That Kefka match is an interesting case to make though.
---
"I dont know what I am talking about? I know more about gaming than 90 percent of the users on the internet." -xmagxus

so I completely agree with Leon about it being silly to bash the stats regarding this.


But since the stats have been off almost every match, I think there is plenty of reason to bash them. They have lost al credibility.
---
I HAVE FURY!

I wouldnt say every match. Really the only bad examples are Riku/Frog, Vivi/Zelda, Magus/Knuckles, Luigi/KOS-MOS, and Ocelot/Pac-Man. But thats 5 examples too much.
---
Leon's Quest For Ultimate Glory:
1. Gordon Freeman; 2. Mega Man

Well, again, we can't blame the stats for this match, as pac-man was simply under-rated.
---
"I dont know what I am talking about? I know more about gaming than 90 percent of the users on the internet." -xmagxus

But since the stats have been off almost every match

The stats are off every now and then, and people start claiming that they don't EVER work. Gimme a break.
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

Well we can blame the Spring stats, Pac-Man has never been as strong as Ocelot was in the Spring.
---
Leon's Quest For Ultimate Glory:
1. Gordon Freeman; 2. Mega Man

I long for the days when we could go three hours without people calling the stats into question.

when the contest is over, we'll have a new set of x-stats that will have been accurate for this contest. I look forward to those, if only to compare them to previous results.
---
http://picturesofwalls.com

It's funny how the s-stats are bein' ripped on so hard (harder than ever, I believe)...yet, just you wait, so many of the same ones are goin' to look at 'em next year because of how badly we've screwed up this year.
---
Hillary Clinton: We need to treat violent video games the way we treat tobacco or alcohol.
Lewis Black: Namely as valued contributors to our re-election fund.

I long for the days when we could go three hours without people calling the stats into question.

Hey, don't worry. On days when they're right or reasonably close (such as yesterday), nobody says a thing, positive or negative.
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

Well, you've got to expect the stats to be off for a lot of matches, and it shows, thanks to the SFF riddled 2004 bracket. With so many things off even at the base, it's hard to determine any changes a character gets in strength, and thus can surprise even more when a decent bracket rolls around again.
---
"i started playing mario brothers as luigi and threw the controller down in disgust in the middle of the game because he lacked character"--gokujont

Well I am just wondering if we cant redo the formula a bit. Maybe factor in vote totals/pictures/characters that dont change like others (Frog).
---
Leon's Quest For Ultimate Glory:
1. Gordon Freeman; 2. Mega Man

That would be really hard to do. I don't see how it's feasible--most of it would just be guessing how much of an impact certain factors have.
---
"i started playing mario brothers as luigi and threw the controller down in disgust in the middle of the game because he lacked character"--gokujont

Maybe factor in vote totals/pictures/characters that dont change like others (Frog).

Those things are virtually impossible to account for. It'd be a worthless venture even to try, and it'd screw up the stats even more.
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

It would be guesswork, but its better than what we are doing now apparently.
---
Leon's Quest For Ultimate Glory:
1. Gordon Freeman; 2. Mega Man

The 2004 stats are a lot of guesswork, too. We need to stop trying to calculate things that the stats were never meant to account for.
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

Let's look at the stats, shall we?



Mario/Joanna - N/A
Ness/C.J. - C.J. weaker than Jak? I don't think so.
Zero/Ryu H. - Stats were off
Lloyd/Wesker - Stats were WAY off. Lloyd weaker than Mithos?

Samus/Yuri - N/A
Riku/Frog - Riku gets 47% on Snake?
Ganon/Yuna - Stats were off
Auron/Boss - N/A

Snake/Manny - N/A
Vivi/Zelda - The stats WAY off again
Alucard/Kratos - N/A
Sora/47 - N/A

Dream division is N/A, although Rikku is on par with Dante? I doubt it.

Master Chief/CATS - Stats were off
D.K./Sam - Off
Tommy/Kefka - Off
Crono/Zidane - N/A

Devil division is N/A... except for one


Magus/Knux - WAY off

Tifa/Vyse - OFF. Tifa is NOT beating Samus.
Luigi/KOS-MOS - WAY OFF. AGAIN. WHAT WILL IT TAKE FOR YOU IDIOTS TO SEE THEY ARE FAILING BIG TIME?

MM/Conker - WOW, FIRST MATCH THE STATS CALLED! HAIL THE STATS!
Gordon/Leon - Way off. Leon is stronger than Jill
Yoshi/Laharl - Off
Today - Off


Spring Contest - Nuff said


Wow, so the only match the stats have predicted right is MM/Conker.


---
I HAVE FURY!

I see. You're expecting them to be right to the exact percentage, apparently, and you're basing strength on your own personal expectations (such as CJ > Jak). No wonder you're disappointed.
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

there's nothing wrong with the formula at all. you just have to know which numbers to trust and which numbers not to. Ocelot is one of those - he's based off uber-Bowser from the Villain Contest, and he likely overperformed on Sephiroth. trying to account for stuff like pictures numerically is hopeless.
---
http://picturesofwalls.com

Then you've got to call into question just how MUCH said factors had, and we have very little evidence on how much match pictures actually affect a match. Considering the vote-totals is good, but I don't want to see that combined with the x-stats themselves; I'd rather check them out separately, then compare the x-stats, and decide for myself.


I know CJay aimed to make a great bracket (and, obviously, so far so good)...but does anybody else think even he had expected the hell this has caused, especially after just the first round? I can't imagine he actually believed Lloyd/Wesker and Luigi/KOS-MOS to turn out the way they did, for example. Those two matches are good examples of hell raised by us in the stats topic and the board both...though Magus/Knuckles still takes #1 right now.
---
Hillary Clinton: We need to treat violent video games the way we treat tobacco or alcohol.
Lewis Black: Namely as valued contributors to our re-election fund.

They just have been off a few percent Leonhart, they've been 10-15% off several times now.


---
I HAVE FURY!

And not to mention, the stats actually were able to calculate some of those matches based on other performances (Zero/Hayabusa and Magus/Knuckles, the one you LOVE to bring up).
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

They just have been off a few percent Leonhart, they've been 10-15% off several times now.

Once. Luigi/KOS-MOS. That's it.
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

Wow, xstats suck this year... Only one match right? Ulti's logic is really true now.

And yeah, I voted for Pac-man dispite my bracketing of Ocelot. Everyone else who wasn't gutsy enough to call this upset will fail with me anyway.

Though, there are some reasons I'm sad Pac-man wins, but there are many more that I'm happy he won.

I was looking forward to voting for Yoshi over Ocelot next round. Also, Yoshi vs. Pac-man isn't going to be fun to watch, as one of my favorites beats a character who is also one of my favorites.... I like Yoshi more , but still.

But I am very happy Pac-man could pull this win out. His first win since he doubled Kyo in '02. I don't feel much pity for Ocelot here either, Pac-man deserves it.
---
Vote for Yoshi
You have no chance to survive make your time.

cant believe they put ocelot vs pac-man...
---
Mada Mada Dane
Joined 7/21/2002 4:40:02 PM

Wow, xstats suck this year... Only one match right? Ulti's logic is really true now.

*sighs*
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

I don't agree with most of those, but I take issue with...

Lloyd/Wesker: I would bet dollars to donuts Lloyd not only beats Mithos, but scores more on Kefka.

Zelda/Vivi: No excuse for this. Either Zelda massively increased or Vivi was overrated, but the stats never saw a 60/40 coming.

Magus/Knuckles: Self-explanatory. Only Sam Fisher was the prophet of doom.

Luigi/KOS-MOS: Again, while Luigi may have been underrated or KOS-MOS may have dropped, the stats saw this coming.

The other matches people have showed that some characters were likely going to be overrated (Frog/MC, for instance), so I don't exactly fault the stats for that. But there's just some things they can't and won't predict.
---
"i started playing mario brothers as luigi and threw the controller down in disgust in the middle of the game because he lacked character"--gokujont

Mario/Joanna - N/A
Ness/C.J. - C.J. weaker than Jak? I don't think so.
Zero/Ryu H. - Stats were off
Lloyd/Wesker - Stats were WAY off. Lloyd weaker than Mithos?

Samus/Yuri - N/A
Riku/Frog - Riku gets 47% on Snake?
Ganon/Yuna - Stats were off
Auron/Boss - N/A

Snake/Manny - N/A
Vivi/Zelda - The stats WAY off again
Alucard/Kratos - N/A
Sora/47 - N/A

Dream division is N/A, although Rikku is on par with Dante? I doubt it.

Master Chief/CATS - Stats were off
D.K./Sam - Off
Tommy/Kefka - Off
Crono/Zidane - N/A

Devil division is N/A... except for one


Magus/Knux - WAY off

Tifa/Vyse - OFF. Tifa is NOT beating Samus.
Luigi/KOS-MOS - WAY OFF. AGAIN. WHAT WILL IT TAKE FOR YOU IDIOTS TO SEE THEY ARE FAILING BIG TIME?

MM/Conker - WOW, FIRST MATCH THE STATS CALLED! HAIL THE STATS!
Gordon/Leon - Way off. Leon is stronger than Jill
Yoshi/Laharl - Off
Today - Off

All the ones I bold, are actually big WTF for you. I mean, in some of thems you're just saying nonsense (like Tifa =/= Samus, Rikku not being that strong, Leon not being that weak) and some the others were off by a few percentages (like Yoshi/Laharl) so I think you're looking at the stats the wrong way.
---
That's right. Cloud N. Candy would get a NEGATIVE percentage, and RPGuy's extrapolated calculator would crash. - therealmnm

*never saw this coming, rather for Luigi/KOS-MOS
---
"i started playing mario brothers as luigi and threw the controller down in disgust in the middle of the game because he lacked character"--gokujont

Where in the hell are you people getting defenses for Zelda/Vivi from?
---
"i started playing mario brothers as luigi and threw the controller down in disgust in the middle of the game because he lacked character"--gokujont

Leon, don't even bother. arguing with someone who's dead set on attacking the stats is useless. it's funny how the people who attack the x-stats read them much more literally than the people who embrace them - for example, trying to calculate a match down to the exact percentage.

at this rate, we're gonna need two topics - one for discussion and one for stats.
---
http://picturesofwalls.com

Wow, xstats suck this year... Only one match right? Ulti's logic is really true now.

*sighs*

Uh, right, I take back that statement...
---
Vote for Yoshi
You have no chance to survive make your time.

Zero/Ryu was pretty much right if you use unadjusted Zero.

Lloyd being weaker than Mithos? Sure. If Zero blows Lloyd away, that even more assures it.

Riku would get less than 47%, as Frog overperformed on Snake anyway, do keep that in mind.

As for Vivi/Zelda, how can you tell? Zelda simply gained.

Rikku isn't on-par this year's Dante, either. I think Dante is a bit stronger than last year.

As for MC/CATS, CATS had a face-pic, and MC was over-rated as much as Frog, hell, that trio was all over-rated. Ganondorf through Spring would barely triple CATS, it was unreasonable to see MC doing it too.

Magus/Knuckles, Magus was overrated, the stats themselves indicated that.

As for Luigi/KOS-MOS, Luigi was probably under-rated. I can't see KOS-MOS dropping to Pac-Man, but Pac-man going up a little under KOS-MOS, seeing that it's reasonable anyway.

And the rest is just your personal thoughts..

---
"I dont know what I am talking about? I know more about gaming than 90 percent of the users on the internet." -xmagxus

There's really no proper defense for Vivi/Zelda until she faces Snake. At the moment, it looks like she benefitted from the big increase Link and Ganon got, but other than that, we can really only speculate.
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

You know what I'm noticing? Everybody was SOOOO quick to throw away and piss on the x-stats for the Villains Contest because of how crazy it seemed. Now, lots of people're wantin' to throw away the stats we've had for the past several summers and even just stats in general, though the stats have generally worked well the past few years...

...so, I'm bringin' this up again: maybe there's a big change in the voter demographic this year like there was in 2k3, due to it both being more recent and us having high vote totals again. However, what I don't think many people will rally behind, is that alot of this change of demographic can be blamed on the merge with GameSpot; the boards and user-names were put together this Christmas, afterall. I'm not sayin' this is the case for sure (we saw how MC did), but it could still be the reason (we saw how Luigi did).
---
Hillary Clinton: We need to treat violent video games the way we treat tobacco or alcohol.
Lewis Black: Namely as valued contributors to our re-election fund.

Lloyd being weaker than Mithos? Sure. If Zero blows Lloyd away, that even more assures it.

You're delusional. And the rest of what you're saying only reinforces my point. People aren't complaining about the stats, they're complaining about the interpretation of the stats. It's like Slow says--they're more than a guideline, they're a puzzle that needs to be solved. When you start putting the pieces together in the wrong places, either not adjusting for SFF or pic factors or increases or decreases or overrating or underrating, that is a failure of the stats to do what they are supposed to do--which is predict the outcome of the match to a reasonable degree. And it can't do that all the time, which is why you have to take them with a grain of salt and make sure to compensate for other factors. But if you don't, expect to get burned.
---
"i started playing mario brothers as luigi and threw the controller down in disgust in the middle of the game because he lacked character"--gokujont

I don't believe in stats, I believe in the power of my heart!
---
Yoshifan scorecard: Next Match: Bowser Vs. Chun-Li: 86.58%
Points:Contest:14/14 Oracle:609.14/9th place

And at this point, this match is pretty much over. I don't see Ocelot being able to come from behind now.
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

I think there's a change in the demographic every year, as there's always a new "in" game and those are the people who are visiting gamefaqs. most people don't visit the site every day. also, the x-stats are based off one result, so the standard error is huge. if you calculated a standard deviation, it would likely be pretty large.
---
http://picturesofwalls.com

Doesn't Leonhart have Tidus winning his division?
---
I HAVE FURY!

What does it matter if I do?
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

Kefka shouldve won, Magus shouldve won, Gordon shouldve won, and now Pacman will finally make up for those 3 that lost.

Obviously, you just let bias and not logic influence your picks.

---
I HAVE FURY!

Hey, I have Tidus taking his division, against all odds and comprising the biggest long shot of my entire bracket. But, hey--I like to live dangerously.
---
"i started playing mario brothers as luigi and threw the controller down in disgust in the middle of the game because he lacked character"--gokujont

Not sure about him (I think he has Ryu, though I know he likes Tidus)...I know I do though. It's lookin' more and more like I'm gonna get burnt by Kirby though.
---
Hillary Clinton: We need to treat violent video games the way we treat tobacco or alcohol.
Lewis Black: Namely as valued contributors to our re-election fund.

What does this have to do with anything?
---
It's just not the same without this name.
Faithful supporter of Squall Leonhart in SC2K5

The stats to me, really is a huge puzzle, as each poll does have a life of it's own. So I do agree with you on that for the most part..
---
"I dont know what I am talking about? I know more about gaming than 90 percent of the users on the internet." -xmagxus

Knuckles > Squall
---
Knuckles the Echidna's road to the Sc2k5 Title!
Round 2: (2)Squall Leonhart

The first round >>> us all, stat-lovers and stat-haters alike.
---
Hillary Clinton: We need to treat violent video games the way we treat tobacco or alcohol.
Lewis Black: Namely as valued contributors to our re-election fund.

KNUX IS TEH PWNZOR!
---
Knuckles the Echidna's road to the Sc2k5 Title!
Round 2: (2)Squall Leonhart

[This message was deleted at the request of a moderator or administrator]