Online RPS

[nutsinho]

http://www.rubl.com/rock-paper-scissors.html


The ultimate game of reading one's opponent. In the version preferred on this site, a win with rock is worth 1 pt, a win with paper is worth 2 points, and a win with scissors is worth 3 points. Matches are decided by the first to win a certain number of rounds (up to best-of-13 series), with each round decided by the first to get a certain number of points (from like 6-15). There is a TON of strategy involved. I got this link off 2p2 and I had a blast playing last night, going 6-0. If anyone wants to challenge me, let me know what time you want to play tonight (1o pm est or later). Or just challenge each other; we could get some type of FTR RPS ladder going.

[givememyleg]

dude im good i beat this one guy like 6-1

[nutsinho]

let me know when your rating is 1700...holla

[mcatdog]

I love playing this type of game but let me think about the strategy for a few minutes first.

[nutsinho]

my observations thus far:

with this scoring format, picking randomly is a bad strategy because that can be exploited by picking scissors very often (scissors vs random would have a positive expectation of 2/3 pt per turn, rock and paper vs random has exp. of -1/3 pt per turn each).
So generally against people with a losing or breakeven record, I just try to outplay what I expect them to be thinking on the first level; against winning opponents I'm either trying to outplay their 2nd/3rd level thinking on each individual turn or trying to make them fire paper too often on the whole.

In 1-1-1 RPS it becomes basically a crapshoot between top players thinking at the umpteenth level because picking randomly is an unexploitable breakeven strategy. However, in 1-2-3 RPS I think there is a ton of strategy involved and playing optimally would have to involve degrees of both good reading of opponent and some randomization (in a non 1:1:1 ratio)

[mcatdog]

my observations thus far:
with this scoring format, picking randomly is a bad strategy because that can be exploited by picking scissors very often

Picking randomly is unexploitable as long as you have the right frequencies of scissors, paper, and rock (which isn't 1:1:1). I'll let you guys figure out exactly what the right probabilities are.

[nutsinho]

my observations thus far:
with this scoring format, picking randomly is a bad strategy because that can be exploited by picking scissors very often

Picking randomly is unexploitable as long as you have the right frequencies of scissors, paper, and rock (which isn't 1:1:1). I'll let you guys figure out exactly what the right probabilities are.

you'd also have to randomize in a different fashion at the end of each round if the score is close; if you only need one point and your opponent needs three, it is dumb to pick paper even if you know your opponents knows you know this.

[nutsinho]

i also don't know why randomizing by any ratio would be unexploitable; seems like if your opponent figured out your ratio he could select one that wins against it. I want to see your logic/math.

[mcatdog]

I just beat a guy pretty horribly. I came up with a way to randomly select rock, paper, or scissors according to the probabilities that I figured out, and used it for the first two rounds, which he happened to win. During those two rounds I noticed that he was picking rock a lot more often than he should be. I adjusted my method slightly so that 25% of the time it told me to pick scissors, I picked paper instead. Then I beat him the next 3 rounds to win the match, and then won the next match 3-1. Most of those 7 rounds weren't even close.

[mcatdog]

i also don't know why randomizing by any ratio would be unexploitable; seems like if your opponent figured out your ratio he could select one that wins against it. I want to see your logic/math.

Figure out how to exploit a given ratio and you'll quickly see that there's one ratio that's unexploitable. Good point on adjusting the probabilities near the end of the match.

[nutsinho]

thoughts on those matches mcatdog? Those were by far the most challenging matches I've played. I'd only lost 2 rounds in 8 matches going in.

[mcatdog]

My thoughts are that the correct strategy picks scissors a slight majority of the time, but in the beginning you were picking paper almost never and rock way too often. I tried to exploit that by picking rock and paper a little bit more often than I should. I still picked scissors about a third of the time hoping that you wouldn't catch on to what I was doing.

I think I need to occasionally what the coins are telling me to pick and try to get into the other person's head, instead of just playing like a computer.

[nutsinho]

Yeah, I HATE picking paper if I have no reason to believe my opponent isn't picking scissors. It generally takes me a while into each round before I feel like it's time to use paper. This is obviously a mistake if you know what I'm up to but I think by getting you to change your strategy to one that is exploitable I can make up more ground later on.

[swiggidy]

i also don't know why randomizing by any ratio would be unexploitable; seems like if your opponent figured out your ratio he could select one that wins against it. I want to see your logic/math.

This should be pretty easy for a poker player to figure out.

You adjust your ratios so that everything he picks has the same eV. If your opponent is not playing perfectly you can adjust your ratios for more eV, as mcatdog is doing.

[nutsinho]

i also don't know why randomizing by any ratio would be unexploitable; seems like if your opponent figured out your ratio he could select one that wins against it. I want to see your logic/math.

This should be pretty easy for a poker player to figure out.

You adjust your ratios so that everything he picks has the same eV. If your opponent is not playing perfectly you can adjust your ratios for more eV, as mcatdog is doing.

Well there's a difference between an unexploitable winning strategy and an unexploitable breakeven strategy. I don't know how much EV can be gained from a pure randomization approach, but I don't think it's as great as the edge you have if you've got a decent idea what your opponent is thinking a lot of the time.

[bigred]

wow...

[mcatdog]

Well there's a difference between an unexploitable winning strategy and an unexploitable breakeven strategy. I don't know how much EV can be gained from a pure randomization approach.

None, and in fact, I think that between two players who understand the mathematics of this game, no one should ever have any edge. The reason you can win so often is that most players are fish because they pick paper way too much. A mixed random strategy of 70% scissors, 15% rock, and 15% paper should be able to crush 90% of the players on that site.

[nutsinho]

haha,that reminds me I should start calling my defeated opponents "fish"