How To Win At Rock-Paper-Scissors With Science

aopsan / Image ID: 76804822 / Shutterstock

Can’t agree on who should get that last slice of pie? ”Rock, Paper, Scissors!”

While it may seem like there is no way to win at rock-paper-scissors above chance level, if you throw in a bit of game theory and psychology, you may well turn those odds in your favor. (And get that extra slice of pie while you are at it.)

Rock-paper-scissors originates in China, dating back centuries before it gained traction in Europe in the late 1920s. The game is meant to produce a one in three chance of winning, a one in three chance of drawing a tie, and a one in three chance of losing. But people do not play at random—they abide by subconscious patterns and tendencies that an opponent can exploit.

To determine how to do this, a group of scientists from Zhejiang University in China studied 360 students in a massive rock-paper-scissors tournament. As an incentive to do well, the winners were paid in direct relation to their number of wins.

The researchers found that players who won a match tend to repeat their winning hand. Losers, however, tend to switch it up. The “one in three” rule was no longer true. If a player lost playing rock, there was a greater than one in three chance of them playing paper the next round. 

To find out more, watch this six-minute video by Numberphile below. For even more fun, play a computer opponent that exploits a person’s game patterns over at The New York Times.

 

 

At the University of Tokyo, researchers have created a robot that wins rock-paper-scissors 100% of the time. However, the robot has an unfair advantage: It is equipped with a high-speed camera that recognizes within a millisecond which shape its opponent’s hand is going to make and then, of course, produces the winning hand.

Comments

If you liked this story, you'll love these

This website uses cookies

This website uses cookies to improve user experience. By continuing to use our website you consent to all cookies in accordance with our cookie policy.