Support NeoGAF

[Ahnez]

So basically you're saying that you're right and everyone one else is wrong. lol ok

No, im saying that saying something MUST be 100% right or 100% wrong is ridiculous

He got things right and things wrong, one right in 9 wrongs dont make a something true, and one wrong in 9 rights dont make something wrong

Mainly when we're talking about something that came from a human. one wrong in a million is inaceptable for a computer, but one wrong in ten is aceptable for a non-algorithmically being

Edit:

Giving every possible character a fair 50/50, its a combination of probabilities

If you get 6 characters right, it is 1/64 to pick the right ones.

But, you need to put the incorrect combinations on the calculation.

Lets say there are 7 characters to choose, just one incorrect, and the other 6 correct, its 1/128 to get those 7, and C(7,6), there are 7 combinations, just 1 right, so 1/7 to get the right one.

So, its already 1/896 to get.

Put 15~25 characters there, and it will be easily > 2^-32

 

[Juansho]

I've seen someone use one of these at a tourney:

Would be interesting to see someone use this one...

 

[Man God]

I'm cutting none of the 64 cast if I get to cut Brawl characters instead.

This is some crazy thought experiment where any of the original 12 were cut from the game, kinda like Street Fighter III.

 

[Odie_Esty]

This is some crazy thought experiment where any of the original 12 were cut from the game, kinda like Street Fighter III.

I...don't think I understand your point.

 

[Tripon]

Time for another flood of fanart pics.

Here's some of Palutena.

A chart of the sizes of Samus vs. her Rouge's Gallery, and more Samus fanart.

This pic will be useful for future discussions.

20XX has arrived.

Pit is rather conflicted on Palutena's making it into Smash 4.

Ness is still waiting for his invitation.

Pikachu cosplayer

Little Mac does not care for your trolling.

Lucina, Wario and Fox cosplaying as ZSS

And Snake is above all this.

 

[GreenLiquid]

No, im saying that saying something MUST be 100% right or 100% wrong is ridiculous

He got things right and things wrong, one right in 9 wrongs dont make a something true, and one wrong in 9 rights dont make something wrong

Mainly when we're talking about something that came from a human. one wrong in a million is inaceptable for a computer, but one wrong in ten is aceptable for a non-algorithmically being

I don't think this is quite what the argument over the leak's validity is about. To use statistical terms: there are basically two possible states of nature. Either what the leaker has said up until now, including the excuses for incorrect info is right, or what the leaker has been saying is false (there's also the possibility that has been suggested that only the first bit of leaked info was knowingly correct). So while wrong against nine rights doesn't invalidate the possibility of the things the leak says being true, it acts as evidence against the proposition that the state of nature is that the leak is true. If someone's required level of confidence to accept the leak as true is high enough, such a mistake could cause them to reject the leak as one would an unsubstantiated hypothesis.

Again, nothing against anyone who believes the leak, but I think the "90% right = 100% wrong" criticisms misses the actual cause of doubt in the leak -- (edit) which, I should clarify, isn't that what the leaker says (such as Shulk being confirmed) is false, but, rather, that the notion that one should be more likely to believe that Shulk is confirmed than they would otherwise because of the leak's existence is false.

 

[ffdgh]

Time for another flood of fanart pics.

Here's some of Palutena.

A chart of the sizes of Samus vs. her Rouge's Gallery, and more Samus fanart.

This pic will be useful for future discussions.

20XX has arrived.


Pit is rather conflicted on Palutena's making it into Smash 4.


Ness is still waiting for his invitation.

Pikachu cosplayer

Little Mac does not care for your trolling.


Lucina, Wario and Fox cosplaying as ZSS

And Snake is above all this.

Hmm...little mac's face reminds me of something.

 

[Videoneon]

No, im saying that saying something MUST be 100% right or 100% wrong is ridiculous

He got things right and things wrong, one right in 9 wrongs dont make a something true, and one wrong in 9 rights dont make something wrong

Mainly when we're talking about something that came from a human. one wrong in a million is inaceptable for a computer, but one wrong in ten is aceptable for a non-algorithmically being

See this is my problem with the Gematsu thing and I hope no one here takes any offense. First, on the surface loads of other "purported leakers" got things right and wrong, or maybe even mostly right. We have to presumably give more credit because he/she was the first to make validated E32013 guesses or something. As for the human element, I don't think that matters here - presumably the leaker isn't "predicting" anything or doing something that resembles the use of an algorithm, the person is just relaying information. More broadly, the point is the Gematsu leaker has incomplete information and this has been the case more than once.

Now I don't disagree that an absolute categorization like "right" or "wrong" is not helpful, but it's just, qualitatively, a "worse leak" compared to lupinko knocking out the entire MVC3 roster 100%. I mean, the Gematsu person makes a bad call in one case and in other cases retains his/her credibility ostensibly by omission. Oops, Rosalina. Oops, Lucina. I mean, come on.

Is it not telling that the leaker has all but disappeared since Shulk's not-reveal?

 

[NeonZ]

90% correct = 100% wrong

Gematsu leak had genuine info from some point in development, clearly some of his info was either outdated or misinformed (chrom instead of robin/lucina, maybe marshall instead of chorus men) but to completely deny him when he has gotten more right then wrong is silly.

"90% correct" only works if you assume that the later leaks have just as much validity as the first one just because they all supposedly come from the same source even though no one aside from Gematsu itself can verify that.

But, really, the only thing the second leak has gotten undoubtedly correct right now is Palutena's addition. Greninja only was foreseen as "Pokemon X/Y" and now Chrom is out - and, as far as we know, was never part of the development.

The excuse that it was from early on in the development clashes with the third and fourth "leaks" which claimed to have recent information, and yet didn't mention anything about changes. You're just closing your own eyes if you can't see that these later leaks have been rather spotty so far, in spite of the perfect first one.

 

[zroid]

oh dear god...

 

[Calvero]

Lucina, Wario and Fox cosplaying as ZSS

oooh can anyone make an avatar out of that Wario ZSS?

 

[afoni]

it's the weekend in my timezone how about you guys

 

[Dimitri LH]

Would be interesting to see someone use this one...

I feel proud that i won a MvC1 match on a DDR pad.....doubt its doable on smash though. >_>;

 

[superNESjoe]

Oh....I thought we settled this whole Gematsu argument last thread.

Ok then, best two out of three. I want a good clean debate bump here people.

 

[The12thGripper]

how did fox grow breasts?

 

[Makai]

I feel proud that i won a MvC1 match on a DDR pad.....doubt its doable on smash though. >_>;

I think I could do it if pressing the arrows makes me dash instead of walk.

 

[afoni]

how did fox grow breasts?

Medical science hit that point a long time ago.

Keep up.

 

[The12thGripper]

No, im saying that saying something MUST be 100% right or 100% wrong is ridiculous

He got things right and things wrong, one right in 9 wrongs dont make a something true, and one wrong in 9 rights dont make something wrong

Mainly when we're talking about something that came from a human. one wrong in a million is inaceptable for a computer, but one wrong in ten is aceptable for a non-algorithmically being

...Did you just continue to defend the second Gematsu rumor and not answer my questions?

 

[Ahnez]

...Did you just continue to defend the second Gematsu rumor and not answer my questions?

I didnt answer?

lol sorry, Im studying english just for 5~6 months, my understanding is very limited.

sorry

 

[BassForever]

"90% correct" only works if you assume that the later leaks have just as much validity as the first one just because they all supposedly come from the same source even though no one aside from Gematsu itself can verify that.

But, really, the only thing the second leak has gotten undoubtedly correct right now is Palutena's addition. Greninja only was foreseen as "Pokemon X/Y" and now Chrom is out - and, as far as we know, was never part of the development.

The excuse that it was from early on in the development clashes with the third and fourth "leaks" which claimed to have recent information, and yet didn't mention anything about changes. You're just closing your own eyes if you can't see that these later leaks have been rather spotty so far, in spite of the perfect first one.

Except we know it was x/y pokemon early in development straight from Sakurai. So that's 2/5, if anything the confirmation that greninja was known as x/y pokemon early in development shows that he had info from an early point in development, and things changed. It means his leak isn't 100% gospel, but it does give us a pretty good idea of at least two more newcomers

 

[Makai]

No, im saying that saying something MUST be 100% right or 100% wrong is ridiculous

He got things right and things wrong, one right in 9 wrongs dont make a something true, and one wrong in 9 rights dont make something wrong

Mainly when we're talking about something that came from a human. one wrong in a million is inaceptable for a computer, but one wrong in ten is aceptable for a non-algorithmically being

Edit:

Giving every possible character a fair 50/50, its a combination of probabilities

If you get 6 characters right, it is 1/64 to pick the right ones.

But, you need to put the incorrect combinations on the calculation.

Lets say there are 7 characters to choose, just one incorrect, and the other 6 correct, its 1/128 to get those 7, and C(7,6), there are 7 combinations, just 1 right, so 1/7 to get the right one.

So, its already 1/896 to get.

Put 15~25 characters there, and it will be easily > 2^-32

You're using probability inappropriately. P.S. I called 6 of the newcomers.

 

[zroid]

Keep up.

ಠ_ಠ

 

[GreenLiquid]

If every character has a 50/50 shot, then that means it's reasonable to conclude that about half of all characters ever will be in the new Smash. Probably not a reasonable premise.

 

[Makai]

If every character has a 50/50 shot, then that means it's reasonable to conclude that about half of all characters ever will be in the new Smash. Probably not a reasonable premise.

Yeah, exactly. The probability for newcomers is way below 0.5. Wii Fit Trainer is probably close to 0.0000

 

[theprodigy]

afoni is back, now the hype is real

 

[BassForever]

Yeah, exactly. The probability for newcomers is way below 0.5. Wii Fit Trainer is probably close to 0.0000

I don't think anyone is denying the first leak, I mean how can they its 6/6 and had the far more "never in a million years" picks ala wft and villager.

Its the second leak that's suspect, which is 2/5 with one character being flat out wrong in chrom

 

[zroid]

I'm still not clear on how one determines the "probability" of a certain character being playable in an upcoming Smash Bros. game.

Is there an equation for this?

 

[Makai]

I'm still not clear on how one determines the "probability" of a certain character being playable in an upcoming Smash Bros. game.

Is there an equation for this?

No, but if a betting market was opened for Smash Bros newcomers, we could have reasonable estimates.

 

[GreenLiquid]

I don't think anyone is denying the first leak, I mean how can they its 6/6 and had the far more "never in a million years" picks ala wft and villager.

Its the second leak that's suspect, which is 2/5 with one character being flat out wrong in chrom

I think this is the most reasonable conclusion, but I go one step further and cast doubt on the first leak for one simple reason: it wasn't presented as a leak, but as a set of "unlikely predictions"

True, it's nigh impossible for a single made-up leak to happen to correctly guess the things that the first Gematsu leak did. But the Gematsu leak, being presented only as predictions, wasn't special: many hundreds if not thousands of Smash fans around the Internet were hyped for E3 and probably making similar leaks [EDIT: predictions, my bad >.>]. The real question, to me, is not "what's the chance that Gematsu guy got it right by luck" but "what's the chance that anyone on the Internet got it right by luck?" I think the credibility drops quite a lot when that's considered.

 

[Ahnez]

Well, considering that almost anything from Sakurai is unpredictable, everything that comes from pattern analysis is pretty much worthless..

Well, at least it will be possible to verify those leaks in a month

 

[Boss Doggie]

oh god why is zero suit fox in gaf

 

[Storm Kyleis]

No, but if a betting market was opened for Smash Bros newcomers, we could have reasonable estimates.

This should be a thing.

 

[Monkey Blue]

Except we know it was x/y pokemon early in development straight from Sakurai. So that's 2/5, if anything the confirmation that greninja was known as x/y pokemon early in development shows that he had info from an early point in development, and things changed. It means his leak isn't 100% gospel, but it does give us a pretty good idea of at least two more newcomers

And on this 2 year old list that he carries around mii magically changed to Mii Fighter ONE day before e3 2014?
I'm not saying he doesn't have info, I'm saying he only had info from e3.
please don't talk about the 2 year old list of newcomers that he has but never mentioned or leaked info from until just before e3. It's the most absurd theory around.

 

[-Horizon-]

Time for another flood of fanart pics.

Here's some of Palutena.

A chart of the sizes of Samus vs. her Rouge's Gallery, and more Samus fanart.

This pic will be useful for future discussions.

20XX has arrived.


Pit is rather conflicted on Palutena's making it into Smash 4.


Ness is still waiting for his invitation.

Pikachu cosplayer

Little Mac does not care for your trolling.


Lucina, Wario and Fox cosplaying as ZSS

And Snake is above all this.

Awww yissss fan art. ZSS Lucina? All of my money. Palutena hug? Yes please.

 

[Freezie KO]

So here's a question: if you had to cut four characters from the original 12 and they could never be in another Smash game again, who would you choose?

I have to cut four characters from the original Smash brothers?! Please, no. Fuck it. Take my mother instead.

 

[The Great Gonzales]

oh god why is zero suit fox in gaf

An avatar quote would be too easy, so I won't do it.

 

[BassForever]

I think this is the most reasonable conclusion, but I go one step further and cast doubt on the first leak for one simple reason: it wasn't presented as a leak, but as a set of "unlikely predictions"

True, it's nigh impossible for a single made-up leak to happen to correctly guess the things that the first Gematsu leak did. But the Gematsu leak, being presented only as predictions, wasn't special: many hundreds if not thousands of Smash fans around the Internet were hyped for E3 and probably making similar leaks. The real question, to me, is not "what's the chance that Gematsu guy got it right by luck" but "what's the chance that anyone on the Internet got it right by luck?" I think the credibility drops quite a lot when that's considered.

He predicted Wii Fit Trainer and was the only person on the ENTIRE INTERNET to do so, predicted villager who sakurai had said "wasn't smash viable", predicted little mac before we saw the boxing ring after he had been assist trophies in brawl and had been considered for both melee and brawl but chosen against due to "all he can do is punch things", and predicted mega man who was DEAD at the time as Capcom was trying to bury the character.

The first leak is either some god tier prediction, or an honest to goodness leak. I'd say the probability of guessing all of those correctly is less then .01%

 

[Black-Wind]

I don't think anyone is denying the first leak

I am.
WFT is one of those "it makes perfect sense but Smash fans don't think about that" characters.

 

[GreenLiquid]

He predicted Wii Fit Trainer and was the only person on the ENTIRE INTERNET to do so, predicted villager who sakurai had said "wasn't smash viable", predicted little mac before we saw the boxing ring after he had been assist trophies in brawl and had been considered for both melee and brawl but chosen against due to "all he can do is punch things", and predicted mega man who was DEAD at the time as Capcom was trying to bury the character.

The first leak is either some god tier prediction, or an honest to goodness leak. I'd say the probability of guessing all of those correctly is less then .01%

I agree that it's still quite unlikely and honestly I think your reasoning that it's a leak is sound given your premises, but I think of those only the WFT and Villager reveals are particularly impressive, since there was a heck of a lot of Mega Man and Little Mac predicting going on at the time. But yeah, while it was in hindsight brilliant, WFT was a completely bizarre, off-the-wall choice. I'm glad it was made though!

 

[theprodigy]

No, but if a betting market was opened for Smash Bros newcomers, we could have reasonable estimates.

Nah, certain crazies would rig the odds on some characters.

 

[Tripon]

I am.
WFT is one of those "it makes perfect sense but Smash fans don't think about that" characters.

Uh, all 6 characters from that first leak was confirmed.

 

[GreenLiquid]

Nah, certain crazies would rig the odds on some characters.

I can see the inevitable Geno speculation bubble already. D:

 

[Beelzebufo]

I am.
WFT is one of those "it makes perfect sense but Smash fans don't think about that" characters.

Yeah, I'd agree with this. The same thing happened with Brawl. People thought "leaks" were legit because they listed characters like ROB and Toon Link, but they were obvious choices. He just got lucky, like the Gematsu "leaker".

 

[BassForever]

I am.
WFT is one of those "it makes perfect sense but Smash fans don't think about that" characters.

No, no it doesn't regardless of how much you want to try and rewrite history it's not a pick that would have ever made any kind of sense. It STILL doesn't really make sense despite it being a good pick in the end.

 

[Makai]

I think this is the most reasonable conclusion, but I go one step further and cast doubt on the first leak for one simple reason: it wasn't presented as a leak, but as a set of "unlikely predictions"

True, it's nigh impossible for a single made-up leak to happen to correctly guess the things that the first Gematsu leak did. But the Gematsu leak, being presented only as predictions, wasn't special: many hundreds if not thousands of Smash fans around the Internet were hyped for E3 and probably making similar leaks [EDIT: predictions, my bad >.>]. The real question, to me, is not "what's the chance that Gematsu guy got it right by luck" but "what's the chance that anyone on the Internet got it right by luck?" I think the credibility drops quite a lot when that's considered.

I think you have the right conclusion. This is an idea I have trouble explaining to people who put stock into seemingly extraordinary coincidences they notice. However, I believe in the Gematsu leaks. I think it is likelier than not that a large development team working on a closely followed game would have at least one person attempt to reveal its details to the public. I think Chrom's exclusion is due to changes in development, and the same could be true of both Shulk and Chorus Kids.

He predicted Wii Fit Trainer and was the only person on the ENTIRE INTERNET to do so, predicted villager who sakurai had said "wasn't smash viable", predicted little mac before we saw the boxing ring after he had been assist trophies in brawl and had been considered for both melee and brawl but chosen against due to "all he can do is punch things", and predicted mega man who was DEAD at the time as Capcom was trying to bury the character.

The first leak is either some god tier prediction, or an honest to goodness leak. I'd say the probability of guessing all of those correctly is less then .01%

I think it's closer to one in a billion.

 

[Green Mamba]

I am.
WFT is one of those "it makes perfect sense but Smash fans don't think about that" characters.

Have you ever heard of hindsight bias? Because you have it pretty strongly here.

 

[ShockingAlberto]

Gematsu leaker clearly had some valid information, but also filled in blanks himself with what he thought were shoo-ins and ended up with egg on his face because of it. That's what credibility is, it gets damaged when you're wrong.

Not sure why people have a problem with that.

 

[Makai]

Nah, certain crazies would rig the odds on some characters.

I can see the inevitable Geno speculation bubble already. D:

Sounds like an easy way for me to make some money, then

I wouldn't be the only one putting pressure in the other direction.

 

[-Horizon-]

Back to talking about leaks huh. We have two plumbers but never use them :/

Spoiler

where's that smash cycle chart?

I think WFT being leaked was legit. I do honestly think that Sakurai knew about the Shulk leak as well and changed up the reveal schedule to throw us off.

 

[Makai]

Yeah, I'd agree with this. The same thing happened with Brawl. People thought "leaks" were legit because they listed characters like ROB and Toon Link, but they were obvious choices. He just got lucky, like the Gematsu "leaker".

How is ROB an obvious choice? He's not even a shoo-in for Smash 4.