Can the computer read your habits?
Story: The computer is not seeing your move early. It is just watching for habits, like “I often throw rock after I lose.”
Rock beats scissors. Paper beats rock. Scissors beats paper.
The AI is trying to guess what you will do next. Can it spot your pattern before you spot it yourself?
Try it. Play 30 rounds and watch the clue box after each turn:
Why the AI often wins
People are not very good at being random. We often have tiny habits we do not notice:
- After losing with scissors, most people switch to rock
- After winning, many people play the same move again
- Most people throw rock more than a third of the time
The AI watches your last few throws and guesses what comes next.
Big idea: predictable beats clever
If you play rock 33% of the time, paper 33%, and scissors 33%:
- Against pure rock: you win paper 33% of rounds, lose scissors 33%, draw rock 33%
- Your expected win rate: exactly 33% — same against any strategy
This is called a mixed strategy. That is a grown-up phrase for a simple idea: do not become easy to guess.
Big idea: when the other player can read your habits, “being tricky” is often worse than being balanced.
Practice
Rock beats scissors, paper beats rock, scissors beats paper. What is the result of rock vs rock?
draw — same move always draws. Rock vs rock = draw.
Why does a pattern-learning AI beat most humans at RPS?
The AI wins because it notices habits that people repeat without meaning to.
The "perfect" RPS strategy plays rock 33%, paper 33%, scissors 33%. What's your expected win rate with this strategy against any opponent?
Exactly 33%. The mixed strategy doesn't give you an edge — it removes the other player's edge. You win, lose, and draw equally often, no matter what they play.
What's the problem with 'thinking' of a random move instead of actually rolling a die?
Our brains often make patterns even when we try not to.