Overswarm
is laughing at you
- Joined
- May 4, 2005
- Messages
- 21,181
I'm starting to think you don't know much about numbers unless they're given to you arbitrarily.
-
Welcome to Smashboards, the world's largest Super Smash Brothers community! Over 250,000 Smash Bros. fans from around the world have come to discuss these great games in over 19 million posts!
You are currently viewing our boards as a visitor. Click here to sign up right now and start on your path in the Smash community!
Official Metaknight Discussion
- Thread starter Hylian
- Start date
- Status
•Col•
Smash Champion
- Joined
- Nov 19, 2007
- Messages
- 2,450
Whoa, settle down there. I was just making a joke before... ._. I wasn't really trying to make any sort of point, other than really say to Nanaki "People have used the EDC in tournament matches before and gotten away with it, it's not that hard to do"
Although, I do believe it was more than once... If I remember correctly, there was at least one other video of him doing it.
You asked me where I heard/saw that. I go out of my way to get a nice little bit of evidence. Then you get back to me with a sarcastic, snide remark. Way to be an *******. :/
adumbrodeus
Smash Legend
Actually, I was sort of agreeing with him, simply averaging it draws bad conclusions.
The problem is, so does not averaging it.
Random sampling is amazing.
Red Arremer
Smash Legend
@Colaya: Well, I guess it's because you made it sound like he does it all the time.
Delvro
Smash Ace
Decreasing marginal benefit is a property that applies to ALL data with these specifications (Ankoku's and OS's (which is a subset of Ankoku's).
The conclusion is that:
for any percentage increase in popularity of a character in tournaments, the percentage increase in point value will be LESS than this increase in popularity. Always.
Exactly how much less depends on the data, but you can see from the "very simple" sample tournament that it is possible to be very much less (a 100% increase in MK popularity resulted in a 17% increase in point value).
Therefore, MK's dominance in 100+ man tournaments cannot be adequately explained by simply saying that MK is the most popular character.
Overswarm
is laughing at you
- Joined
- May 4, 2005
- Messages
- 21,181
FINALLY.
We're making progress now.
So according to Omni, you, you are okay with the following in the top 12:
2/3rds (8/12ths) of the top 12 having used Metaknight in tournament
2/3rds (8/12ths) of the top 12 having lost to at least one Metaknight
Eliminating the players beaten by the 3 obvious outliers (Mew2King, ADHD, Ally) removes ADHD, M2K, KSizzle, Shadow, Havok (giving 7 players left of the top 12)
5/7ths of the remaining top 12 having been beaten by a Metaknight
2/7ths of the remaining top 12 having been beaten by two Metaknights (technically 3, but Ally lost to M2K once so he only lost once in this example)
Looking at top 3:
3rd losing to two MKs, one of them getting second place and the other 4th
2nd losing to the same Diddy twice
1st losing no matches and beating second twice
So this is all perfectly acceptable for you, right?
Overswarm
is laughing at you
- Joined
- May 4, 2005
- Messages
- 21,181
Turbo Ether
Smash Master
- Joined
- Apr 12, 2006
- Messages
- 3,601
I can't wait to see the results of Viridian City 7.
It can't; however, we can't say MK is a character that warrants a ban because he is so popular.
Imagine you had 100 marbles of 5 different colors representing smashers of different skill levels. Red is terrible, Yellow is bad, Blue is average, Green is good, and Purple is amazing.
100 marbles selected at random are put into seperate bags representing characters.
MK's bag gets 40 marbles to show for his popularity
Snake's bag gets 10 marbles
Wario's bag gets 8 marbles
Falco's bag gets 6 marbles
Various other characters get 5 or less marbles in each bag to add up to 100
After all of the marbles are distributed in each of the bags, we take each bag and count up the purple marbles in each bag.
MK: 4 purple marbles
Snake: 1
Wario: 2
Falco: 0
Other character etc:
Probability states that MK's bag will probably have the most amount of purple marbles for the simple fact that it has the most marbles in it overall.
CRASHiC
Smash Hero
That assumes that those charecters won by chance and this is a random occuring, like seeing a random Olimar, a Tager/Bang, a Mewtwo, Zangeif, or Q. This is not that, this is consitent and persistent dominance at tournaments with the largest tournouts and the highest number of top Metaknights: WHOBO, Genesis, and now Pound 4.
Delvro
Smash Ace
If the assumptions you make are true, then yes, you would be exactly right. However, you are assuming that skill is normally distributed with respect to some mean (in your case, a blue marble).
Because of this, the likelihood of top level players is difficult to determine, because there are inter-region influences. If one purple marble mains metaknight, will that increase or decrease the likelihood of other purple marbles in the nearby region (because regions with one great player tend to have many great palyers) maining Metaknight? Unfortunately, I can't answer that question for certain. =[
adumbrodeus
Smash Legend
But is MK just 100% more popular at the top levels of play?
Regardless, the real issue is that since the question is, "how much?" not "is MK a dominant character?", we need a statistically valid way to track PRECISELY how much more powerful he is. And that data is useless for tracking that. Now if you were simply arguing that MK's results are at least partially due to him being the best character, I see no reason to disagree, but if you're saying that this shows him as dominant enough to be clearly banworthy (as in fitting as supermajority of smashboard's criterias), well that's what I was arguing against.
Especially when we consider that this tournament could easily be an unusual case (probably not a true outlier, but certainly pretty out there).
Considering they're the best players in the country playing the best characters in the country... yea, sure.
I think OS states what all pro-ban smashers should do more clearly than I could.
That only works if Smash was a game of probability which it isn't.
For instance I have almost never used MK in any serious competition so I'm not very good as him. So if I was one of these marbles there is no way I would end up in the "winning" bag. Because I suck. So you have to take away any sucky marbles like me because they don't matter.
But are they the best players in the country? Or are they just top level players using the best character?
edit:
My question is, how many other players that use other characters could be top MKs, and place like they do if they had used MK from day 1 instead of their character...
Delvro
Smash Ace
Unfortunately, we don't know for certain. The best we can do is estimate, unless someone is willing to put forth monumental effort in gathering this kind of data.
However, that doesn't mean that our estimations are necessarily bad, although they would probably be subject to some bias. This is where things get messy >_<
He was right to include them. He was taking a number of samples with size of sample based on popularity, which means a random sample all players of a character, good and bad =]
Putting yellow marbles into the MK bag doesn't make them purple ;-)
#HBC | Red Ryu
Red Fox Warrior
- Joined
- Jun 15, 2008
- Messages
- 27,486
- Location
- Milwaukee, Wisconsin
- NNID
- RedRyu_Smash
- 3DS FC
- 0344-9312-3352
Wait if MK was popular due to being good then why was Link so popular in melee?
Eh I guess I misunderstood the example. I thought the colors were representing tiers not player skill.
If MK is only doing so well because of popularity why didn't Link dominate Melee..?
Btw does anyone have the data that shows Link being the most common tourney character?
adumbrodeus
Smash Legend
Therein lies the problem.
But I say again, random sampling side-steps a lot of these issues.
Me and a friend are actually discussing a way to produce relevant data.
The real problem is it doesn't mean our estimations are good either.
#HBC | Red Ryu
Red Fox Warrior
- Joined
- Jun 15, 2008
- Messages
- 27,486
- Location
- Milwaukee, Wisconsin
- NNID
- RedRyu_Smash
- 3DS FC
- 0344-9312-3352
Someone said they had it in here. I remember a purple name.
Lmfao totally not a logical arguement.
Link is popular in the same way that Mario and Sonic are popular in brawl...at birthday parties in FFAs with items.
I'm gonna need a lot more than that to accept Link's popularity. Purple names sometimes spout out BS too.
This made me want to compare pigeons to dragons... Pigeons might be known for pooping everywhere, but dragons poop too.I'm gonna need a lot more than that to accept Link's popularity. Purple names sometimes spout out BS too.
Yeah, dragons don't exist, whatever.
#HBC | Red Ryu
Red Fox Warrior
- Joined
- Jun 15, 2008
- Messages
- 27,486
- Location
- Milwaukee, Wisconsin
- NNID
- RedRyu_Smash
- 3DS FC
- 0344-9312-3352
Look someone claimed Link was popular in Melee, even if he didn't show up a lot in top spots in Melee, I'm trying to figure out who said it.
Dragons do exist.This made me want to compare pigeons to dragons... Pigeons might be known for pooping everywhere, but dragons poop too.
Yeah, dragons don't exist, whatever.
Well you shouldn't just post arguments that you heard somewhere for something. That's how rumors start...
Komodo Dragons aren't winged so they don't poop like pigeons.
And the whole MK discussion seems to have dissolved yet again, now we're talking about non-MK-related stuff... I wonder, OS... Is there any talk at all in the SBR-B currently?
No I mean real dragons exist... Just because you haven't seen one doesn't mean they don't.Komodo Dragons aren't winged so they don't poop like pigeons.
And the whole MK discussion seems to have dissolved yet again, now we're talking about non-MK-related stuff... I wonder, OS... Is there any talk at all in the SBR-B currently?
And we're semi on topic here. The question is whether MK's dominance is simply because of saturation or if his saturation is a side effect of his dominance.
CRASHiC
Smash Hero
Link isn't really the most used in tournaments, however, what happened was every so often there would be a huge surge in Link players. This was a sign of when the new players entered the field, and when they died out, it was a sign that things had died down a bit and those Link players switched to viable mains.
Sort of like Lucario mains in Brawl, except they just picked up a Metaknight secondary
Sort of like Lucario mains in Brawl, except they just picked up a Metaknight secondary
When Brawl started out, everyone went to whoever they felt was gonna be their favorite (or the best) character. While people kept playing, they started to lose to Snake so they used him to win tourneys since he was the best in the game back then. Once MK started beating Snake overall, some Snake mainers (and the people who wanted to main the best character) changed to MK and also started dominating heavily. Skills kept improving from there til this day, MK's still the best in the game, and people who lose vs MK no matter how much time and effort they put into that one single MU compared to every other MU in the game, changed their mains to MK because it was easier than Snake/Diddy/Wario whose MUs aren't as even as MK:MK (50:50).
The more tourneys happened back then, the more MK won, and the more people realized that MK was in a league of his own. Remember MK being in SS Tier due to tourney rep? That was the time people realized MK was way too good compared to all the other characters. As the best MKs lost to Snakes/Diddies/Warios/Falcos, as bored mains quit the game because it was the same old thing (MK beating them even if they put in all their effort), as frustrated mains stopped attending tourneys and as the MK hate grew and grew, some people stopped using MK and moved onto the characters who apparently had "even MUs" with him. Our current top-level-tourney winners in the metagame are a Snake main, a Diddy main, and an MK main. One level below is a noticeable amount of MK mains, with a couple of random joes here and there (Falco, Wario, Snake, Diddy, IC, Marth).
Think about this... A large amount of top-players are using MK. We can say that there's a saturation of MK mains... Why is MK the most saturated character, and why are the other characters not as dominant as MK? In MY opinion, it's because MK is a far larger "tourney-placing modifier" than the 2nd best character. People are calling these mainers "top-players" without keeping in mind that MK might be the reason they're doing so well (which should be obvious, he IS the best in the game)... What if their skills are lower than the other mainers who struggle to maintain their top-player status (should be obvious)? It would then mean that their tourney placings are what they are because of MK, which should be okay if MK had some sort of exploitable weakness that doesn't revolve around theoretical assumptions and can be exploited consistently during a match! Snake has cooldown on some moves and can be gimped, Diddy's bane is for someone who can use items better than him to come around and beat him in both item and non-item usage, Falco suffers from his fastfall and below-average recovery, and so on... If MK had an actually exploitable weakness, would these people actually do this good in tourneys and be considered "top-level players"?
To me, you need the data to build views and arguments. But sometimes the data might eman something, but our eyes might show us something different. Believe it or not, calculators give out faulty info when you try to punch in tougher equations at higher-leveled Math problems, but how can this be if calculators give out calculations and is all about data? Sometimes our own eyes can tell us something's wrong with the picture, even if everything else points to it being perfect.
The more tourneys happened back then, the more MK won, and the more people realized that MK was in a league of his own. Remember MK being in SS Tier due to tourney rep? That was the time people realized MK was way too good compared to all the other characters. As the best MKs lost to Snakes/Diddies/Warios/Falcos, as bored mains quit the game because it was the same old thing (MK beating them even if they put in all their effort), as frustrated mains stopped attending tourneys and as the MK hate grew and grew, some people stopped using MK and moved onto the characters who apparently had "even MUs" with him. Our current top-level-tourney winners in the metagame are a Snake main, a Diddy main, and an MK main. One level below is a noticeable amount of MK mains, with a couple of random joes here and there (Falco, Wario, Snake, Diddy, IC, Marth).
Think about this... A large amount of top-players are using MK. We can say that there's a saturation of MK mains... Why is MK the most saturated character, and why are the other characters not as dominant as MK? In MY opinion, it's because MK is a far larger "tourney-placing modifier" than the 2nd best character. People are calling these mainers "top-players" without keeping in mind that MK might be the reason they're doing so well (which should be obvious, he IS the best in the game)... What if their skills are lower than the other mainers who struggle to maintain their top-player status (should be obvious)? It would then mean that their tourney placings are what they are because of MK, which should be okay if MK had some sort of exploitable weakness that doesn't revolve around theoretical assumptions and can be exploited consistently during a match! Snake has cooldown on some moves and can be gimped, Diddy's bane is for someone who can use items better than him to come around and beat him in both item and non-item usage, Falco suffers from his fastfall and below-average recovery, and so on... If MK had an actually exploitable weakness, would these people actually do this good in tourneys and be considered "top-level players"?
To me, you need the data to build views and arguments. But sometimes the data might eman something, but our eyes might show us something different. Believe it or not, calculators give out faulty info when you try to punch in tougher equations at higher-leveled Math problems, but how can this be if calculators give out calculations and is all about data? Sometimes our own eyes can tell us something's wrong with the picture, even if everything else points to it being perfect.
moyshe
Lazer Fox
I'm so tired of this constant complaining of banning MK. Honestly hes not even that broken. Yes I agree he is broken in some ways but I think snake is a lot more broken then MK. the MK match up is the first match up you should learn when picking up a character and just because you decided to not learn a match up doesn't mean hes broken, it just means your lazy and like to blame your loss in a tourney to something anyone can agree with. And playing meta on a higher level is a lot harder because people know how to actually play even if they never played you before, so you have to play meta differently or just not suck. I dont play meta but if hes banned I'll quit brawl and start playing melee or something because banning a character just because a bunch of kids are to lazy to learn a match up.
Nanaki
Smash Lord
I have a bit of a problem with this - we can statistically prove that MK is by far the best player in the game, sure. We can get a p<.001 scenario that proves, beyond a shadow of a doubt, that he's the best.
...does that matter at all?
People will still say 'yeah, he's the best, but every game has a best'.
The statistics can show he's the best, they can show he's 5 times better than everyone else, but he still won't be 'broken', and therefore still won't meet most people's 'objective criteria' (paradox much?) for a ban.
You can look at his moveset and rant about how his dtilt outranges this, and uair outspeeds that, etc., but will that ever lead to the conclusion that he's 'broken'?
The way these 'criteria' are being brought about makes zero sense to me. We say other communities "Just looked at the character and saw that he was broken", but we can't take a huge pile of tournament data, frame data, matchup discussion, and so on and make any kind of decision?
I think you need to shut the **** up. Seriously. Saying 'get better' 84839120583 times in a thread doesn't make you sound cool, it makes you a ****ing spammer. Contribute or shut the **** up.
But does putting green marbles in the MK bag make them as good as purple marbles?
Does putting purple marbles in the MK bag make them SUPER purple marbles?
Justblaze647
Smash Lord
And why is MK still doing so well? Are the other top players just not good enough to exploit this flaw, or does MK have ways of taking care of this flaw? I've never seen MKs jumping at you from across the stage, they always run and tay on the ground unless a jump is called for, and when they get close enough THEN they jump and attack. Tornado is a very good aerial plus an anti-air attack AND has GREAT aerial mobility.
That might seem like it's a bad weakness when compared to other characters... But when you factor everything else in like dash speed, walk speed, jump speed, acceleration, start-up on all attacks, endlag on all attacks, disjoints, priority, and aerial movement-altering maneuvers (glide, tornado, drill, downB)... His slow horizontal aerial movement isn't as exploitable as the data implies. The only character that I've seen that literally gives MK no choice but to take to the air earlier than he should is Diddy, and it's because his ground game is the best in the game.
- Status