GTO stands for game theory optimal. A GTO poker strategy is one your opponent cannot beat by changing their own strategy: whatever adjustment they make, they gain nothing against you. In game theory terms it is a Nash equilibrium: no player can improve by deviating alone.
That one sentence explains almost everything about how solvers play, and most of the confusion about GTO comes from forgetting it.
Unexploitable is not the same as maximally profitable
A GTO strategy is built to not lose against the best possible response. It is not built to win the most against the player in front of you. Against a recreational player who calls too much, a pure GTO bluffing frequency leaves money on the table: you should bluff less and value-bet thinner.
So GTO is best understood as a baseline. It is the strategy you fall back on when you have no reliable read, and the reference point you deviate from when you do. Exploitative play is not the opposite of GTO; it is GTO plus a reason to change it.
Why solvers mix actions
Open a solver output and you will see hands that raise 60% of the time and call 40%. Players often read that as noise. It isn't.
At equilibrium, a hand that mixes is indifferent: both actions have the same expected value. The mix exists so that the range as a whole stays balanced. If you always raised that hand, your calling range would lose a piece and become easier to attack; your opponent's best response would change, and you would no longer be unexploitable.
Two practical consequences:
- A mixed hand is a close decision, not a random one. Getting it "wrong" by picking the less frequent action usually costs almost nothing in EV.
- Pure hands matter most. The hands that always fold, always raise or always call are where large mistakes live. Learn those first.
What a solver actually does
A solver takes a set of inputs and searches for the equilibrium of that game:
- Starting ranges for both players (for postflop spots, usually taken from a preflop solution).
- Stack depth and pot size.
- The bet sizes it is allowed to use: the betting tree.
- Rake, antes and payout structure where they apply.
The output is only as good as those inputs. Change the bet sizes or the starting ranges and you get a different "GTO". That is why libraries of pre-solved spots are labelled by format, stack depth and sizing, and why a 100bb cash solution is the wrong reference for a 20bb Spin & Go.
There are 22,100 possible flops, which collapse to 1,755 strategically distinct ones once you account for suits that play the same way. Nobody memorises that. Players learn how strategy shifts across textures: high and dry, low and connected, paired, monotone.
GTO vs exploitative play
| GTO baseline | Exploitative | |
|---|---|---|
| Goal | Can't be exploited | Win the most against a specific opponent |
| Needs reads | No | Yes, and they must be right |
| Risk | Leaves money vs weak players | Can be counter-exploited |
Strong players combine both. They know the baseline well enough to notice where an opponent deviates, then adjust in the direction that punishes it. Without the baseline you can't tell a real tendency from variance.
How to study GTO without drowning in it
Reading solver output feels like studying, but recognition is not recall. You can nod along at a chart and still misplay the spot at the table. A study loop that works:
- Start preflop. Preflop decisions happen every hand and are the easiest to memorise. A preflop leak repeats thousands of times.
- Pick the most common spots first: opening from the button, defending the big blind, the small blind vs big blind battle. See our guides to BTN vs BB and SB vs BB.
- Drill, don't just view. Use a GTO poker trainer that deals you the spot and grades your decision.
- Check your real hands against the strategy to find the leaks that actually cost you money.
- Measure in EV, not frequency. Matching a 60/40 mix exactly matters far less than never folding a hand that should always continue.
Common misconceptions
"GTO means always doing the same thing." No. GTO is full of mixed strategies, and the same hand plays differently depending on position, stack and board.
"Pros play pure GTO." Nobody can execute a full equilibrium at the table. Pros play simplified strategies close to GTO, and deviate when they have a reason.
"Solvers have solved poker." Heads-up limit hold'em was essentially solved in 2015. No-limit hold'em, especially multiway, is not. Solvers produce very close approximations for well-defined spots, which is more than enough for study. How close they get is covered in how accurate GTO solutions are.
For the fundamentals GTO builds on (ranges, position, pot odds), see our poker strategy guide.
Frequently asked questions
What does GTO mean in poker?
GTO stands for game theory optimal: a strategy that can't be exploited, because no change in the opponent's strategy lets them win more against it. In game theory terms it's a Nash equilibrium.
Is GTO the most profitable way to play poker?
Not against every opponent. GTO guarantees you can't be exploited, but against players with clear mistakes an exploitative adjustment from the GTO baseline wins more.
Is poker solved?
Heads-up limit hold'em was essentially solved in 2015. No-limit hold'em is not solved, but solvers produce close approximations for well-defined spots, which is enough for study.
Should beginners study GTO?
Yes, starting with preflop. Preflop ranges are easy to memorise and come up every hand, so a GTO preflop baseline is the fastest way to remove large, repeating mistakes.