Become a fan of Slashdot on Facebook

typodupeerror

## Winning Algorithms For Rock, Paper, Scissors65

Celarent Darii (1561999) writes "The probability of winning at Rock-Paper-Scissors is about 1 in 3. However, people do not play entirely randomly, a study has revealed. People tend to follow hidden patterns that can be used to win more games. A short article on the BBC gives hints on the strategies to be used to get a competitive advantage with your Rock-Scissors-Paper nemesis." Remember, these strategies are for use against people, not robots.
This discussion has been archived. No new comments can be posted.

## Winning Algorithms For Rock, Paper, Scissors

• #### This is useless (Score:2, Funny)

by Anonymous Coward

When I play rock, paper, scissors with someone, we just play once unless it's a tie, so there are no patterns.

• #### Re:This is useless (Score:4, Insightful)

on Saturday May 03, 2014 @09:41AM (#46907481) Homepage

This may come as a surprise to you, but you're not the only person in existence.

• #### Re: (Score:1)

by Anonymous Coward

This may come as a surprise to you, but you're not the only person in existence.

Thanks, imaginary person, for your imaginary post.

• #### Re: (Score:2)

And /. only gives me mod points on the weekends...except this time. Dammit!
• #### Telling a Pinocchio? (Score:3)

I thought the more common protocol was to win 2, as seen in the arcade game Fist Talks.
• #### Re: (Score:1)

When I play rock, paper, scissors with someone, we just play once unless it's a tie, so there are no patterns.

I guess that now that you know there are patterns, you'll play it for hours on end.

• #### Re:This is useless (Score:5, Funny)

on Saturday May 03, 2014 @11:08AM (#46907949)

When I play rock, paper, scissors with someone, we just play once unless it's a tie, so there are no patterns.

If you play to the death there are no patterns either.

• #### Re: (Score:1)

I like to play it solitaire, right hand against left hand.
• #### A robot (Score:2)

Could watch you muscle movements and with sharper reflexes pick the winning option

• #### Wait. What? (Score:3, Interesting)

on Saturday May 03, 2014 @09:46AM (#46907515)
OK, so two people play this game, each has about the same chance of winning, and that chance is 1 in 3? BTW, nowhere in the original article does it state that.
• #### Re:Wait. What? (Score:5, Funny)

on Saturday May 03, 2014 @10:17AM (#46907699)

When you play high stakes Rock, Paper, Scissors in a Casino in Las Vegas, the house wins on a tie. That's how they make their money.

Also note that in Las Vegas, they can boot you for life for trying to "play smart", like just happened to Ben Affleck, for allegedly counting cards.

It doesn't seem fair to me . . . if you are good at math, and can count cards, why shouldn't you be able to use your intellectual skills at a game?

Those Las Vegas Casinos are discriminating against geeks. Someone outta sue.

• #### That's way too much of a house edge (Score:2)

When you play high stakes Rock, Paper, Scissors in a Casino in Las Vegas, the house wins on a tie.

Citation needed. The house has far less of an edge in craps, basic strategy blackjack, roulette, or regulated slots.

• #### Coin-op RPS (Score:2)

I wasn't quite sure if it was a joke. (Damn you Poe's law!) So here's another citation [wikipedia.org].

Now here's what brought it up: A local arcade has a coin-operated RPS machine called Fist Talks, and I wondered if casinos had been installing similar RPS machines that pay out. I know blackjack tables don't take pushes unless both the player and dealer bust, which provides the fundamental house edge in blackjack. Every rule that benefits the player (standing below 17, double down 10 and 11, split A-A and 8-8, increase

• #### Re: (Score:3)

The New York Times made a Rock-Paper-Scissors Flash-based bot [nytimes.com] a few years back. It's essentially exactly what the summary is talking about, since it learns your patterns and will in short order begin winning against you far more than merely 1-in-3 times. Alternatively, if you play the Veteran version of the bot, it has knowledge of all of the patterns from anyone who has ever played it, and it starts off beating you right from the start. If you could smuggle that into the casino, I'd be willing to bet that

• #### Re: (Score:2)

Or you could remember that you were in a casino and borrow a six-sided die in case the other guy was using a bot.
• #### Will they fucking let this go already? (Score:1)

by Anonymous Coward

We got it, brains are pattern-based machines at thus have a hard time generating pattern-less random sequences.
Can we now leave this trivial game alone and do some real science please?

• #### Re: (Score:2)

We got it, brains are pattern-based machines at thus have a hard time generating pattern-less random sequences. Can we now leave this trivial game alone and do some real science please?

Exactly. Now as to the algorithmic approaches for winning Rock-Paper-Scissors-Lizard-Spock...

• #### Re: (Score:2)

Play Lizard 100% of the time.

• #### Re: (Score:2)

Oh, no! Someone is doing something that you think is a waste of time!

The horror!

• #### The brain evolved... (Score:1)

... to survive, find mates and have kids. Everything else we got is extra. No one should expect the brain to do much more than what it was originally selected to do.

• #### Only most of the time? (Score:2)

Here's how to win all the time: http://www.k2.t.u-tokyo.ac.jp/... [u-tokyo.ac.jp]

Slashdot thinks I type too fast, so they won't take this comment until I wait a while. I can only guess that they want more drivel to fill the white space on the page.
• #### Re: (Score:3)

It's because you are typing too fast, if you had taken the time to follow all the links in TFS, you wouldn't have posted :)

But who RTFAs anyway, right ? :)

• #### Obligatory (Score:4, Funny)

on Saturday May 03, 2014 @11:14AM (#46907981)

Lisa's Brain: Poor predictable Bart. Always takes 'rock'.
Bart's Brain: Good ol' 'rock'. Nothin' beats that!
Bart: Rock!
Lisa: Paper.
Bart: D'oh!

• #### Re: (Score:1)

Rock rips through paper because of gravity. Rock wins always!
• #### Well, I don't know.. (Score:3)

on Saturday May 03, 2014 @11:18AM (#46908005)
Well, I don't know. Let's get a game started to test this..

Rock!

• #### Re: (Score:2)

Scissors! Oh rats...

• #### Friend of Mine and I Play (Score:3)

on Saturday May 03, 2014 @11:37AM (#46908123) Homepage Journal
A friend of mine and I play to determine exit order on the aircraft. I've beaten him 4 out of the last 5 times we've done it playing rock. Next time we play I'm going to point that out to him in advance and then the head games begin! Will I play rock again, because it's the predictable move, or am I just setting him up to throw paper because I'm planning to throw scissors?
• #### Re: (Score:2)

Same here. I'm one step ahead of you. My friend and I play for "who has to go deal with the stupid user"... Last time we played, he said "wow. When you said you always pick rock, you really meant you _always_ pick rock"...

He's not a stupid man... So he could be setting me up anticipating that I will play scissors next so I should pick paper...

Good thing there's no iocane powder involved.

• #### Win-stay lose-switch strategy (Score:2)

This strategy "win-stay lose-switch" is an very important and significant finding in game theory and evolution. The tournament of algorithms conducted in U Mich in late 1980s was a breakthrough in explaining the evolution of cooperation and altruism. But its winning strategy tit-for-tat was indistinguishable from "always-cooperate". So any mutant that does not follow "always-cooperate" would gain a very strong foot hold. This strategy was not evolutionarily stable strategy, ESS.

The win-stay lose-switch st

• #### Re: (Score:2)

Of course it is. That's how a neural network works.
It's hardly surprising that humans are terrible RNGs. It is, however, a good thing to keep in the back of your mind when in a competitive situation where that is relevant.

• #### Doesn't everybody knows that already? (Score:2)

(obvisouly I did not RTFA.)

When I took Simulations in gradschool 10 years ago, one of our assignment was to train a markov chain to predict the player next move at rock-paper-scissors. Using simply as state "lastmove, lastoutcome" is enough to learn what humans (read the students of the class) do.

• #### Am I missing something? (Score:2)

If the winner tends to stick with the same hand, while the loser tends to switch, doesn't that imply the loser will tend to win the next round? I mean, the only way the winner can win the next round by sticking with the same hand is if the loser also sticks with the same hand.
• #### Re: (Score:2)

that's what people tend to do, not the winning strategy. use the trend to your advantage. for example, if someone just beat your scissor with rock, he will tend to play rock again, so you should play paper next. whereas you might've been half-tempted to play rock also

• #### Down here (Score:1)

we tend to call it paper-scissors-rock, so...
• #### Best 2 out of 3 (Score:2)

Next time I play, I'll suggest playing best 2 out of 3 so as to be able to use this tactic.

• #### Maybe... (Score:1)

They say that the person chooses the next in sequence by the name of the game, but that may not be why. Perhaps when they lose they choose the pattern that beats the pattern they just used. It still results in the same sequence, R>P>S, but for a different reason.
• #### Not truly random (Score:2)

FTA:

When players won a round, they tended to repeat their winning rock, paper or scissors more often than would be expected at random (one in three).

If it was truly random then anything could happen, including a game where the opponent only chooses scissors the entire time.

• #### Did they consult Dr. Sheldon Cooper? (Score:1)

I'm sure Dr. Cooper would have told them the winning algorithm for Rock-Paper-Scissors-Lizard-Spock.

#### Related LinksTop of the: day, week, month.

"Never face facts; if you do, you'll never get up in the morning." -- Marlo Thomas

Working...