A zero-sum game with no pure-strategy equilibrium: the only stable choice is to randomise in equal parts. It is the basic model of mixed-strategy equilibrium.

Example — Zero-sum game: odds or evens

Two players simultaneously show 11 or 22 fingers. G1 wins 11 EUR if the sum is even; G2 wins 11 EUR if it is odd. Payoff matrix (for G1):

G1 / G212
1+1+11-1
21-1+1+1

(G2’s payoffs are the opposites: zero-sum game.)

Pure strategies: none dominant; no pure-strategy equilibrium (if G1 fixes a choice, G2 deduces their own winning move).

Mixed strategies: G1 plays “1” with probability pp, G2 with probability qq. G1’s expected winnings: Vˉ1(p,q)=pqp(1q)(1p)q+(1p)(1q)=(2p1)(2q1).\bar V_1(p,q) = pq - p(1-q) - (1-p)q + (1-p)(1-q) = (2p-1)(2q-1). G2 chooses qq to minimise Vˉ1\bar V_1; if p1/2p\neq 1/2, G2 chooses qq at the extremes to push Vˉ1\bar V_1 negative. The only robust choice for G1 is p=1/2p=1/2 (G1 indifferent to the opponent). By symmetry q=1/2q=1/2.

Mixed-strategy Nash equilibrium: both play 1/21/2-1/21/2. Expected winnings of each: Vˉ=0\bar V = 0 (fair game).

Topics: Probabilita
Concepts: Equilibrio di nash · Gioco a somma zero · Strategia mista
Methods: Equilibrio nash · Gioco somma zero · Strategia mista
Skills: Calcolo probabilita · Modellizzare