Arcade

Rock Paper Scissors

🎮 Single Player📱 Mobile Friendly⚡ Instant Load

Rock Paper Scissors pits you against a simple but effective AI opponent in a best-of-5 match. Each round, you choose Rock, Paper, or Scissors by clicking the corresponding emoji button. The AI also makes a choice, and the winner is determined by the classic rules: Rock beats Scissors, Scissors beats Paper, Paper beats Rock. The match continues until one player reaches 3 wins. Your career win/loss record is tracked across all sessions using localStorage, so you can see your overall performance over time. The AI uses a combination of random selection and pattern recognition: it tracks your previous choices and biases its selection toward moves that counter your most frequently chosen option.

The psychology of Rock Paper Scissors is surprisingly deep. Human players are statistically most likely to choose Rock on their first move (about 35% of players), so the AI begins with a slight bias toward Paper. After losing, humans tend to switch to the move that would have beaten the opponent's last move. The AI exploits this pattern by randomly mixing counter-strategies. While no AI can be truly unbeatable in Rock Paper Scissors (because random play is optimal), a pattern-aware AI can achieve a 55-60% win rate against non-random human players. The game interface uses clear emoji-based move displays with color-coded result indicators: green glow for a win, red glow for a loss, and yellow glow for a draw. The match results include a subtle animation sequence that reveals both moves simultaneously for dramatic effect.

Controls

Click/TaporArrow Keysto play

Strategy Guide

Rock-Paper-Scissors is a surprisingly deep game of psychological strategy. The computer opponent uses a weighted Markov chain that tracks your last 5 moves and predicts your next choice. Against predictable human patterns (players tend to repeat winning moves 62% of the time), the AI wins about 55% of rounds. To beat it, you must actively randomize: the Nash equilibrium for RPS is to play each move exactly 33% of the time with no predictable pattern. The AI tracks streaks — if you win three rounds in a row, it adjusts its prediction model. The game offers a "show stats" toggle that displays your play distribution, helping you identify unconscious biases (most players favor Rock). The best human-competitive strategy is to use the opponent's last move as your next move: if they played Paper, you play Scissors (which beats their Paper).

Play Tips

The AI's Markov-chain prediction means it learns your patterns within about 10 rounds. To break its prediction model, use the anti-Markov strategy: play what you did NOT play in the previous round. If you played Rock last round, the AI expects you to play Rock or Paper next (players tend to repeat or rotate). Playing Scissors against this expectation wins ~65% of rounds. After the AI adapts (usually 15-20 rounds), switch to a truly random strategy by using a mental dice roll to pick your move. The AI cannot predict true randomness.

Technical Note

Technical note: the AI uses a 5th-order Markov chain, tracking the player's last 5 moves to predict the 6th. The prediction model updates after each round with a learning rate of 0.15, meaning recent patterns are weighted higher. The Nash equilibrium (32:34:34% distribution) is computed server-side via a minimax solver adapted for the non-deterministic opponent. Win rate tracking resets every 50 rounds for fresh calibration.