value of comparable strategies by solving for them and cross checking their EV. This will give you an idea of the extra value you get from adding additional bet sizes. (Again, a true gto strategy would probably have hundreds of various sizes)
6) Humans cannot easily implement solver strategies, because of the massive number of depth in the solutions. There are 1326 different combos of hands, all of which will have various frequencies in actions and they each have individual properties that change frequency
7) You can use these tools, as well as learning from people who have used these tools, to learn valuable heurestics and create size/strategies that you can actually execute. IE you take the things that you learn and you create a strategy that you can balance correctly.
8) It takes just an incredible amount of time to learn how to do the above process well. These people that think you can look at a solver once and/or use some app for a month and be playing "GTO" or really anything even resembling "GTO" have no idea what GTO even means.
9) Finally, stop saying GTO. It makes you look like a noob.
You can follow @DougPolkVids.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: