I thought that 0% percent was maximally paradoxical. I selected the option that even I, myself, did not choose.
A fine warm-up for US elections this Fall, when one might ask the question: What percentage of the candidates on this ballot are not already pwnz0r3d?
"Well... Let's just call that a rounding error!" We are not (and do not want to be) the infamous corporation located in Redmond called Microsoft. We do not do <insert poor excuse here> errors and <insert lame excuse here> bugs.
I think you're just rationalising this whole thing like into something you've done on purpose! I think we're stuck with a very stupid and a very dismal looking poll!
I think we're stuck with a very stupid and a very dismal looking poll!
Funny, I thought it was an interesting and clever question despite its apparent simplicity. Anyone who actually thought about it would realise that answering "correctly" without cheating would not be straightforward, and would also have realised that the questions it raised were just the tip of a very large iceberg.
Such as; self-reinforcing behaviour based on the correct answer- or rather, possibly *not* self-reinforcing based on what people *thought* the correct answer would be. And whether that would end up being self-reinforcing anyway, or not.
Of course, there may be more than one correct answer; you could consider whether the categories chosen made it possible for *everyone* to be correct, and if so, what (various) pattern of voting would let everyone chose a correct answer.
So already we've got sociological, statistical and other mathematical questions from one simple poll.
But I'm sure others would rather vote on some obscure in-joke about Star Wars again.
This is a great game-theoretic question. Obviously it's possible for eveyone to be right by everyone just selecting "100%". The issue is that in order make a wise choice we have to have a theory of how others will choose--just as we do when deciding which highway to take to avoid traffic, or what day to go to the barber shop to avoid the wait, etc. Sometimes our reasoning works, other times the exact opposite of our intention comes about by the very fact that others reasoned the same way. Remember that old Yogiism: "Nobody goes to the ballpark anymore: it's too crowded."
For example, you could argue that 25-50% is a reasonable choice based on the number of possible choices, but if enough people buy into that argument then 25-50% will be too low. That may make you think that 50-75% is a better choice, but then you have to assume that 50-75% of the respondants will think like you. Any determination about which answer is "best" has to rely on some theory of how others will respond.
This is somewhat related to the prisoner's dilemma, as well as what happens in voting booths--it may be that the majority of a populace actually favors one candidate, but believe that that candidate has no chance of winning, so will vote for a less desirable candidate instead to avoid wasting their vote (hey guys, let's try not to do this this election season please;-p). Maybe pollsters should realize this and instead of asking people "Who would you vote for?", ask "If you could appoint the president yourself without having to worry about everyone else's choices, who would you choose?"
I picked 1-25%, and kinda peeved that my group was off (by 4%). Yes, this is definitely an excellent statistical/game theory question, and I should have been able to determine "the right answer". I figured #1, #6, & #7 combined HAD to make up a significant chunk of the answerees, because its popular to be contrarian here. (But I overestimated the numbers of #1 & #6, because I underestimated the intelligence of the audience to exclude those possibilities.) That left four choices, and random distrib
For example, you could argue that 25-50% is a reasonable choice based on the number of possible choices, but if enough people buy into that argument then 25-50% will be too low. That may make you think that 50-75% is a better choice, but then you have to assume that 50-75% of the respondants will think like you. Any determination about which answer is "best" has to rely on some theory of how others will respond.
It's never rational to pick 100%. Suppose you would win if you picked 100%. In that case, you would also win if you picked "under 25%." That's because if you had picked that, you would have made 100 wrong and "under 25" right, so you would have been (the only one to be) right. So in every situation where 100 wins, "under 25" also wins (so long as there are more than four players).
But the converse is not true: Sometimes, "under 25" would win where 100 loses. So the answer "under 25" dominates the answer "100", and it's never rational to say "100".
I remember in college our prof sat at his desk with a single 3x5 notecard and, with no other explanation, told everyone to pick a number between 1-5 and write it down. As we each read our number aloud (and showed it written down as proof), he tallied them on the board. If I remember right, 3 won overwhelmingly. Once the results were tallied, he flipped over his 3x5, which read "3 will win". He said he had done that every class for a decade and was never wrong. Apparently that's just how we think...1 is to close to the beginning to be safe, 2 is closer but "they'll be expecting that", 4 is too late in the game, and 5 is a loser. Think about that game when you pick a urinal, grab something from a line of objects, etc.
It's been years, though...it might have been 2. Cool trick, nonetheless.
You only think I guessed wrong! That's what's so funny! I switched glasses when your back was turned! Ha ha! You fool! You fell victim to one of the classic blunders! The most famous is never get involved in a land war in Asia, but only slightly less well-known is this: never go in against a Sicilian when death is on the line! Ha ha ha ha ha ha ha! Ha ha ha ha ha ha ha! Ha ha ha...
A more sophisticated contest entrant, wishing to maximize his chances of winning a prize, would think about what the majority perception of beauty is, and then make a selection based on some inference from his knowledge of public perceptions. This can be carried one step further to take into account the fact that other entrants would also be making their decision based on knowledge of public perceptions. Thus the strategy can be extended to the next order, and the next, and so on, at each level attempting to predict the eventual outcome of the process based on the reasoning of other rational agents.
" . . . entrant, wishing to maximize his chances of winning a prize, would think about what the majority perception of beauty is, and then make a selection based on some inference . . . "
Hell with that, I just looked at the results before choosing
Not quite. The choices aren't evenly distributed: 0%, 76-99%, and 100% choices aren't credible, so fewer people will vote for them. CowboyNeal usually doesn't pick up a huge number of votes.
The three remaining choices are unlikely to be evenly distributed, so it's a fairly good bet that one of them will get at least 25% of the vote. So lots of slashdotters (including myself) are picking the 26-50% option, a sort of self-fulfilling prophecy.
...and enough of the rest will choose the option lower than that, fulfilling two of the prophecies.
Don't forget: this isn't a mutually exclusive poll. Many multiple options could hypothetically have won. If one person voted for the Cowboy Neal option (just Cowboy Neal), nobody voted for 0%, and the 1-25% and 26-50% vote went as it has done...four of the seven options would be correct.
It's kind of an interesting exercise in game theory. You have a few objective facts (number of options, etc) and a bunch of people. How will they all act?
The facts alone won't cut it - given 7 options with even distribution, 1-25% is the obvious choice. But we're all (for the most part) pretty good at math here so we all know that, so if we all go on numbers alone that screws us. 100% of people would choose 1-25% and we'd all be wrong.
So we have to try to predict just how cooperative everyone will be. If we were totally cooperative, we'd all choose 100% - but if one person didn't choose that, it screws the rest of us. Most of us aren't that trusting.
I chose 26-50% because I figure about half of us at best will think that hard about this one, and I really hope at least a quarter of us do. At the moment we're at 28% that voted for that category.
Now for the unexpected dynamic - how may people will check out the results and vote based on them in order to be right? That could FSCK the numbers up something royal.
Just CowboyNeal I actively buy into slashdot mythology
But slashdot's technical audience (and relatively deep understanding of statistics) will throw that off. The 1-50% range will get disproportionately many votes. Therefore, 50-100% are the correct options.
No, because the stated question is: "How many people will select the same option as you?"
For example: for the 51-75% to be correct, 51-75% of the people voting, including you, would have to select 51-75%. There is no option that says "51%-75% of the people will select 26-50%". So if you select 51-75%, and it turns out in the end that 55% selects 26-50%, then you are still wrong because it will be physically impossible for more than 45% to select 51-75%.
No, it is not paradoxically correct, because if the option you selected is not labeled "51-75%", yet 51-75% do select it, then it fails one of the two conditions. The first condition is that you select that option, and the second condition is that some X percentage of people also select that option. It's only the correct option if both cases hold true. There is no "paradoxically correct" about it.
Your interpretation is reasonable if the trials were independent. But they aren't, and there is a strong semi-psychological strategic feedback element in the question. If everyone picked the "right" bread-and-butter statistics answer of 1-25% then the "right" answer would actually shift to 100%. But if everyone was paying attention they might assume other people would know that too, so everyone should pick 100%. But then again other people would assume not everyone would figure that out, and many people
It's way early for me, and I had to read this a couple of times before I realized you weren't talking about dipping Picaso's and Rembrant's into vats of hot oil...
Statistically I thought less than a quarter of people would choose the same as me, then I realised that most people here would know that, and so I went the next one up (currently at 28% or something so I was right:P ) This is probably the best quiz I've ever seen:D
Fun poll when you can read so much meaning into it.
0% - Nobody agrees with anything I say so why should this poll be any different 1-25% - 1 out of 7 choices equals 14%, so this is correct if people choose randomly 26-50% - WTF? Take me to the results 51-75% - WTF? Take me to the results 76% - 99% - Most people SHOULD agree with me because I'm right, dammit 100% - I am God and everyone agrees. Those who don't, don't count Cowboy Neal - I am really hurting for Karma and stroking Cowboy Neal's ego might help
Nope quite. I think this is probably going to be the most popular individual option, and one of two where the voter will have answered the "correctly", the other being 1-25%. The 0%, 100% and the Cowboy Neal option are already out of the running by virtue of them each having more than one vote (think about it). Factor in those who click at random and hit submit (typically a bell curve distribution) and those who took a more holistic view that enough voters would think 1-25% would be too popular and voted 25-50% as possibly being able to get enough votes to be right.
Then again, if 1-25% proves too popular, then there probably won't be enough votes spare to enable 25-50% to be right either, so my guess is we'll have one of two outcomes:
No options match their percentage range
Both the 1-25% & 25-50% options match their percentage range
Absolutely correct. I'm astonished that as I write this the 76-99% option has the lowest number of votes.
Of course, if everyone were perfectly rational, then 100% would be the correct choice, but I figured that there would be a small percentage of non-rational people (which would make 1-25% equally valid, but that seemed like a second-order result, so I plumped for the first-order 76-99% instead). Obviously, like you, I vastly overestimated the rationality of/. readers. Which is actually quite interesting, I suppose.
I wonder what Mr. Spock would have chosen. The 1-25% I suspect, since even though it's logically a derivative result, it's more robust in that there other, non-rational arguments that can lead to it being correct. Now I'm perilously close to rambling, though...
As more and more people pick the right choice, it becomes the wrong choice.
That's true for all of the choices except 100%, in the limit of an infinite number of voters. 100% will become the correct answer once everybody realizes that it's correct. Since not everybody will realize this, I voted for 76-99%.
Of course, once everybody realizes that 100% is the correct answer, then 0% is also correct, for one person.
Nice little lesson in probability. Except......IF 100% were correct, then 0% could not possibly be correct, as 100% represents the total number of entries, therefore that one individual that chooses 0% would invalidate the 100% crowd simply by participating.
UNLESS.....any of 3 conditions apply.
1)That one person who fails to select 100% also fails to participate in the survey altogether.
when people start picking the NEW right choice, the OLD right choice becomes correct again.
This is just a really BIG view of the Prisoner's Dilemma [wikipedia.org]. The correctness of your answer depends on the correctness of others.
I think it break down into several groups:
Realistic Pessimist - chooses 1-25%
Realistic Optimist - chooses 26-50%
Unrealistic Pessimist - chooses 0%
Unrealistic Optimist - chooses 51-74% or 75-99%
Idealistic Perfectionist - chooses 100%
Masochist - chooses CowboyNeal option
I am going to guess that the realistic pessimists and the realistic optimists will balance themselves correctly because of the nature of the audience of people who read/.
If this same poll were posted on Digg or as a Facebook Application I would bet on the unrealistic nature of the rest of the world to ruin any chance of anybody getting it right.
If Oprah were to hold the same poll on her website, I think her sheeple would vote 75-99% because they all mindlessly think the same way, but realize there are always going to be dissenters who they still need to brainwash into having their way of thinking.
There is only one option such that everyone chooses a correct answer (correct in the sense of being true): 100%.
*BZZT!*. Thank you, but that is the wrong answer. That kind of logic is what leads to software bugs that kill people. If 23% vote for 1-25%, 26% vote for 25-50% and 51% vote for 51-75%, then they'll all be right.
I have some notes about your lists... First I would say that choices aren't wrong per se, but people choosing that option can be wrong. So if 100% of people vote for 100%, everybody is right.
I prefer to understand your question with the precondition "there's at least one people having voted for each option"
Now, let's see:
Max satisfied at the same time:
What's asked isn't the final percentage, but the percentage of others voting the same. So one single vote for 0% satisfies the condition : nobody (0%) other th
Actually, if the 0% choice got exactly one vote, that one vote would be correct, since that one person said zero percent will select the same option as them, which means it is exclusive of that person. The moment zero percent had two votes, it's wrong forever.
There's 7 options meaning roughly 14% of people who pick options randomly will select your option. However, some people also reason this way, or in a similar manner, thus the options bellow 50% will get more votes than the options above it. Not knowing the psychology of slashdot poll voters I thus made a guess most peopel would vote for either the second or the third option, with some people voting for the other ones. The 2nd and 3rd options, if picked with equal preference, would thus probably get between 14% and 50% of the votes, which is consistent with either of them. The options above 50% for a single option seem unlikely because of people who vote randomly, and for the same reason the option bellow 14% seems unlikely. Hence I would expect the second and third options to get the most votes.
Yea, as a matter of fact I tend to be "that guy" at parties. How did you know ?
Hence I would expect the second and third options to get the most votes.
Although that's not what's being asked - if too many pick the options (as is in danger of happening with option 2), then you haven't chosen a correct answer.
There's 7 options meaning roughly 14% of people who pick options randomly will select your option. However, some people also reason this way, or in a similar manner, thus the options bellow 50% will get more votes than the options above it.
There are three realistic possibilities, for those who are trying to choose correctly.
The 0% option is a silly one, as identified by all of the comedians early in the comments tree.
76% through to 100% are highly unlikely, and the CowboyNeal option is... a CowboyNeal option.
That leaves the 1-25%, which would be correct if everybody chose randomly (as in my parent post), and may also be correct if people try to outsmart themselves (see below).
The 26-50% would be correct if everybody chose randomly between the three plausible categories (hence 33.33% each). By rights, it should be the most chosen by sentient/.ers (if there is such a thing), which could put it over 50%, but allowing for humourists and CowboyNeal fans, I'm confident it will stay a little below 50%. That's where I voted, and at this stage I'm correct.
The 51-75% is an outside chance. If there was a general consensus on one answer, the most popular answer would probably have 51-75% of votes. But if any single category does get over 50% of votes, I don't think it will be this one.
So really, that leaves two categories that can be seriously voted for. Allowing for leakage into nonsense categories, that should still keep the 26-50% category below 50%.
What does it mean if too many people select '25 - 50%'? E.g. what if 75% of people predict that 25 to 50% of voters will choose the same option?/gonna cut the grass and not think very hard about it.
7 choices, but 0%, 100%, and Just Cowboy Neal can be thrown out since those are discrete choices- it only takes 1 other person to make choosing one of those choices invalid. The given choices don't address what happens between the choices- zero point something, for example. Leaving that aside since we can't determine what to do there, we're left with 4 non-discrete choices: 1% - 25%, 26% - 50%, 51% - 75%, and 76% - 99%. There is a slight difference because the 4th of these choices stops at 99% (maybe), making it smaller than the other 3 choices. If we call these 4 choices roughly equal, then there should be an average of 25%. I don't remember if this would be a normal distribution or not, but it's like flipping a coin 1,000 times. On average, the coin should come up heads 500 times, but the likelihood of the coin coming up heads exactly 500 times is very small. This would seem to make 1% - 25% and 26% - 50% equally likely (roughly).
But how are people going to answer the poll? How many people always choose the Cowboy Neal option? How many people choose randomly (7 choices, 100/7, ~14.29%, 1% - 25%)? There are going to be people who will choose 0%, 100%, or Just Cowboy Neal, so this would nudge the relative balance between 1% - 25% and 26% - 50% in favor of 1% - 25%.
Just Cowboy Neal can be thrown out since those are discrete choices
Hmm... I voted for CowboyNeal, because the other options are guesses of what others will guess. The more people vote for _some_ of the options, the less likely they are to win. In contrast, the CowboyNeal option is one that counters logic, and I figured people would go for it precisely because it "isn't right", and it's not a function of time (or number of voters).
I voted 1 to 25 percent which currently has 21% of the vote and, thus, is also correct. Although, if too many more people are correct like me then we'll all be incorrect.
If CowboyNeal votes for anything but the CowboyNeal option, then the CowboyNeal option is wrong. And, if anybody besides CowboyNeal votes for the CowboyNeal option, and he does too, CowboyNeal is wrong in his choice.
The only valid answers are 1-25, 26-50, 51-75, 76-99.
Um, no.
The other three answers are entirely valid. They're IMPROBABLE, but quite valid.
Besides that, you just know that a handful of schmucks are gonna say 0 or 100 just to be dorks, or CowboyNeal because, hey, it's CowboyNeal!
I figured that based on that, the number of "sames" would be in the low 20 to mids, all things being average otherwise. I picked 1-25%, erring to the low side because of the other fringe options. I was close, at least:)
when i took the poll, i picked 26-50% and was right, b/c that selection had 29% at the time there are only 2 choices that are truly likely, 26-50%, and 51-75%, this was an intuitive choice...i post on slashdot quite abit, and i just guessed that between 26-75% of the voters would try to think out the options like myself
of these, i figured that 26-50% was the best b/c there were only 70 comments, so I knew the poll was new...the newness of the poll led me to think that fewer would choose like I did, between t
As will you do with the Rohrsach [tickle.com] test (also known as the inkblot test), to which any answer may be "wrong" or "right" more depending on the person asking the question than the person answering it.
Mostly answers to a Rohrsach test is kind of right if you say "butterfly", "Janus-face" or "Horsehead". If you on the other hand answer with references to promiscuous features you may be wrong.
Answers related purely to computers and similar may be a good confuser. "Looks like a squashed DB9...", "That's what My
Most people are drawn to 50%. I don't know why, they just are. Two of the answers shoot off from 50%. It is unlikely that more than 50% of people would actually choose one with more than 50%, so I chose 25-50%, to account for the smartasses who choose 0, 100, and Cowboy Neal, as well as the ones who would choose 1-25%.
I think this poll would be relatively the same if it were posted to Digg, or even a MADD, religious, or any other forum.
As others have mentioned, this poll is a game similar to the Prisoner's Dilemma [wikipedia.org], whereby the correctness of your own choice depends on the choices made by the other players.
The starting point for my reasoning was that, obviously, if everyone were perfectly rational and that fact was common knowledge [wikipedia.org] (in the technical, game-theoretic, logical sense), everyone would realize that whatever the correct answer was, everyone else (being perfectly rational) would select that option too, leaving 100% as the clear correct answer.
However, knowing that not all players in this game will even try to be rational - some will just be funny and choose 0% or Cowboyneal - that immediately eliminates 100% as a possibility. Still presuming MOST Slashdotters to be rational people though, I reasoned that most of them would also acknowledge the presence of some irrationality amongst us, and realize that everybody else realized that, and so most of us would choose 76-99%, which would thus make that option the correct answer.
But then, it dawned on me that there is likely a significant degree of randomness in Slashdot polls, both from people who just select at random and people who through some other line of reasoning selected a different option than that which I reasoned to be most reasonable, which would dilute the percentage of people who chose what I determined to be the reasonable answer. Once again imagining the remainder of Slashdot (excluding the jokesters, chaotes, etc) would reason along these same lines, I figured a good chunk of people would select the 51-75% option, so that's what I selected.
Where I seem to have failed is in accounting for all the people who would reason from the other end, assuming that the poll options would all get about equal distribution (thus 1-25%), and then realizing that some options (0%, 100%, Cowboyneal) were obvious losers, and thus that the concentration of people voting on the remaining options would actually be higher than if there were an even distribution (thus 26-50%). Of course, there remains the possibility that so many people will select those options, that they too will be false, and everyone will lose this game.
The lesson I've learned from all this? That it is unreasonable to reason from the assumption that other people are reasonable.
I had to vote for 100%, not because I thought it would happen, but in the unlikely event that everyone else voted for it, I didn't want to be the one person who didn't go along with it.
About 15 years ago, Scientifi American ran a contest (with prize money) that was simple in appearance: pick a number from 0 to 10^6. The number closest to 2/3 of the average wins. Then they published a paper analyzing the strategies. Very interesting.
Some interesting comments. The experiments were done in three papers, with numbers between 0 and 100. Results of all three experiments, and what the theoretically rational answer would be (everybody picking 0).
8: I have no life, and would prefer to discuss the merits of all options, rather than be outside enjoying the world.
Although, that option probably dropped out before the Iowa primary; it probably assumed, and rightly so, that the target market would never vote for it.
It is known that there is an infinite number of worlds, simply because there is an infinite amount of space for them to be in. However, not every one of them is inhabited. Therefore, there must be a finite number of inhabited worlds. Any finite number divided by infinity is as near to nothing as makes no odds, so the average population of all the planets in the Universe can be said to be zero. From this it follows that the population of the whole Universe is also zero, and that any people you meet from time to time are merely the products of a deranged imagination.
This poll is nothing but a product of a deranged imagination:-)
I know this was supposed to be funny, but that kind of argument has always bothered me. Not all infinities are equal. For example, there are an infinite number of natural numbers... only half of them are even... but there's still an infinite number of even natural numbers. I believe (though mathematicians feel free to correct me) that they're even both countably infinite, i.e. the SAME value of infinity, so even though only half of all natural numbers are even numbers, there are still just as many even natu
Seriously. Because this is the first time in months I've actually used my brain as a result of surfing the web. Sure it hurts, but it's all coming back to me now...
When I first encountered/. (must be close on ten years ago, I'd guess?), the running poll at the time was something like "Correct orientation for toilet paper: - under; - over". Actually, the goofiness of that poll what what made me join:)
Note that it's in principle possible to have as many as 5 options to be true at the same time: + 1 Vote at 0% + 23% (minus 2 votes) at 1-25% + 26% at 26-50% + 51% at 51-75% - 0% at 76-99% - 0% at 100% + just CowboyNeal at Just CowboyNeal.
Whoah (Score:2)
*went for 51-75%
Re:Whoah (Score:5, Funny)
Bet you feel stupid now (51-75% currently 19%)
Re:Whoah (Score:5, Insightful)
A fine warm-up for US elections this Fall, when one might ask the question: What percentage of the candidates on this ballot are not already pwnz0r3d?
Re:Whoah (Score:4, Funny)
Re: (Score:2, Funny)
We are not (and do not want to be) the infamous corporation located in Redmond called Microsoft.
We do not do <insert poor excuse here> errors and <insert lame excuse here> bugs.
Re:Whoah (Score:5, Funny)
Approximately 0 (Score:5, Funny)
Re: (Score:3, Funny)
Re:Whoah (Score:5, Interesting)
Such as; self-reinforcing behaviour based on the correct answer- or rather, possibly *not* self-reinforcing based on what people *thought* the correct answer would be. And whether that would end up being self-reinforcing anyway, or not.
Of course, there may be more than one correct answer; you could consider whether the categories chosen made it possible for *everyone* to be correct, and if so, what (various) pattern of voting would let everyone chose a correct answer.
So already we've got sociological, statistical and other mathematical questions from one simple poll.
But I'm sure others would rather vote on some obscure in-joke about Star Wars again.
Re:Whoah (Score:5, Interesting)
For example, you could argue that 25-50% is a reasonable choice based on the number of possible choices, but if enough people buy into that argument then 25-50% will be too low. That may make you think that 50-75% is a better choice, but then you have to assume that 50-75% of the respondants will think like you. Any determination about which answer is "best" has to rely on some theory of how others will respond.
This is somewhat related to the prisoner's dilemma, as well as what happens in voting booths--it may be that the majority of a populace actually favors one candidate, but believe that that candidate has no chance of winning, so will vote for a less desirable candidate instead to avoid wasting their vote (hey guys, let's try not to do this this election season please
Re: (Score:3, Funny)
I figured #1, #6, & #7 combined HAD to make up a significant chunk of the answerees, because its popular to be contrarian here. (But I overestimated the numbers of #1 & #6, because I underestimated the intelligence of the audience to exclude those possibilities.) That left four choices, and random distrib
Re:Whoah (Score:5, Funny)
The choice "under 25%" dominates the choice "100%" (Score:5, Interesting)
But the converse is not true: Sometimes, "under 25" would win where 100 loses. So the answer "under 25" dominates the answer "100", and it's never rational to say "100".
Re: (Score:3, Funny)
my fortune cookie said:
42.7% of statistics are made up on the spot.
that seems to work for me.
Re:Whoah (Score:4, Funny)
Re:Whoah (Score:4, Insightful)
Re:Whoah (Score:5, Interesting)
It's been years, though...it might have been 2. Cool trick, nonetheless.
Re: (Score:3, Funny)
Re: (Score:2)
Ha! Trick question: you cannot appear on the ballot unless you are already pwnz0r3d.
I selected 0% (Score:5, Funny)
Dammit!
Re:I selected 0% (Score:5, Informative)
A more sophisticated contest entrant, wishing to maximize his chances of winning a prize, would think about what the majority perception of beauty is, and then make a selection based on some inference from his knowledge of public perceptions. This can be carried one step further to take into account the fact that other entrants would also be making their decision based on knowledge of public perceptions. Thus the strategy can be extended to the next order, and the next, and so on, at each level attempting to predict the eventual outcome of the process based on the reasoning of other rational agents.
Re: (Score:2)
Re: (Score:3, Funny)
Re:I selected 0% (Score:5, Funny)
Re:I selected 0% (Score:5, Funny)
Hell with that, I just looked at the results before choosing
Superrationality (Score:4, Interesting)
http://en.wikipedia.org/wiki/Superrationality [wikipedia.org]
Interpretation (Score:5, Funny)
1-25% I understand statistics
26-50% I'm better than statistics
51-75% I don't need statistics
76-99% I'm an egomaniac
100% What am I voting for?
Just CowboyNeal I actively buy into slashdot mythology
Re:Interpretation (Score:5, Insightful)
The three remaining choices are unlikely to be evenly distributed, so it's a fairly good bet that one of them will get at least 25% of the vote. So lots of slashdotters (including myself) are picking the 26-50% option, a sort of self-fulfilling prophecy.
Re: (Score:2)
Don't forget: this isn't a mutually exclusive poll. Many multiple options could hypothetically have won. If one person voted for the Cowboy Neal option (just Cowboy Neal), nobody voted for 0%, and the 1-25% and 26-50% vote went as it has done
Game Theory (Score:5, Interesting)
The facts alone won't cut it - given 7 options with even distribution, 1-25% is the obvious choice. But we're all (for the most part) pretty good at math here so we all know that, so if we all go on numbers alone that screws us. 100% of people would choose 1-25% and we'd all be wrong.
So we have to try to predict just how cooperative everyone will be. If we were totally cooperative, we'd all choose 100% - but if one person didn't choose that, it screws the rest of us. Most of us aren't that trusting.
I chose 26-50% because I figure about half of us at best will think that hard about this one, and I really hope at least a quarter of us do. At the moment we're at 28% that voted for that category.
Now for the unexpected dynamic - how may people will check out the results and vote based on them in order to be right? That could FSCK the numbers up something royal.
Re: (Score:2)
But slashdot's technical audience (and relatively deep understanding of statistics) will throw that off. The 1-50% range will get disproportionately many votes. Therefore, 50-100% are the correct options.
Re: (Score:2, Redundant)
For example: for the 51-75% to be correct, 51-75% of the people voting, including you, would have to select 51-75%. There is no option that says "51%-75% of the people will select 26-50%". So if you select 51-75%, and it turns out in the end that 55% selects 26-50%, then you are still wrong because it will be physically impossible for more than 45% to select 51-75%.
Re: (Score:2)
Re:Interpretation (Score:5, Insightful)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
51-75% will get 35%.
I voted for 26-50%.
Re:Interpretation (Score:5, Insightful)
Re: (Score:2)
I'm rather hoping it isn't a question of either. I'm a statistician working in psychology, and I went for 51-75%
Re: (Score:3)
Re: (Score:3, Funny)
It's way early for me, and I had to read this a couple of times before I realized you weren't talking about dipping Picaso's and Rembrant's into vats of hot oil...
Re: (Score:2)
Re: (Score:2)
http://www.google.com/search?q=%22they+told+me%22+bush+re-elected+site%3Ainstapundit.com&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a [google.com]
Re: (Score:3, Insightful)
Re: (Score:3, Funny)
Bizarre (Score:4, Interesting)
0% - Nobody agrees with anything I say so why should this poll be any different
1-25% - 1 out of 7 choices equals 14%, so this is correct if people choose randomly
26-50% - WTF? Take me to the results
51-75% - WTF? Take me to the results
76% - 99% - Most people SHOULD agree with me because I'm right, dammit
100% - I am God and everyone agrees. Those who don't, don't count
Cowboy Neal - I am really hurting for Karma and stroking Cowboy Neal's ego might help
Re:Bizarre (Score:5, Insightful)
Nope quite. I think this is probably going to be the most popular individual option, and one of two where the voter will have answered the "correctly", the other being 1-25%. The 0%, 100% and the Cowboy Neal option are already out of the running by virtue of them each having more than one vote (think about it). Factor in those who click at random and hit submit (typically a bell curve distribution) and those who took a more holistic view that enough voters would think 1-25% would be too popular and voted 25-50% as possibly being able to get enough votes to be right.
Then again, if 1-25% proves too popular, then there probably won't be enough votes spare to enable 25-50% to be right either, so my guess is we'll have one of two outcomes:
Re:Bizarre (Score:5, Insightful)
Assuming that most voters are rational:
If 1-25% was the most rational option then most voters would pick this, making it wrong. So it can't be the most rational option.
That line of reasoning led me to pick 76-99%. Of course, if less than 76% of
Re:Bizarre (Score:5, Insightful)
Of course, if everyone were perfectly rational, then 100% would be the correct choice, but I figured that there would be a small percentage of non-rational people (which would make 1-25% equally valid, but that seemed like a second-order result, so I plumped for the first-order 76-99% instead). Obviously, like you, I vastly overestimated the rationality of
I wonder what Mr. Spock would have chosen. The 1-25% I suspect, since even though it's logically a derivative result, it's more robust in that there other, non-rational arguments that can lead to it being correct. Now I'm perilously close to rambling, though...
Re:Bizarre (Score:5, Insightful)
if I vote for the outcome 1-25% then:
for any result where 76-99 would be correct, 1-25 would also be correct. And there exists outcomes where 1-25 is correct, but 76-99 is not.
So the outcome where 76-99 is correct, is a propper subset of the outcomes where 1-25 are correct, if I vote for 1-25%
Can I Change My Vote? (Score:5, Interesting)
However, when people start picking the NEW right choice, the OLD right choice becomes correct again.
Re: (Score:2, Insightful)
Of course, once everybody realizes that 100% is the correct answer, then 0% is also correct, for one person.
Re: (Score:2, Interesting)
UNLESS.....any of 3 conditions apply.
1)That one person who fails to select 100% also fails to participate in the survey altogether.
2)Our wonderful survey masters opt to invok
Re:Can I Change My Vote? (Score:5, Interesting)
This is just a really BIG view of the Prisoner's Dilemma [wikipedia.org]. The correctness of your answer depends on the correctness of others.
I think it break down into several groups:
I am going to guess that the realistic pessimists and the realistic optimists will balance themselves correctly because of the nature of the audience of people who read /.
If this same poll were posted on Digg or as a Facebook Application I would bet on the unrealistic nature of the rest of the world to ruin any chance of anybody getting it right.
If Oprah were to hold the same poll on her website, I think her sheeple would vote 75-99% because they all mindlessly think the same way, but realize there are always going to be dissenters who they still need to brainwash into having their way of thinking.
Re:Can I Change My Vote? (Score:5, Funny)
Re: (Score:2)
Re: (Score:3, Interesting)
Re:Can I Change My Vote? (Score:5, Interesting)
So one question is: how many votes can be correct at the same time? I say five:
Another question: can all the choices be wrong at once? Yep:
Great poll!
Re: (Score:3, Interesting)
First I would say that choices aren't wrong per se, but people choosing that option can be wrong.
So if 100% of people vote for 100%, everybody is right.
I prefer to understand your question with the precondition "there's at least one people having voted for each option"
Now, let's see
Max satisfied at the same time
What's asked isn't the final percentage, but the percentage of others voting the same.
So one single vote for 0% satisfies the condition : nobody (0%) other th
Re: (Score:3, Insightful)
Reasoning (Score:3, Interesting)
Yea, as a matter of fact I tend to be "that guy" at parties. How did you know ?
Re: (Score:2, Insightful)
Although that's not what's being asked - if too many pick the options (as is in danger of happening with option 2), then you haven't chosen a correct answer.
Re:Reasoning (Score:5, Insightful)
There are three realistic possibilities, for those who are trying to choose correctly.
The 0% option is a silly one, as identified by all of the comedians early in the comments tree.
76% through to 100% are highly unlikely, and the CowboyNeal option is... a CowboyNeal option.
That leaves the 1-25%, which would be correct if everybody chose randomly (as in my parent post), and may also be correct if people try to outsmart themselves (see below).
The 26-50% would be correct if everybody chose randomly between the three plausible categories (hence 33.33% each). By rights, it should be the most chosen by sentient /.ers (if there is such a thing), which could put it over 50%, but allowing for humourists and CowboyNeal fans, I'm confident it will stay a little below 50%. That's where I voted, and at this stage I'm correct.
The 51-75% is an outside chance. If there was a general consensus on one answer, the most popular answer would probably have 51-75% of votes. But if any single category does get over 50% of votes, I don't think it will be this one.
So really, that leaves two categories that can be seriously voted for. Allowing for leakage into nonsense categories, that should still keep the 26-50% category below 50%.
trivail, only 100% has some sense (Score:2)
Results would be interesting to see over time (Score:2)
Hello, I'm a Statistician (Score:5, Funny)
Hmmm... (Score:2)
Most Likely Answer (Score:3, Informative)
But how are people going to answer the poll? How many people always choose the Cowboy Neal option? How many people choose randomly (7 choices, 100/7, ~14.29%, 1% - 25%)? There are going to be people who will choose 0%, 100%, or Just Cowboy Neal, so this would nudge the relative balance between 1% - 25% and 26% - 50% in favor of 1% - 25%.
I'd say my money's on 1% - 25%.
Re: (Score:2)
Just Goes To Show... (Score:2)
In this one, for example, it's trivial to show that over half the respondents are simply wrong.
(I voted "26%-50%", so at least I'm correct.)
Interestingly (Score:3, Insightful)
My head hurts.
Cheers,
Dave
Oblig. missing option (Score:5, Funny)
yup, it is the... (Score:2)
I win! (Score:2)
What did CowboyNeal vote for? (Score:2)
Re: (Score:2)
The poll is flawed!
Easy (Score:2)
The only valid answers are 1-25, 26-50, 51-75, 76-99.
Odds are that people will trend toward the middle (psychology), but lower because of the randomness. 26-50 must win the day!
Re: (Score:2)
Um, no.
The other three answers are entirely valid. They're IMPROBABLE, but quite valid.
Besides that, you just know that a handful of schmucks are gonna say 0 or 100 just to be dorks, or CowboyNeal because, hey, it's CowboyNeal!
I figured that based on that, the number of "sames" would be in the low 20 to mids, all things being average otherwise. I picked 1-25%, erring to the low side because of the other fringe options. I was close, at least
best answer changes over time (Score:2)
there are only 2 choices that are truly likely, 26-50%, and 51-75%, this was an intuitive choice...i post on slashdot quite abit, and i just guessed that between 26-75% of the voters would try to think out the options like myself
of these, i figured that 26-50% was the best b/c there were only 70 comments, so I knew the poll was new...the newness of the poll led me to think that fewer would choose like I did, between t
You will always fail at this test. (Score:2)
Mostly answers to a Rohrsach test is kind of right if you say "butterfly", "Janus-face" or "Horsehead". If you on the other hand answer with references to promiscuous features you may be wrong.
Answers related purely to computers and similar may be a good confuser. "Looks like a squashed DB9...", "That's what My
Wow... Two accurate answers! (Score:2)
But 26-50% right now has 28%, so anyone that voted there is also accurate as well!
100% had the theoretical possibility of being accurate, but, this being Slashdot, obviously wasn't.
0% is just for smart asses.
51-75% and 76-99% are both again theoretically possible, but highly doubtful.
Pick the middle option. (Score:2)
I think this poll would be relatively the same if it were posted to Digg, or even a MADD, religious, or any other forum.
Next Week, On Slashdot: (Score:2)
Why I Picked 51-75% (Score:5, Interesting)
The starting point for my reasoning was that, obviously, if everyone were perfectly rational and that fact was common knowledge [wikipedia.org] (in the technical, game-theoretic, logical sense), everyone would realize that whatever the correct answer was, everyone else (being perfectly rational) would select that option too, leaving 100% as the clear correct answer.
However, knowing that not all players in this game will even try to be rational - some will just be funny and choose 0% or Cowboyneal - that immediately eliminates 100% as a possibility. Still presuming MOST Slashdotters to be rational people though, I reasoned that most of them would also acknowledge the presence of some irrationality amongst us, and realize that everybody else realized that, and so most of us would choose 76-99%, which would thus make that option the correct answer.
But then, it dawned on me that there is likely a significant degree of randomness in Slashdot polls, both from people who just select at random and people who through some other line of reasoning selected a different option than that which I reasoned to be most reasonable, which would dilute the percentage of people who chose what I determined to be the reasonable answer. Once again imagining the remainder of Slashdot (excluding the jokesters, chaotes, etc) would reason along these same lines, I figured a good chunk of people would select the 51-75% option, so that's what I selected.
Where I seem to have failed is in accounting for all the people who would reason from the other end, assuming that the poll options would all get about equal distribution (thus 1-25%), and then realizing that some options (0%, 100%, Cowboyneal) were obvious losers, and thus that the concentration of people voting on the remaining options would actually be higher than if there were an even distribution (thus 26-50%). Of course, there remains the possibility that so many people will select those options, that they too will be false, and everyone will lose this game.
The lesson I've learned from all this? That it is unreasonable to reason from the assumption that other people are reasonable.
My philosophy (Score:3, Interesting)
Reminds me of a SA contest (Score:4, Interesting)
Re:Reminds me of a SA contest (Score:4, Informative)
Yay, found a link: http://www.jstor.org/view/00028282/sp030006/03x0205w/0 [jstor.org].
Some interesting comments. The experiments were done in three papers, with numbers between 0 and 100. Results of all three experiments, and what the theoretically rational answer would be (everybody picking 0).
Forgotten 8th option (Score:5, Funny)
Although, that option probably dropped out before the Iowa primary; it probably assumed, and rightly so, that the target market would never vote for it.
The Right Answer is 0 (Score:3, Funny)
It is known that there is an infinite number of worlds, simply because there is an infinite amount of space for them to be in. However, not every one of them is inhabited. Therefore, there must be a finite number of inhabited worlds. Any finite number divided by infinity is as near to nothing as makes no odds, so the average population of all the planets in the Universe can be said to be zero. From this it follows that the population of the whole Universe is also zero, and that any people you meet from time to time are merely the products of a deranged imagination.
This poll is nothing but a product of a deranged imagination:-)
Re: (Score:3, Interesting)
Best. Poll. Ever. (Score:4, Interesting)
Re:Wow (Score:5, Funny)
Re: (Score:3, Informative)
Re: (Score:2)
Welcome to
Re:Wow (Score:5, Insightful)
Re:Wow (Score:5, Interesting)
Re:Wow (Score:5, Funny)
Re: (Score:2, Interesting)
+ 1 Vote at 0%
+ 23% (minus 2 votes) at 1-25%
+ 26% at 26-50%
+ 51% at 51-75%
- 0% at 76-99%
- 0% at 100%
+ just CowboyNeal at Just CowboyNeal.
Re: (Score:2)
Re: (Score:2)