GTO? More like GT-NO!
Today we’ll be debunking 4 big myths about game theory optimal poker strategy, or “GTO,” as it’s now commonly called.
With insights from Ryan Fee and acclaimed poker writer Matthew Janda (from their new GTO module in the Upswing Lab), we’re determined to clear up these misconceptions once and for all.
A GTO strategy is a strategy that cannot be beaten in the long run, regardless of how our opponents play.
The best players in the world currently use a GTO strategy
The GTO strategy for No Limit Hold’em exists, but is not yet known by any human, or even computer. The game is just too complicated, and hasn’t been solved. Less complex poker games, such as heads-up Limit Hold’em, have been solved (or at least essential solved).
GTO has become a buzzword synonymous with anything from “best possible strategy” to “good,” depending on the user. You might hear someone describe their steak dinner as GTO, or even their dog.
When you hear people talk about a GTO poker strategy, they probably mean to say a game theory-based strategy. Even veteran pros regularly get this wrong.
That said, we do understand some GTO concepts for some of the simpler parts of the game, and we recommend using them to construct your baseline strategy (more on this later).
(Note: Want to turn your poker hobby into a profitable side job? Start crushing your competition with expert strategies taught to you by world-class players when you join the Upswing Lab. Learn more now!)
A GTO strategy would always take the most profitable line
Wrong. A GTO strategy is the strategy that maximally exploits an omniscient opponent. In other words, it’s the strategy that wins the most even if your opponent knows it perfectly.
A GTO strategy does not take into account the type of strategy an opponent is employing and thus is not always the most profitable strategy. For instance, a GTO strategy bluffs on the river against a calling station just as often as it bluffs against a nit.
Again, since poker is not a solved game, no player employs a perfectly optimal and balanced strategy. All players will be exploitable in some way, and some will be more exploitable than others. The most profitable line to take against opponents will involve maximally exploiting their imbalances through a specific counter-strategy.
However, exploitable plays have their own drawbacks. The two most significant are:
1. Sometimes your exploitative play will be wrong.
By definition, exploitable plays are a deviation from a theoretically optimal strategy. This is why it’s important to make exploitable plays based on reliable information–not assumptions. If you make exploitative plays based on assumptions, you will sometimes make massively -EV plays as a result of shaky information.
2. Whenever you make an exploitative play, you leave yourself vulnerable to being exploited.
Consider an example. We are playing online against a regular, and have a huge database of hands which shows that they over-fold from the SB versus 3-bets. Against this player, we are incentivized to 3-bet from the BB aggressively as we know the SB will fold often.
However, by making this adjustment we ourselves become unbalanced and open to being exploited. With any luck, the opponent we are attempting to exploit won’t notice that we are 3-betting a wide range, which could be punished by light 4-bets.
You must be a master of math to use a GTO-influenced strategy
There is obviously math involved in playing optimal poker, but you don’t need to be a math expert to get your head around it.
Almost all of the calculations you need to know involve simple operations—there’s no need for algebra or linear equations and there’s no scary notation. The most commonly used formulas are available online. For example, you can read about two of the most important pieces to a game theory-based strategy here: calculating pot odds and minimum defense frequencies.
For example, the formula for calculating the optimal ratio of value bets/bluffs that you need based on your bet size is simple and can quickly be memorized. Let’s say we are heads-up and bet $100 into a $200 pot on the river, giving our opponent 3 to 1 odds on a call. In this instance, our betting range should consist of 75% value hands and 25% bluffs to make their bluff-catchers indifferent to calling or folding (the best they could do is break-even).
This ratio changes depending on our bet size. Here’s a table with a few of the most common bet sizes to help you remember them:
Size of Bet
2 bluffs for every 3 value bets
1 bluff for every 2 value bets
3 bluffs for every 7 value bets
1 bluff for every 3 value bets
As the table shows, the smaller we bet, the more value hands we need to have, and the larger we bet, the more we should bluff.
There’s no point to learning GTO since an exploitable strategy has a higher profit ceiling
While we’ve already established that an effective exploitative style will yield more expected value (EV) than a GTO strategy, this doesn’t mean you should disregard GTO altogether.
Playing in a well-balanced and optimal way is best when you know little or no information about your opponents. It often takes thousands of hands before you can confidently spot potential exploits in others, so it’s wise to employ a baseline strategy that protects you from getting exploited while you can gather this information. When you do pick up on potential exploits, then by all means devise a counter-strategy and feel free to deviate from your starting strategy.
Additionally, a sound understanding of GTO concepts will help you identify potential exploits in your opponents’ games. If you are well-versed in what an optimal strategy should look like, then you will be able to clearly spot when a player is deviating from it, and know how to punish them accordingly.
Time to GT-Go
GTO strategy is a complex topic that deserves more attention than we’ve given it here today, but hopefully we’ve cleared up some of the more common confusions.
Drop any comments, queries, or suggestions for future articles in the section below, and good luck on the felt!
Read more from Upswing Poker: