RoShamBo - Rock,Paper,Scissors
I think its along the same lines of poker. Humans can try as much as they want but thy are bad when trying to be random. I know a few techniques and the theories behind them however, I'm not sure if it would be enough to beat the commission. I would guess if you found a RPS fish you could clean up.
My question is, is this beatable even with a house rake/percentage, assuming you played a random person every time? Some good some bad.
It sounds like a computer program could be easily made to play within a set of rules that would be advantageous.
this is a very indepth website that I had saved about the game.
http://www.datagenetics.com/blog/march52013/index.html
Quote: AxelWolfI see they now have a bit coin gaming site that offers peer vs peer RPS. I find this somewhat interesting and think it would be a good topic.
I think its along the same lines of poker. Humans can try as much as they want but thy are bad when trying to be random. I know a few techniques and the theories behind them however, I'm not sure if it would be enough to beat the commission. I would guess if you found a RPS fish you could clean up.
My question is, is this beatable even with a house rake/percentage, assuming you played a random person every time? Some good some bad.
It sounds like a computer program could be easily made to play within a set of rules that would be advantageous.
There are (or, at least, used to be) RPS programming competitions where the computers would play against each other. Of course you can just pick randomly and break even against everyone, but, to prevent people from doing that, they would run your program against several easy-to-beat programs (like, for example, against a program that always picked rock), so if you pick randomly no matter what, you will not beat those and it will hurt your overall score.
Quote: AxelWolfI Humans can try as much as they want but thy are bad when trying to be random.
I don't agree with this. I have a system of RNG in my head that I use when I play poker. Say if I calculate in a certain situation I should call 27% of the time and fold 73% of the time I actually have a way of generating that RNG. Feels pretty awkward to do at first but once I'm used to it it becomes somewhat natrual. Not sharing it though.
You cant share it because it probably dose not work. Someone would find a way to poke holes many holes in it. People like to think they can do things they can't, like control dice.Quote: NeutrinoI don't agree with this. I have a system of RNG in my head that I use when I play poker. Say if I calculate in a certain situation I should call 27% of the time and fold 73% of the time I actually have a way of generating that RNG. Feels pretty awkward to do at first but once I'm used to it it becomes somewhat natrual. Not sharing it though.
I'm sure the same type of thing you are doing, can be achieved by looking at where the 2nd hand on your watch is at the time you need to make the random decision.
sounds like a ton of effort to go through in the limits you play. I highly doubt anyone you are playing poker with is catching on to if you call hands like 10, 10 fold or raise them.
Quote: AxiomOfChoiceThere are (or, at least, used to be) RPS programming competitions where the computers would play against each other. Of course you can just pick randomly and break even against everyone, but, to prevent people from doing that, they would run your program against several easy-to-beat programs (like, for example, against a program that always picked rock), so if you pick randomly no matter what, you will not beat those and it will hurt your overall score.
That's pretty cool. So basically two separate programs battle it out? That's cool. :)
Proper RPS strategy -- go paper every time, if you think enemy catches on, switch to scissor. Only play rock if you are certain enemy is playing scissor (low chance). Never play scissor for first time against someone as they are likely to pick rock. Only play scissor (at start) if you're certain they'll choose paper (less likely than rock but more likely than scissor).
@Axel,
What's the vig? How does it work (house always collects vig, only collects vig on win...)?
If it's something astoundingly high, I'd say avoid it.
If it's cheap enough (each game is low cost) and wanna test it out, I'd say give it a shot. Just bet paper every time.
If it works keep playing. If you're losing after some X hands (and don't wanna lose more than $100 or w/e) then quit. Most you can lose is $100 and it can be a potential gold mone (maybe?).
Quote: NeutrinoI don't agree with this. I have a system of RNG in my head that I use when I play poker. Say if I calculate in a certain situation I should call 27% of the time and fold 73% of the time I actually have a way of generating that RNG. Feels pretty awkward to do at first but once I'm used to it it becomes somewhat natrual. Not sharing it though.
What do you mean by RNG? Like a number that's kinda sorta random that ends up not being random at all because you're applying a simple (or complex) formula to it? Or we talking a legitimate random number? I think there's more behind RNGs than you may think.
Quote: RSThat's pretty cool. So basically two separate programs battle it out? That's cool. :)
I took an AI class in school and we did this, but we were playing the game Fox and Geese instead of RPS. All the students had a tournament playing against each other, sometimes as the Fox and sometimes as the Geese.
I got shafted because at the beginning the rules were "you can't take more than X seconds/minutes per turn." Come day of the tournament, everybody's programs were slow as anything, so the teacher limited it to calculating Y moves in the future instead of by time. Mine was written in a way that it was very fast but needed to look at many moves in the future to determine the next turn :(
Quote: NeutrinoNot sharing it though.
If you ever change your mind, I'd be very interested to know your secret.
Personally, I've used the second hand on my watch as the basis of drawing random numbers at times.
Quote: WizardIf you ever change your mind, I'd be very interested to know your secret.
Personally, I've used the second hand on my watch as the basis of drawing random numbers at times.
I played poker at a table where a guy rolled dice before some decisions.
It was all for show, though. He was also wearing a ridiculous costume and trying very hard to project an image. He was not a very good player.
Quote: AxelWolfYou cant share it because it probably dose not work.
Lool. If that makes you feel better to say.
Quote: WizardIf you ever change your mind, I'd be very interested to know your secret.
Personally, I've used the second hand on my watch as the basis of drawing random numbers at times.
This should be fine for the *most* part. But once other people catch on to it it's exploitable. (i.e. wait 36 seconds so your decision would be to fold). But yeah, if I change my mind I'll let you know.
Quote: NeutrinoI don't agree with this. I have a system of RNG in my head that I use when I play poker. Say if I calculate in a certain situation I should call 27% of the time and fold 73% of the time I actually have a way of generating that RNG. Feels pretty awkward to do at first but once I'm used to it it becomes somewhat natrual. Not sharing it though.
Are you a pro? Otherwise it looks like another year where an amateur wins the Main Event. Thanks for saving a lot of people 10K!
Quote: NeutrinoI don't agree with this. I have a system of RNG in my head that I use when I play poker. Say if I calculate in a certain situation I should call 27% of the time and fold 73% of the time I actually have a way of generating that RNG. Feels pretty awkward to do at first but once I'm used to it it becomes somewhat natrual. Not sharing it though.
I think I could probably do that if I wanted to, at least, I can think of an easy enough way to do 10-20-30-40-50-60-70-80-90%'s.
Basically, I would just think of five two-digit numbers, which I can do faster than adding them, and then I would add them up after having thought of them. If I wanted a 70/30, for example, the last digit being an eight, nine or ten would be on the 30% side and the last digit being 1-7 would be on the 70% side.
Quote: Mission146I think I could probably do that if I wanted to, at least, I can think of an easy enough way to do 10-20-30-40-50-60-70-80-90%'s.
Basically, I would just think of five two-digit numbers, which I can do faster than adding them, and then I would add them up after having thought of them. If I wanted a 70/30, for example, the last digit being an eight, nine or ten would be on the 30% side and the last digit being 1-7 would be on the 70% side.
What's the point of the first digit of each number?
I'd bet (not much mind) that Mission's numbers he produced from this system were biased. I'd also wager no-one can be bothered to do the number of trials to find out :)
Quote: AxiomOfChoiceWhat's the point of the first digit of each number?
If I used single-digit numbers, I'd add them as quickly as I was thinking about them and could not help but do so.
You randomly think of the numbers to add up? If so I don't think you can randomly think of numbers to add up. You will tend to pick non random numbers. I'm sure this is good enough for poker or even RPS.Quote: Mission146If I used single-digit numbers, I'd add them as quickly as I was thinking about them and could not help but do so.
https://www.youtube.com/watch?v=PCZbp-hZfjo&feature=kp
Here it is in a nutshell:
After you lose: Throw what wasn't played. This is because the winner will have a better than 1/3 chance of repeating whatever he just played and won with. Playing whatever would beat that, or whatever wasn't played last time, is what will defeat him if he does.
After you win: Play whatever your opponent just played. This is because your opponent will want to switch things up and have a better than 1/3 chance to switch to whatever wasn't played. Playing whatever he just played will beat that.
In other words, rock paper scissors players play like gamblers in games like baccarat or roulette. They tend to stick with whatever way they are betting if they are winning and switch to the other side if losing. Just try to exploit this mentality in RPS.
A question for the forum is what should you play after a tie?
A disagreement I have with the video is that the strategy against a perfect randomizer is to randomize yourself. I submit that against a perfect randomizer, it doesn't make any difference what you do.
Must be a slow day, the Wiz just went back two years.
Must be sumptin' good here.
I'll read again, I'll be back (maybe ;-)
Unlike dice You can probably influence RSB.Quote: surrender88sJust wait until the RoShamBo Opponent Influencers hijack this thread...
Quote: Wizard
A question for the forum is what should you play after a tie?
A disagreement I have with the video is that the strategy against a perfect randomizer is to randomize yourself. I submit that against a perfect randomizer, it doesn't make any difference what you do.
I don't know what you would want to throw after a tie, unless you have experience with that opponent and tend to have an idea of what the opponent goes for after a tie. In my experience, people often seem hesitant to throw what would have lost to the thing that you tied with and tend to think that the other person is going to throw the same thing to try to, 'Trick,' them. Therefore, after a tie, I tend to throw what would have lost. For instance, if you tie with scissors, then my tendency is to throw paper thinking that my opponent will throw rock at least more often than anything else. Admittedly, I haven't even kept any Empirical data on that, so I'm going based on pure memory.
I agree with you, obviously, against a perfect randomizer. I submit that the law of large numbers dictates that the randomizer is going to throw everything closer and closer to 33.3~% of the time as one gets further and further into the long run, therefore, a non-random player could throw nothing but paper and it would make no difference.
Say something like "I'm going to rush you!" Then subconsciously your opponent thinks you're throwing rock, and he'll throw paper, to which you actually throw scissors and collect your winnings!Quote: AxelWolfUnlike dice You can probably influence RSB.
Ask him to come back here, and we can have a serious whack at the Riemann Hypothesis. https://www.youtube.com/watch?v=VTveQ1ndH1c . And put this forum on the map.Quote: Wizardteliot turned me onto a great set of math videos called numberfile.
I would play for tie all along to confuse 'em, have some fun with the dummies as well as the smarties.Quote: WizardA question for the forum is what should you play after a tie?
Intuitively, you have to at least believe that strategy, something which is universally open to anyone who seeks it, can't end part way through. That would imply some part of the universe is without purpose. Hawking seems to think that time wraps around like space. A waste of time.Quote: WizardA disagreement I have with the video is that the strategy against a perfect randomizer is to randomize yourself. I submit that against a perfect randomizer, it doesn't make any difference what you do.
And low and behold, if you think about this long enough, you find yourself going beyond the N=2 Surprise Paradox for the Reals, and into new strategies to beat the random stuff in the Naturals.
Hey, hey. Didn't see that one coming. Ouch. (Under the "you gotta be s****ing me" category.)Quote: TwoFeathersATLI'm not 100% sure, but I thought I read just read the best strategy was to just 'kick em in de balls' ;-?
This is supposed to be PG. I will not report you, and the mgt may miss your post. 'Not my job' so to speak.....Quote: surrender88sAnother method is to not change at all, countering your opponent's assumption that you will change. My friend had a signature sequence: rock - rock - rock - rock - rock. He called it the avalanche.
Quote: WizardAfter you win: Play whatever your opponent just played. This is because your opponent will want to switch things up and have a better than 1/3 chance to switch to whatever wasn't played. Playing whatever he just played will beat that.
If you win, shouldn't your strategy be: switch to whatever wasn't played? If you assume the opponent will switch too often, then you want to play whatever loses to whatever he just threw. Which, since you won, is whatever was unthrown.
Example: I throw scissors, you throw paper. I win! Now I think you are going to switch with a probability more than 2/3, but I don't know what you'll switch to, so I want to do well against the set {rock, scissors}, meaning I should go with rock.
Or another way to look at it, you lost with paper, so you're not going to throw it again, so I should be safest with rock.
Quote:A disagreement I have with the video is that the strategy against a perfect randomizer is to randomize yourself. I submit that against a perfect randomizer, it doesn't make any difference what you do.
Yes, you are definitely correct.
Quote: MrGoldenSunIf you win, shouldn't your strategy be: switch to whatever wasn't played? If you assume the opponent will switch too often, then you want to play whatever loses to whatever he just threw. Which, since you won, is whatever was unthrown.
I'm only quoting what was in the video. However, I can see how psychologically someone would change things up after a loss. So, if I just won scissors over paper, I could assume my opponent would either play rock or scissors. If I play paper, I'll have a 50/50 chance of winning or losing, for a net win of 0. However, if I play rock, I'll have a 50/50 chance of winning or a tie, for a net win of 0.5. Thus, switch to what wasn't played.
Quote:Example: I throw scissors, you throw paper. I win! Now I think you are going to switch with a probability more than 2/3, but I don't know what you'll switch to, so I want to do well against the set {rock, scissors}, meaning I should go with rock.
Or another way to look at it, you lost with paper, so you're not going to throw it again, so I should be safest with rock.
I don't see why the opponent would favor rock over scissors after a loss with paper.
Quote: DRichMy question is what do you throw on your first throw against an opponent you have zero knowledge of?
Paper!
Quote:I don't see why the opponent would favor rock over scissors after a loss with paper.
That is not what I am saying. My impression was that the assumption is if your opponent loses, they will switch with a greater probability than if they were random uniform every round.
If my opponent is going away from paper, then I should play rock. It seems you agree with this, but your prior post says that the video advises after I win, "play whatever your opponent just played," which in my example is paper.
That is what I am disputing. I contend that if my opponent is switching too often, but I don't know what they're switching to, I should play rock next round, not paper.
Quote: IbeatyouracesTherefore, my first throw will be scissors :-D
...or will it :-X
