Verification
Last updated: July 6, 2026 · measured on the shipping engine
Every trainer claims its coaching is accurate. OptimaPoker measures it, the same way you'd audit a trading system: against ground truth, at scale, adversarially, and again after every change. This page is the summary of those receipts. The numbers below are actual outputs of the audit harnesses, not marketing copy, and they are re-run continuously — a dedicated 32-thread lab machine re-verifies the engine at full precision after every significant change.
1. Coach vs. solver ground truth
The core question: when the coach recommends a play, how much EV does that play give up against an exact game-theory solution of the same spot? The audit deals full games, harvests the coach's real decisions, then solves each spot to equilibrium with the CFR solver and measures the gap.
2. Solver convergence
The live in-app solver solves your actual hand to a measured exploitability of ≤0.5% of the pot — the same convergence bar used by professional desktop solvers. Convergence is pinned by automated proofs on both flop-rooted and turn overbet trees; those tests must pass before any release ships.
3. Realism at scale
The opponents are only worth training against if they behave like real players. Every archetype is calibrated against published population statistics (VPIP, PFR, 3-bet, c-bet, aggression factor, donk/probe and check-raise line shapes), and the whole field is stress-tested in mass simulation:
- 100,000,000 hands simulated with zero hard violations — chip conservation, action legality, and hand termination asserted on every single hand.
- Per-style × per-skill statistics verified against the calibrated bands on every run.
4. Unexploitability
A training field you can beat with one weird trick teaches you the trick, not poker. An adversarial harness plays nine fixed exploit strategies — always-fold, maniac shoving, light 3-bet spam, float-and-stab, river overbet bluffing, and more — against the shipping bot field. Every one of them loses. The field also adapts: it tracks public frequencies and punishes pattern abuse, the way a live table would.
5. Permanent gates
The engine carries 230+ automated tests, including three permanent validation harnesses that run before every release: a coverage sweep proving every reachable decision is coached and legal (zero coaching gaps), preflop invariants (position monotonicity, no junk opens, push/fold ordering pinned to Nash), and postflop invariants (MDF, pot odds, and bluff ratios pinned to the textbook numbers). The GTO math isn't believed — it's asserted.
Method notes
- Audits grade the coach's real in-game decisions, not hand-picked spots.
- The coach never sees opponent hole cards or the runout — it reasons from ranges, with exact card removal, like a player must.
- Full-precision re-verification runs after every engine change; this page is updated as new receipts land.