Optimal strategies in two-player Kuhn poker

Each of two players receives one of three ranked cards, pays an ante, and can make or respond to a single fixed bet. The problem is to determine which betting and calling probabilities guarantee the value of the game at each ratio of bet to ante.

Why it matters and what counts as progress

Why it matters: The small game has incomplete information and allows a full description of bluffing and calling strategies. Changing the bet changes which strategies remain optimal.

What counts as progress: Compute the game value or prove an exact condition for an optimal strategy. State the stake range, the family of strategies and any choices at decision points that are never reached.

How the problem developed

  1. 1950 · Kuhn’s three-card game [1]

    Harold W. Kuhn Harold W. Kuhn introduced the two-player game and analyzed the game value and strategies, including several cases for general stakes.

  2. 19 Oct 2026 · A proof of the corrected Case 3 interval [2]

    JD Jones with Claude JD Jones’s project reports that Claude produced the exact certificates yielding the allowable middle-card calling interval for Kuhn’s specified Case 3 family. The proof also shows why the printed upper endpoint fails.

AI contribution

Kuhn poker: the value −1/18, all optimal strategies, and the solution at every ante and bet

JD Jones

For Kuhn’s three-card poker with positive ante a and a≤b≤2aa \le b \le 2a, corrects the Case 3 calling-probability interval for the middle card to [(2a−b)/(2a+b),b/(2a+b)][(2a-b)/(2a+b), b/(2a+b)] for the specified player-II strategy family. The printed upper endpoint fails, including at a=2a=2 and b=3b=3.

AI involvementAI-centered
ImportanceModest
Technical verificationFormal (Lean)
Human understandingFew
Paper date:

Sources

  1. Kuhn’s original paper, page 102
  2. Kuhn poker: the value −1/18, all optimal strategies, and the solution at every ante and bet
  3. Pinned Lean theorem statements
  4. Pinned project description and provenance
  5. Pinned project scope and AI disclosure
  6. A simplified two-person poker
  7. Project AI disclosure
Publication history