The mind games that actually work in rock paper scissors are about habits, not hands: people repeat patterns after wins and losses, favour certain opening throws, and are bad at being random, so you win by noticing those patterns and nudging them. No trick can read a hidden throw, and anything that relies on seeing a hand early is a foul, not a mind game.
Here are the tactics I think are worth using, the research behind them, and one popular claim you should drop.
First, a myth to drop
You may have seen articles claiming a psychologist's investigation found your favourite throw predicts your personality (rock players are aggressive, paper players are diplomats, and so on). I looked for that study and could not find one. An older article on this site repeated the idea; it had no source, and I have removed it. Treat any "your throw reveals your personality" claim as entertainment.
The biases that make people predictable
People are poor at producing random sequences. Psychology research on this goes back decades; a classic review by W. A. Wagenaar (1972, Psychological Bulletin) found that people asked to generate random sequences avoid repeats and alternate too much. In rock paper scissors that shows up as players avoiding the same throw three times running, even though a truly random player would do it often.
Win-stay, lose-shift. In a 2014 study, Zhijian Wang, Bin Xu and Hai-Jun Zhou at Zhejiang University had 360 students play repeated rounds for small cash payments. Overall frequencies were close to one third each, but at the individual level, winners tended to repeat their winning throw more than chance, and losers tended to shift, most often to the throw that would have beaten the one they just lost to.
Opening habits. Across a very large dataset of online games, Batzilis and colleagues (2019, Games) found that players respond to what their opponents did in previous games rather than playing purely at random. And in our own data, rock is the most common first throw.
What our data shows, with caveats
On wrpsa.com, between 29 July and 27 September 2026, we recorded 3,020 throws from 943 players, mostly against the computer. That is a small sample, from one site, against a machine, so treat it as a hint rather than a law.
| Pattern | What we saw |
|---|---|
| Overall | Rock 36.2%, paper 35.6%, scissors 28.2% |
| First throw | Rock 38.1%, paper 36.0%, scissors 25.9% |
| After a win | Repeated the same throw only 20.0%; moved to the throw that beats it 42.3% |
| After a loss | Upgraded 38.5%, downgraded 34.0%, stayed 27.5% |
The after-a-win line differs from the Zhejiang lab result, where winners tended to stay. Our players moved on after winning. That may be because they were playing a computer, or because casual online players expect the opponent to adjust. Either way it shows why you should read the person in front of you rather than rely on a rule from a study.
Tactics that work (and are fair)
1. Open with paper against a new opponent. If rock is the most common opener, paper is the percentage play for round one. It is a small edge, but it is free.
2. Track your opponent's response to winning. After they win a round, note whether they repeat or move. Two or three rounds usually reveal a lean. If they move to the throw that beats their winner (our most common pattern), throw the thing that beats that.
3. After they lose, expect a shift. Losers rarely stay. If you just beat their rock with paper, they are more likely to come back with scissors (to beat your paper) than with rock. Rock beats scissors, and rock was also what they just lost with, so many players will not expect it.
4. Announce a throw. Saying "I'm going rock" before a round is a classic bluff. It works because the opponent has to decide whether you are telling the truth, double bluffing, or just trying to rattle them. Keep it legal: announce before the count, then play on the beat like normal. Use it once or twice a match, not every round, or it stops working.
5. Break your own patterns. Every tactic above can be used against you. Every few rounds, check whether you have thrown the same thing after a win twice in a row, and change it.
6. Pace, don't stall. Keeping a brisk, even count takes away thinking time from an opponent who plays better when slow. Deliberately stalling to look at their hand is a late throw and a foul under our rules.
What does not work
- Watching the hand during the count. At normal speed the shape forms too late to read. Players who seem to do it are usually throwing late.
- "Nobody throws scissors twice." Some do. Rules of thumb are guesses, not guarantees.
- Playing perfectly random. It cannot lose on average, but it cannot win on average either. If your opponent is predictable, pure randomness wastes the edge.
For the game-theory view of why random play is safe but unexploitative, see game theory. For the fuller picture of the research, see our psychology page. To test these tactics against a patient opponent, play online.
FAQ
Can you read someone's hand before they throw?
Not reliably at normal speed. What you can read is their history, which is where every legitimate edge comes from.
Does your throw reveal your personality?
There is no study I can find showing that. Treat that claim as a myth.
What is the best first throw?
Paper is a reasonable default, because rock is the most common opener in the data I have seen. It is a small edge, not a guarantee.