Rock Paper Scissors gets dismissed as pure chance, but repeated play against another thinking opponent — human or a reasonably designed CPU — is actually a pattern-reading game disguised as a coin flip.
Humans are structurally bad at randomness
Given a free choice, most people don't distribute their picks evenly across rock, paper, and scissors. People tend to under-repeat their previous choice and subtly favor certain sequences without realizing it — which means a genuinely random-feeling sequence to the person making the picks is often statistically detectable to an opponent paying attention.
Avoid repeating your last winning move
A common (and exploitable) habit is repeating whatever choice just won. If you win with rock, there's a real temptation to throw rock again — which is exactly the kind of predictable pattern a good opponent, or a well-tuned CPU, will learn to counter.
Our CPU actually adapts
Our Rock Paper Scissors build features an adaptive CPU that tracks your pattern across the match, not just your last move. Beating best-of-5 against it consistently requires actively varying your strategy — including occasionally making a choice that feels "wrong" purely to stay unreadable.
The best counter-strategy is inconsistency, on purpose
Because the goal is to be unpredictable, the best long-run approach is deliberately breaking your own patterns — if you notice you've favored one choice twice in a row, force a different one next, specifically because an adaptive opponent is likely tracking exactly that tendency.
Why this matters beyond the game
The same "avoid being predictable to an adaptive opponent" logic shows up in other competitive games with hidden information — poker bluffing frequency, penalty kick placement, even negotiation tactics. Rock Paper Scissors is a genuinely useful, low-stakes place to practice noticing your own patterns before an opponent does.
Play Rock Paper Scissors against the adaptive CPU, or if reading an opponent's patterns over many rounds sounds appealing, Uno rewards the same kind of adaptive thinking with a full deck of cards instead of three choices.