Solver
Also known as: GTO solver, poker solver, PioSolver, GTO Wizard, MonkerSolver
Software that approximates Nash-Equilibrium strategies for a poker spot by iterating ranges toward minimal exploitability.
A solver is software that computes an approximate Nash Equilibrium for a defined spot: given both players' starting ranges, stacks, and allowed bet sizes, it iterates (typically via counterfactual regret minimisation) until neither range can improve — reported as exploitability in bb/100, driven toward zero. Output is a full strategy: action frequencies and EV for every hand at every node, range vs range.
The standard tools: PioSolver and GTO Wizard (cash + MTT, with ICM-aware solves), Simple Postflop, and MonkerSolver (multi-way, preflop trees). They are the source of every modern baseline — c-bet frequencies, 3-bet ranges, bluff-to-value ratios.
Limits worth respecting:
- Abstraction: solvers use a discrete set of bet sizes, not the continuous real game — give it the wrong sizes and the solution shifts.
- Heads-up bias: multi-way solves are slower, less stable, and less authoritative.
- Inputs are assumptions: the solve is only as good as the ranges you feed it. Garbage ranges, garbage equilibrium.
- It's a baseline, not a read: the equilibrium assumes a GTO opponent. Against a real pool, layer node-locking and exploitative play on top.
Example
You suspect your turn check-raise frequency is too low. Build the spot in GTO Wizard with the exact preflop ranges and the sizes actually in use, run it, and read the check-raise %. If the solver mixes 22% raise and you raise 5%, you're under-pressuring — a measurable, fixable leak.