GameFAQs Contests

Contest Stats and Discussion - Part 881

UltimaterializerX | Posted 3/16/2010 3:41:37 PM | message detail

From: creativename | #450
Transivity means A>B>C.

Linearity is just the multiplication we use. x-stats. A gets 60% on B, B got 60% on C, so A gets 68% on C.


These both make sense, alright. So why do you say everything is transitive? You don't think Missingno's run this year (and several other runs in other contests) had enough variables not to be transitive? You could very easily redo this contest and get very different results for MN's matches.
---
·´¯`·.,¸¸,.·´¯`·.,¸¸,.~*ST*~.,¸¸,.·´¯`·.,¸¸,.·´`·
A wise man gets more use from his enemies than a fool from his friends.
tgs2 | Posted 3/16/2010 3:44:59 PM | message detail
This is an interesting result...
---
You may take Mr. Rogers down, but I will be back and become more powerful than you ever imagine
"tgs is taking it alone"
creativename | Posted 3/16/2010 3:49:54 PM | message detail

From: UltimaterializerX | #2251
These both make sense, alright. So why do you say everything is transitive? You don't think Missingno's run this year (and several other runs in other contests) had enough variables not to be transitive? You could very easily redo this contest and get very different results for MN's matches.


Missingo's run doesn't mean non-transitive. Transivity is pretty much a given.

Missingno is (presumably) non-linear, and so you can't extrapolate from him and get accurate %age predictions. But if Sephiroth beats Missingno and MN beats Crono, transitivity says Seph beats Crono. Or if random variation changed the result to Crono>MN, then Seph beats Crono, and thus Seph would beat MN. Transitivity remains in place.
---
gamefaqscontests.com
gamefaqscontests.com/gallery
nintendogirl1 | Posted 3/16/2010 3:55:45 PM | message detail
Best stats ever

http://thengamer.com/xstats/?name1=Guybrush&year1=2008c&name2=Lucas&year2=2008c
---
Have fun Albion.
nintendogirl1 | Posted 3/16/2010 3:59:33 PM | message detail
http://thengamer.com/xstats/?name1=Big%20Daddy&year1=2008c&name2=Tifa&year2=2008c

Whoo!
---
<.<
FFDragon | Posted 3/16/2010 4:01:08 PM | message detail

From: LeonhartFour | Posted: 3/16/2010 6:26:39 PM | #442
Well, we've got Cloud/Sephiroth and Cloud/Link for that!


That would be 48 hours straight, no?
---
If you wake up at a different time, in a different place, could you wake up as a different person? [HERO'S PLUNGE]
http://img.imgcake.com/wesker.gif
LOLContests | Posted 3/16/2010 4:02:10 PM | message detail
4ways aren't consistent from day to day, let alone year to year.

With the exception of things like LFF and Last Place Factor (which is mostly due to LFF/SFF anyway), they are fairly consistent, at least as much as 1v1s are. 1v1s aren't always consistent either, but since we don't have characters/games facing off in multiple rounds together we just have no opportunity to see it. Stuff like LFF exists in 1v1s as well, and low votals in SFF matches proves it, but it's just competley irrelevant since there's no third party to take advantage. It's not that the sprite round or other factors has more of an impact in 4 ways, but due to characters advancing together to the next round, we have a way to *prove* all if these outside factors.
---
"In the end, you have to choose whether you're going to live as a soldier, or just another man with a gun."-- The Boss
Yesmar
UltimaterializerX | Posted 3/16/2010 4:06:34 PM | message detail

From: nintendogirl1 | #454
Best stats ever

http://thengamer.com/xstats/?name1=Guybrush&year1=2008c&name2=Lucas&year2=2008c


Oh sweet, you finally finished this. Awesome job.
---
Has anyone really been far even as decided to use even go want to do look more like?
red sox 777 | Posted 3/16/2010 4:17:51 PM | message detail
I think MMX/Sonic would be very close, yes. Before the contest I was expecting Link to SFF Sonic, but after he failed to SFF Luigi more than mildly I've given up on that.

The variance in 4-way stats from round to round is higher than in 1v1, although they are generally good as long as you don't extrapolate through too many matches (that's a flaw in 1v1 stats too, but much less severe).

The thing I was talking about is not that though. Our most commonly used version of 4-way stats (pretending 2 of the characters in a poll don't exist) not only produces inconsistencies from day to day or round to round.....but on the very same day, in the very same match. Extrapolate through different characters, all in the same match, and you will get different results. Of course the errors are generally pretty small, so it's more or less okay.
---
2010 Extrapolated Standings: Crono gets 46.85% on Sephiroth.
All you need now is KH3 or SSB4, Crono. Then your quest shall be fulfilled!
Denzokuken | Posted 3/16/2010 4:23:10 PM | message detail
new picz

http://image.com.com/gamefaqs/images/cb8/cb8-121-1.jpg
red sox 777 | Posted 3/16/2010 4:25:57 PM | message detail
And here is an example to illustrate that:

Link 51.52% 74138
Zack Fair 22.09% 31794
Duke Nukem 14.45% 20797
Altair 11.94% 17184
TOTAL VOTES 143913

Zack gets 30.01% on Link.
Duke gets 39.54% on Zack.

So far so good. But.....

By extrapolation, Duke gets 39.54*30.01/50 = 23.72% on Link.
Actually, Duke got 21.91% on Link!

If we go another step, and extrapolate "Link" based on Duke, we get.....

"Link" gets 21.91/23.72*50 = 46.18% on Link

And we did it all in the same match.
---
2010 Extrapolated Standings: Crono gets 46.85% on Sephiroth.
All you need now is KH3 or SSB4, Crono. Then your quest shall be fulfilled!
UltimaterializerX | Posted 3/16/2010 4:27:07 PM | message detail
http://image.com.com/gamefaqs/images/cb8/cb8-121-1.jpg
http://image.com.com/gamefaqs/images/cb8/cb8-121-2.jpg
http://image.com.com/gamefaqs/images/cb8/cb8-121-3.jpg
http://image.com.com/gamefaqs/images/cb8/cb8-121-4.jpg
http://image.com.com/gamefaqs/images/cb8/cb8-121-5.jpg
http://image.com.com/gamefaqs/images/cb8/cb8-122-1.jpg
http://image.com.com/gamefaqs/images/cb8/cb8-122-2.jpg
http://image.com.com/gamefaqs/images/cb8/cb8-122-3.jpg
http://image.com.com/gamefaqs/images/cb8/cb8-122-4.jpg
http://image.com.com/gamefaqs/images/cb8/cb8-122-5.jpg
---
`·.,¸¸,.·´¯`·.,¸¸,.·´¯`·.,¸¸,.~*ST*~.,¸¸,.·´¯`·.,¸¸,.·´`·.,¸¸,.·´
"You are the king of derailing topics." -Lucid
Shoenin_Kakashi | Posted 3/16/2010 4:28:17 PM | message detail
http://image.com.com/gamefaqs/images/cb8/cb8-122-4.jpg

Black Charizard awesome


http://image.com.com/gamefaqs/images/cb8/cb8-121-1.jpg

Guy who does Berserk art/LTTP awesome.
---
Ah, you know it's funny, these people, they go to sleep. They think everything's fine, everything's good... They wake up the next day and they're on fire.
charmander6000 | Posted 3/16/2010 4:31:31 PM | message detail
http://image.com.com/gamefaqs/images/cb8/cb8-122-2.jpg

Hello Missingno
---
Board 8 BOP: http://www.freewebs.com/charmander6000/CB8%20BOP.xls
Character Battle VIII - 197/240 - Today's Winners: Solid Snake and Sephiroth
nintendogirl1 | Posted 3/16/2010 4:31:33 PM | message detail
http://image.com.com/gamefaqs/images/cb8/cb8-122-5.jpg

I vote we use this as the ONLY pic for that match.
---
:)
red sox 777 | Posted 3/16/2010 4:32:30 PM | message detail
In fact, once we have established that Link both equals Link and doesn't equal Link, we can make Link worth an arbitrarily low percentage on himself by simply extrapolating through Duke over and over. We could make Link weaker than Tanner without even using any other results!
---
2010 Extrapolated Standings: Crono gets 46.85% on Sephiroth.
All you need now is KH3 or SSB4, Crono. Then your quest shall be fulfilled!
UltimaterializerX | Posted 3/16/2010 4:34:29 PM | message detail

From: charmander6000 | #464
http://image.com.com/gamefaqs/images/cb8/cb8-122-2.jpg

Hello Missingno


LOL nice catch.
---
"GameFAQs isn't going to be merged in with GameSpot or any other site. We're not going to strip out the soul of the site." -CJayC
HaRRicH | Posted 3/16/2010 4:38:15 PM | message detail
http://image.com.com/gamefaqs/images/cb8/cb8-122-3.jpg

Got my first pic in this contest! I'm sad these two weren't selected though:

http://img695.imageshack.us/img695/7404/linkvssonicwereanimals.png
http://img251.imageshack.us/img251/3158/mariovscharizardbevel.png


C'este la vie, hopefully I feel inclined to make some other pics, but the rest of them just don't excite me so much right now.
---
Picking Auron > Squall was dumb, but not as dumb as thinking I could beat Leonhart!
Keep on smilin'! http://img718.imageshack.us/img718/8863/72571365.jpg
Gooper Blooper | Posted 3/16/2010 4:46:33 PM | message detail
I can't believe it! I finally broke through! The second Mario/Zard pic is mine!

And someone noticed who's helping Charizard yessssss =D
---
And then the car wash cost five dollars.
Pokemon Platinum: 1719 9497 5082
tgs2 | Posted 3/16/2010 4:46:55 PM | message detail

From: charmander6000 | #464
http://image.com.com/gamefaqs/images/cb8/cb8-122-2.jpg

Hello Missingno


It's like finding Waldo.
---
You may take Mr. Rogers down, but I will be back and become more powerful than you ever imagine
"tgs is taking it alone"
creativename | Posted 3/16/2010 4:53:19 PM | message detail
Yeah, what sox was talking about is a problem with the 4-way method we use, but I couldn't think of a better one and no one else has yet either. The problem is that the geometric relationship between characters are different than their relationship to the winner.

From: HaRRicH | #2308
http://image.com.com/gamefaqs/images/cb8/cb8-122-3.jpg

Got my first pic in this contest! I'm sad these two weren't selected though:

http://img695.imageshack.us/img695/7404/linkvssonicwereanimals.png
http://img251.imageshack.us/img251/3158/mariovscharizardbevel.png


C'este la vie, hopefully I feel inclined to make some other pics, but the rest of them just don't excite me so much right now.


Not vanilla enough! Don't you know this level of creativity cannot be tolerated!?
---
gamefaqscontests.com
gamefaqscontests.com/gallery
EternalxCourage | Posted 3/16/2010 4:54:31 PM | message detail
What the...last time I checked Sephiroth wasn't even at 55% now he's over 56%? What the.
---
The ultimate Snow fanboy - Curt
Yeah...it probably is. But I've got to do something, right?
EternalxCourage | Posted 3/16/2010 4:55:08 PM | message detail
Vote stuffers on Tifa's end?
---
The ultimate Snow fanboy - Curt
Yeah...it probably is. But I've got to do something, right?
Swifticuffs | Posted 3/16/2010 4:56:50 PM | message detail
so due to Sephiroth's performance is Snake now the new favorite or are we going to say this is clearly RSFF and TJF doing him in?

Because now even I think Snake has a damn good chance.
---
*~Swift~* Proud NY sports fan
all hail The Great King of Evil
PartOfYourWorld | Posted 3/16/2010 4:57:48 PM | message detail

From: Swifticuffs | #474
so due to Sephiroth's performance is Snake now the new favorite or are we going to say this is clearly RSFF and TJF doing him in?

Because now even I think Snake has a damn good chance.


Dunno if Snake is a favorite, but the people who have been calling Snake>Seph a ridiculous pick have been a lot quieter.
---
Yoblazer: http://i38.tinypic.com/2s024p4.jpg
Watch and you'll see... someday I'll be... part of your world!
EternalxCourage | Posted 3/16/2010 4:59:29 PM | message detail
Snake > Seph is a ridiculous pick.
---
The ultimate Snow fanboy - Curt
Yeah...it probably is. But I've got to do something, right?
TsunamiXXVIII | Posted 3/16/2010 5:01:11 PM | message detail
Ah, but transitivity can't always be implied, because there are matchups to consider. For a moment, let us use something outside of character battles to demonstrate.

http://espn.go.com/mens-college-basketball/tournament/2010/bracketcaster

If you go into the West region, you'll find that, quite counterintuitively, both 12-seed UTEP and 13-seed Murray State are slightly favored in the first round, and then if both do win, Murray State would be the slight favorite in Round 2. However, if UTEP is instead beaten by Butler in round 1, Butler would be a heavy favorite over Murray State; likewise, if Murray State lost to Vanderbilt in round 1, Vandy would be just as big a favorite over UTEP. The South region has a similar situation, with 12-seed Utah State a slight favorite over 5-seed Texas A&M, 4-seed Purdue a heavy favorite over Utah State in Round 2, and Texas A&M favored in Round 2 against Purdue if they do get past Utah State.

It's the same here. Outside of the 2006 Battle Royale, Mario's only match against Cloud was his win in 2002. Yes, it was probably a fluke, but maybe it wasn't. It would be interesting to see a Mario-Cloud rematch. Why bring these two up? Because we've seen time and time again that Mario falls apart when stuck in a poll with Sephiroth, and Sephiroth always eats the ass end of the SFF against Cloud. Maybe it was just a fluke that Mario won, or maybe, for whatever reason, Mario matches up better against Cloud than he does against Sephiroth.
---
CB8 score: 207
Incorrect picks: More than my signature will allow.
UltimaterializerX | Posted 3/16/2010 5:02:25 PM | message detail

From: HaRRicH | #468
http://image.com.com/gamefaqs/images/cb8/cb8-122-3.jpg

Got my first pic in this contest!


NICE, dude. I love 'mander and 'meleon getting into a pic.
---
"There are no fingerprints deep under water, nothing to tie one to a crime
And if you seek vengeance, all you need are instruments of pain." -Dethklok
UltimaterializerX | Posted 3/16/2010 5:03:40 PM | message detail

From: TsunamiXXVIII | #477
Ah, but transitivity can't always be implied, because there are matchups to consider. For a moment, let us use something outside of character battles to demonstrate.

http://espn.go.com/mens-college-basketball/tournament/2010/bracketcaster

If you go into the West region, you'll find that, quite counterintuitively, both 12-seed UTEP and 13-seed Murray State are slightly favored in the first round, and then if both do win, Murray State would be the slight favorite in Round 2. However, if UTEP is instead beaten by Butler in round 1, Butler would be a heavy favorite over Murray State; likewise, if Murray State lost to Vanderbilt in round 1, Vandy would be just as big a favorite over UTEP. The South region has a similar situation, with 12-seed Utah State a slight favorite over 5-seed Texas A&M, 4-seed Purdue a heavy favorite over Utah State in Round 2, and Texas A&M favored in Round 2 against Purdue if they do get past Utah State.

It's the same here. Outside of the 2006 Battle Royale, Mario's only match against Cloud was his win in 2002. Yes, it was probably a fluke, but maybe it wasn't. It would be interesting to see a Mario-Cloud rematch. Why bring these two up? Because we've seen time and time again that Mario falls apart when stuck in a poll with Sephiroth, and Sephiroth always eats the ass end of the SFF against Cloud. Maybe it was just a fluke that Mario won, or maybe, for whatever reason, Mario matches up better against Cloud than he does against Sephiroth.


heheh, nice post. My original bracket love was not GameFAQs, but the NCAAs. Loved it since I was a kid. Then GameFAQs combined that bracket with video games and I was in heaven. A perfect wedding.
---
¸¸,.·´¯`·.,¸¸,.·´¯`·.,¸¸,.~*ST*~.,¸¸,.·´¯`·.,¸¸,.·´`·.,¸¸
"Do you have ANY idea how low my self-esteem is?" -DYL
Calintares | Posted 3/16/2010 5:08:45 PM | message detail
http://thengamer.com/xstats/?name1=Tifa&year1=2008c&name2=Sephiroth&year2=2008c

Who would have though.

Also, NGamer, did you make the 2009 stats by keeping SSBM constant?
---
Ours is not to reason why, ours is but to do and die.
TheCodeisBosco | Posted 3/16/2010 5:08:47 PM | message detail
I've never considered Snake > Sephiroth impossible, but before today, I did consider it quite unlikely (even after Seph's unimpressive performance against Vincent).

Now, though? I definitely think Snake has a chance, especially since his performance on Squall was rock-solid. I still think Seph will pull through, but who really knows for sure at this point?

---
MORE CHARACTERS TO CONSIDER NOMINATING NEXT YEAR:
Psyduck, Kyle Hyde, Mallow, New Age Retro Hippie, Dan McNeely, Fulgore
Guess how many teens | Posted 3/16/2010 5:08:49 PM | message detail
What is the consensus on Mario vs Charizard? Upset probability ~15%?
---
[[[[[[[[[[[[[[[[[[[ -Red |||||||||||||| -Green ]]]]]]]]]]]]]]]]]]] -Blue
LordOfDabu | Posted 3/16/2010 5:10:09 PM | message detail
Voting trends are not transitive in general (this has a name: The Condorcet Paradox).
---
Watch me play games!
http://www.livestream.com/dabu
LordOfDabu | Posted 3/16/2010 5:12:21 PM | message detail
Here's an example:

Let's say 40% rank the characters A > B > C
20% rank them B > C > A
40% rank them C > A > B

Then if A goes against B, A wins with 80%.

If B goes against C, B wins with 60%

Transitive would mean that if A beats B and B beats C, then A beats C.

But if C goes against A, C also wins with 60%.

It's cyclic instead of transitive!

---
Watch me play games!
http://www.livestream.com/dabu
KanzarisKelshen | Posted 3/16/2010 5:14:52 PM | message detail

From: UltimaterializerX | Posted: 3/16/2010 8:02:25 PM | #478
NICE, dude. I love 'mander and 'meleon getting into a pic.


That pic is just ripe for poetic irony, though. If Mario rolls over 'Zard somehow, that "RRRAAAAAAGH SHAPED LIKE A FIREBREATHING DRAGON/Fireball 'whoops'" bit is gonna be a bittersweet sight.
---
Placeholder anti-capslock sig.
TheCodeisBosco | Posted 3/16/2010 5:14:58 PM | message detail
There's a chance of Charizard pulling off the upset, but it's a small one - probably more like 10% upset potential, if that. Samus > Cloud is probably more likely, to be honest.

---
MORE CHARACTERS TO CONSIDER NOMINATING NEXT YEAR:
Psyduck, Kyle Hyde, Mallow, New Age Retro Hippie, Dan McNeely, Fulgore
TheOneAboveAll | Posted 3/16/2010 5:15:44 PM | message detail
Cloud stood prostrate before Link, begging for mercy after his egregious error. Not only had he threatened to challenge Link, he had mistakenly brought up the Contest-That-Shall-Not-Be-Named. Link was considering not sending him to the level of the complete fodders, when the decision was abruptly made for him.
With a flash and a huge explosion, Cloud went flying out the window of Link's fortress.
"What is this?" Link wondered.
Into the room stormed Solid Snake and Samus.
"What was the meaning of that?" Link almost screamed in fury.
"We have come to inform you of your impending demise." Snake and Samus stated in unison.
"Oh?" Link almost laughed. "Do you not remember years past? Do you really think even the two of you together can defeat me?"
"Of course not." Samus replied.
"We are here to inform you of the one who will." Snake followed.
"It'sa me, Mario." A short red Italian plumber flew into the room.
"You? You will never come near defeating me!" Link fumed.
"Have you forgotten our deal already?" Mario countered. "I promised you dominance over this area as long as you allowed me to remain in the top tier. I understood that I had to be behind your henchmen Cloud and Sephiroth. However, you delayed three years in giving me dominance over my nemesis, Crono. I let that slide because of 2005. But last year, you allowed Crono to utterly embarrass me. Sonic the Hegehog will begin to show tonight that you reign is failing, and my followers Snake and Samus will defeat your henchmen to follow. Not even your pet pokemon can stop me. The days of Clinkeroth are over!!"

Let the quarterfinals begin!
KanzarisKelshen | Posted 3/16/2010 5:16:54 PM | message detail
...Oh god, Mario defeating Link at last would be the best thing ever. Now you got me hoping for an impossible upset!
---
Placeholder anti-capslock sig.
TheOneAboveAll | Posted 3/16/2010 5:23:28 PM | message detail
That was kinda the point. And it's really fun to think about. Especially because I think we all agree that Mario deserves to be king of the character battle world.
Shoud I repost this at the beginning of the next part so more people see it?
creativename | Posted 3/16/2010 5:24:29 PM | message detail
Comparing basketball to polls is obviously a ludicrous comparison. And we're talking intra-year. You can't compare different years obviously, that's silly.

These polls do not work like rock paper scissors. It's a pointless debate confused by confusion on term definition.
---
gamefaqscontests.com
gamefaqscontests.com/gallery
KanzarisKelshen | Posted 3/16/2010 5:25:26 PM | message detail

From: TheOneAboveAll | Posted: 3/16/2010 8:23:28 PM | #489
That was kinda the point. And it's really fun to think about. Especially because I think we all agree that Mario deserves to be king of the character battle world.
Shoud I repost this at the beginning of the next part so more people see it?


Yes, do it. More Mario love is never a bad thing.
---
Placeholder anti-capslock sig.
HaRRicH | Posted 3/16/2010 5:25:53 PM | message detail
Charizard's not beating Mario guys. I'm all for Charizard impressing and I think it'll mostly avoid SFF, but it'd be doing well to break 40%.
---
Picking Auron > Squall was dumb, but not as dumb as thinking I could beat Leonhart!
Keep on smilin'! http://img718.imageshack.us/img718/8863/72571365.jpg
LordOfDabu | Posted 3/16/2010 5:26:39 PM | message detail
Well, it's clear that such a thing would be pretty rare, but I would argue that we've already seen some kind of (Link weighted) version in Mario/Link/Samus
---
Watch me play games!
http://www.livestream.com/dabu
creativename | Posted 3/16/2010 5:28:32 PM | message detail
Again - the only possible scenarios for non-transitivity are a series of extremely close results, where say Crono beats Mario by 200 votes after Mario beat Snake by 200 votes. In this scenario Crono and Snake would pretty much be a toss-up, and non-transitivity is possible.

The other would be a case of rSFF where the result actually changed. But of course we've never actually seen rSFF until maybe today; it's been a theoretical thing. And Tifa isn't actually winning. So non-transitivity isn't ocurring here - although if this is rSFF, that would be non-linearity, since you couldn't extrapolate say Sephiroth vs. Big Daddy properly from this.
---
gamefaqscontests.com
gamefaqscontests.com/gallery
creativename | Posted 3/16/2010 5:29:58 PM | message detail

From: LordOfDabu | #2453
Well, it's clear that such a thing would be pretty rare, but I would argue that we've already seen some kind of (Link weighted) version in Mario/Link/Samus


Where would transitivity have been violated there...? Link beats both Mario and Samus.
---
gamefaqscontests.com
gamefaqscontests.com/gallery
TheCodeisBosco | Posted 3/16/2010 5:30:09 PM | message detail
Mario > Link would be the single greatest result in contest history. Even Kirby > Sonic, Starcraft > Halo, etc. wouldn't be able to top it.

---
MORE CHARACTERS TO CONSIDER NOMINATING NEXT YEAR:
Psyduck, Kyle Hyde, Mallow, New Age Retro Hippie, Dan McNeely, Fulgore
Pirateking2000 | Posted 3/16/2010 5:31:09 PM | message detail
Sonic>Link would be cooler, then we can have Sonic face Mario (assuming he beats Charizard)
---
XBL GT: Demon Ninja X2
KanzarisKelshen | Posted 3/16/2010 5:32:52 PM | message detail

From: Pirateking2000 | Posted: 3/16/2010 8:31:09 PM | #497
Sonic>Link would be cooler, then we can have Sonic face Mario (assuming he beats Charizard)


Sonic/Mario followed by Mario/Cloud would be the best contest finish ever, especially if Sonic hulked up and kept it close. Killer combo of contest and normal VG nostalgia.
---
Placeholder anti-capslock sig.
TheCodeisBosco | Posted 3/16/2010 5:32:58 PM | message detail
Sonic>Link would be cooler, then we can have Sonic face Mario (assuming he beats Charizard)

How terrible would it be if Sonic beat Link... and Mario lost to Charizard? That'd be the definition of "anticlimactic."

---
MORE CHARACTERS TO CONSIDER NOMINATING NEXT YEAR:
Psyduck, Kyle Hyde, Mallow, New Age Retro Hippie, Dan McNeely, Fulgore
transients | Posted 3/16/2010 5:33:41 PM | message detail
lol snake
---
xyzzy