Do This Instead of Rock Paper Scissors If You Are More than 2 People
Posted: 2026-04-21
Intro
You’ve probably been in this situation. There are 5 or 6 or more of you and you need to choose who gets to go first in a board game. Or be the goalkeeper. Or run off to the shop and get coffee, or whatever. Someone says: “Let’s play rock-paper-scissors for it. Rock-paper-scissors-shoot!” And then you look at 2 rocks, 3 papers, and 1 scissor hand and begin to realize that this game wasn’t meant for so many players. But you keep going, round after round, until you eventually eliminate all but one player.
What if I were to tell you there’s a better solution that scales well with arbitrarily many people?
Zimi zami zum
In Serbia, we say “zimi-zami-zum”. When “zum” is said, everyone shows either a fist (like rock) or a relaxed palm (like paper).1 This is played by 3 players. The odd one out (eg. the lone palm vs. 2 fists) is “it”. We can scale this up to more players by deciding that the majority is eliminated and the minority moves to the next round. Repeat this process until there are fewer than 3 people remaining.2
Here’s an example with 5 players. The 3 fists are eliminated in this round:
If everyone chose the same option or it’s a tie (eg. 3 fists and 3 palms), no one is eliminated and everyone plays another round. If 2 players remain, they then switch to playing rock-paper-scissors.
While, yes, the picture above looks the same as rock-paper-scissors, this game has much fewer combinations where no player gets eliminated. In rock-paper-scissors, the more players there are, the higher the chance that all 3 choices will be present and no one is eliminated. In zimi-zami-zum, you will very often more than halve the number of remaining players. In fact, let’s prove that this game is indeed better suited for our task.
Analysis
I wrote a simple C++ program to calculate how many players, on average, will move on to the next round of either game (elimination_analysis.cpp).
In the table below, the left-most column is the starting number of players. The middle column is how many players will, on average, move on to the next round of rock-paper-scissors. The right column is that, but for zimi-zami-zum. A lower number is better, as more progress was made with more players eliminated.
| No. players | RPS | ZZZ |
|---|---|---|
| 3 | 2.00 | 1.50 |
| 4 | 2.96 | 2.50 |
| 5 | 4.07 | 1.88 |
| 6 | 5.23 | 3.19 |
| 7 | 6.40 | 2.52 |
| 8 | 7.54 | 4.06 |
| 9 | 8.65 | 3.30 |
| 10 | 9.74 | 5.02 |
| 11 | 10.81 | 4.16 |
| 12 | 11.86 | 6.01 |
| 13 | 12.90 | 5.04 |
| 14 | 13.93 | 7.00 |
| 15 | 14.95 | 5.93 |
We can see that zimi-zami-zum always eliminates more players on average. In fact, with a lot of players, rock-paper-scissors rarely eliminates anyone.
Next, I wrote a simulation of both approaches to analyze how many rounds it takes to go from N players down to 1 (elimination_simulation.cpp). I ran a million simulations for each starting number of players. Lower numbers are better as that means fewer rounds were required.
First, the results for rock-paper-scissors:
| No. players | avg | min | p50 | p90 | p95 | p99 | max |
|---|---|---|---|---|---|---|---|
| 3 | 2.25 | 1 | 2 | 4 | 5 | 6 | 17 |
| 4 | 3.21 | 1 | 3 | 6 | 7 | 9 | 21 |
| 5 | 4.49 | 1 | 4 | 8 | 10 | 13 | 35 |
| 6 | 6.23 | 1 | 5 | 11 | 14 | 19 | 47 |
| 7 | 8.66 | 1 | 7 | 16 | 20 | 28 | 73 |
| 8 | 12.10 | 1 | 10 | 23 | 29 | 42 | 118 |
| 9 | 17.11 | 1 | 14 | 34 | 43 | 62 | 184 |
| 10 | 24.34 | 1 | 19 | 50 | 63 | 93 | 280 |
| 11 | 34.97 | 1 | 27 | 73 | 92 | 138 | 373 |
| 12 | 50.59 | 1 | 38 | 107 | 137 | 206 | 717 |
| 13 | 73.67 | 1 | 55 | 158 | 202 | 306 | 905 |
| 14 | 107.86 | 1 | 79 | 235 | 302 | 457 | 1368 |
| 15 | 158.99 | 1 | 115 | 350 | 451 | 686 | 1848 |
And then for zimi-zami-zum:
| No. players | avg | min | p50 | p90 | p95 | p99 | max |
|---|---|---|---|---|---|---|---|
| 3 | 1.33 | 1 | 1 | 2 | 3 | 4 | 12 |
| 4 | 2.00 | 1 | 2 | 4 | 5 | 7 | 23 |
| 5 | 2.07 | 1 | 2 | 3 | 4 | 5 | 35 |
| 6 | 2.59 | 1 | 2 | 4 | 5 | 7 | 17 |
| 7 | 2.26 | 1 | 2 | 3 | 4 | 5 | 15 |
| 8 | 2.66 | 1 | 2 | 4 | 5 | 6 | 15 |
| 9 | 2.64 | 1 | 2 | 4 | 5 | 7 | 20 |
| 10 | 3.01 | 1 | 3 | 5 | 6 | 8 | 21 |
| 11 | 2.87 | 1 | 3 | 4 | 5 | 7 | 22 |
| 12 | 3.20 | 1 | 3 | 5 | 6 | 8 | 19 |
| 13 | 3.21 | 1 | 3 | 5 | 6 | 7 | 21 |
| 14 | 3.51 | 1 | 3 | 5 | 6 | 8 | 22 |
| 15 | 3.27 | 1 | 3 | 5 | 5 | 7 | 18 |
As you can see, the number of rounds required for rock-paper-scissors grows fast as you add more players, while for zimi-zami-zum it barely moves up.
Conclusion
I hope I got my point across and have proven to you that zimi-zami-zum is better with any number of players greater than 2. Obviously, if you need to pick 1 person out of 10+, you’ll probably whip up an app, a die, or just force the newest person in the group to run the errand. But zimi-zami-zum is good if you’re in the 3-7 players range, especially if you’re at 3 or 4.
Unfortunately, most people don’t know about it, so you’ll first have to explain the game and convince them to go along with it. Feel free to show them this post :-)
I could not find the origin of the game or its name. Everyone just picked it up from other kids. I did find out that Croatians play “fli-fla-flu”. As far as I can tell, it’s the same game, but the choices are palm-up and palm-down.↩︎
The original game can be played with more than 3 players. A trio plays and the odd one out is eliminated. Then, the next player takes their place. This is repeated until two players remain; they get a third “helper” player to help them eliminate one of the two. (source)↩︎
Used images by Amonrat Rungreangfangsai.↩︎