Infinite Rock Paper Scissors - Play Unlimited Rounds
This is infinite Rock Paper Scissors. There is no round limit, no "best of three", no leaderboard timer cutting you off. Click as many times as you like and every round adds to your running stats. Wins, losses, ties, streaks, choice frequencies, and the last sixty results are stored in your browser's local storage so you can close the tab, come back tomorrow, and pick up exactly where you left off. Open Reset Stats when you want a clean slate.
An infinite RPS game is the cleanest possible test of probability. Against a true random opponent (which is what this page uses), your numbers will converge over time toward 33.3% wins, 33.3% losses, and 33.3% ties. Watching that convergence happen live is half the fun - and the Outcome Breakdown donut above shows you the live shape of your distribution.
How to Play
Click one of the three buttons - Rock, Paper, or Scissors - or use your keyboard (R, P, S). The computer picks at the same instant using a uniform random draw. Classic rules decide the winner:
- Rock crushes Scissors
- Paper covers Rock
- Scissors cut Paper
- Same move on both sides is a tie
How to (Almost) Always Win at Rock Paper Scissors
You can't always win Rock Paper Scissors against a truly random opponent - the math forbids it. But humans aren't random, and decades of research on RPS play have surfaced consistent patterns you can exploit. The strategies below have been documented in academic studies (notably the 2014 Zhejiang University experiment with 360 players) and in years of competitive play at the World RPS Society. Used together, they regularly push human-vs-human win rates well above 50%.
1. Open With Paper
The single most reliable bias in Rock Paper Scissors is the opening throw. Casual players - and men especially - pick Rock first more often than chance would predict. Rock feels "strong" and "default", which is exactly why it's exploitable. Opening with Paper covers Rock and burns the most common beginner mistake before round one is over.
2. Use the Win-Stay, Lose-Shift Rule
This is the best-documented pattern in human RPS play. After winning a round, people tend to repeat the move that just worked. After losing, they shift to whatever would have beaten their opponent. So if your opponent throws Paper and wins, expect Paper again - and play Scissors. If they throw Rock and lose to your Paper, expect them to play Scissors next - and play Rock. Mirroring this rule is how seasoned players grind out a steady edge.
3. After a Tie, Expect a Switch
Humans rarely repeat a tied throw. After a Rock-Rock tie, players overwhelmingly switch to Paper or Scissors. If you can read which way they lean (most lean to the move that beats their own previous throw - so Paper after a Rock tie), you can play the counter. Even a coin-flip guess between the two switches still puts you at a 75% expected win/tie rate that round.
4. Don't Telegraph
The other half of the game is hiding your own pattern. Tension in your throwing hand, eye flicks, micro-pauses before you throw, even how tightly your fist is balled - all leak information. Practice throwing all three moves with the exact same posture and tempo. If you want to be unreadable, you have to look bored.
5. Use Real Randomness Against Strong Players
If you're playing someone good enough to read patterns in your own play, the only winning move is to stop having patterns. Top World RPS Society players have been known to memorise dice sequences, use the seconds hand of a watch, or precommit to a sequence from a random.org list. The Choice Frequency chart on this page is the diagnostic - if your three bars aren't roughly equal, you're leaking signal.
Why Humans Are Bad at "Random"
The reason these strategies work is that the human brain is genuinely terrible at generating random sequences. Asked to produce a random series of throws, people:
- Alternate moves more often than chance would (because long repeats feel "non-random")
- Avoid playing the same move three times in a row, even though that should happen about 11% of the time
- Open with Rock significantly more than 33% of the time
- Default to win-stay lose-shift unless they've been trained out of it
The Choice Frequency chart above is a mirror for these biases in your own play. Equal bars mean you're playing randomly. Lopsided bars mean a smarter opponent could profile you.
What the Stats Show
- Wins / Losses / Ties - Running totals across every round you've played on this device
- Win Rate - Wins divided by decided (non-tie) rounds. Against true randomness this converges to 50%.
- Streak - Your current consecutive wins; resets on a loss or tie
- Best Streak - The longest win streak you've recorded on this device
- Outcome Donut - Proportion of wins, losses, and ties across all rounds, with current win share in the center
- Choice Frequency - How often you and the computer have picked each move - useful for spotting your own habits
- Recent Rounds - Your last 60 results as a strip of W/L/T cells, newest on the right. Hover any cell to see that round's matchup.
Is the Computer Really Random?
Yes. The opponent uses Math.random() to pick uniformly from the three options every round, with no memory of what you played before. That makes it impossible to systematically beat - which is the whole point of an honest infinite RPS game. Over many rounds your win rate will hover around 33% wins, 33% losses, 33% ties, with normal statistical wobble. If your chart drifts far from those numbers it's either small-sample variance or, more interestingly, evidence that you have a bias the computer happens to be punishing or rewarding by chance.
Where Your Stats Live
Everything on this page is stored in your browser's local storage. Nothing is sent to a server, and the data is tied to this device and browser. Use the Reset Stats button to wipe it clean.