On Brawl Tiers

Hello Smashboards,

This post is intended to inform the Smashboards community of some major mathematical and scientific flaws in the creation of a formal “Tier” list, as previously used in melee and currently used in many competitive fighting games. Before critiquing some of the techniques, I want to ensure the readers that in no way am I claiming that all characters are equal, or that this is a perfectly balanced game-- however, I do want to bring up some potential confounds in the methods of data analyses used. My primary reason in addressing this issue is that debates and vehement claims are used regarding the use of such data without much methodological consideration regarding how these numbers were identified. In addition, I hope to inform the competitive community of better and more scientifically sophisticated methods of analyzing data.

The primary source of tier lists are basic statistics that anybody can do with excel (i.e. measures of central tendency, t tests, and regression). However these relatively “simple” methods of data analysis are also very vulnerable to threats to the internal and external validity of these studies. When considering the appropriate method of data analysis you must consider the method in which the data is collected. In all fighting tiers the majority of data is collected via passive observational studies in which tournaments are held and data is collected based on winners and competitors in both matchups and in general. This method of data collection does not account for a large quantity of other factors that contribute to the “win” and the “lose”. A coupe examples are as follows: the popularity of the character will greatly skew mean averages of matchup and tournment wins; different characters generally attract different players and play styles; sampling errors (i.e. fluke loses or lucky wins); regression to the mean (i.e. if more people play certain characters and few people play others the mean of good players from the more widely played character will be matched with the smaller sample of other players); and tournament site characteristics.

Given these limitations, and many others, it is highly inappropriate to make causal inferences based on this data. More appropriate methods of data analysis may include logistic regression, multilevel modeling or structural equation modeling which require statistics software such as Mplus, SYSTAT, or SAS. The use of such statistics could more accurately estimate the so called “tier” status of character qualities. If anyone is interested in these statistics or how too apply these to this game feel free to contact me. Below are some wiki links for basic understanding of how these statistics work, and the extent to the inferences that can be drawn from them.


http://en.wikipedia.org/wiki/Structural_equation_modeling

http://en.wikipedia.org/wiki/Multilevel_models

http://en.wikipedia.org/wiki/Logistic_regression

Yes... and???

However, some characters are simply 'stacked.' Good players intuitively are able to find what makes a character better than another. They then flock to these characters, and use them. Though Tiers may not the 'Scientific Fact' they are 'Smash Fact' which is good enough for the competitive nature of the game.

Read the post again-- The intent of the post is an offer (or suggest) a more appropriate approach to tiering characters, rather than using "intuitive" evidence. Granted, some players are stacked in some games-- the main fallback argument in most of the post i have seen in fighting game boards-- which will inevitably happen here-- is that the tiers are the golden rule. Intuitive evidence (as opposed to empirical) is highly dependent on the observer and likely to be influenced by personal bias.

Since when did super computers frequent Smashboards?

good job, i bet you feel amazingly smart for noticing that tiers cause popularity and that can cause more winning...that isn't like the #1 argument against tiers or anything....

and what kind of smart person links them talking about math stuff to WIKIPEDIA!!! send them to a math website...

did you just get out of a stat class or something and went on a power trip?

EDIT: on further analysis of the argument, i have realized that there is no reason for me to believe the OP because he never shows where in the melee tier list it would be affected. Therefore, "The use of such statistics could more accurately estimate the so called “tier” status of character qualities" is the only thing he mentions about the use of these extra methods. OP gives no example of when or even WHY these formulas give better results, merely stating that it would.

plus, OP here's the tl;dr for your paragraphs:
1- I have formulas that can help, im not a disbeliever in tiers.
2- wins and losses aren't everything
3- I have wikipedia links to some methods i found that could help

OP, you need to work on making your stuff readable. As my english professor and my history professor have told me many times: It takes skill to write intelligently, it takes more skill to write in a method that is both intelligent and easily understood.

tl;dr

10char

i'm glad someone's finally trying to set the record straight on this issue

very well worded and logical argument, i know it will be hard for this community to take

I've taken a number of classes on this subject myself, and his statistical points are valid, even if you don't understand them. The tier list will probably still look mostly the same where it counts; Snake and MK at the top. However, there's nothing wrong with computing the numbers in a more scientifically proven manner, is there? Accuracy is ALWAYS a good thing, especially for those who like to win so much that they only play top-tiers.

Enough of the "power trip" comments. He saw a flaw in the way these things are computed, and he's pointing it out. Instead of taking offense that your tier lists are statistically innacurate and attacking the guy, use his advice to improve your list and, through that, your own competitive game.

We know this already. It might be better if, you know, you showed an example.

i like the way you talk .

but sometimes you broke off the fun with a not-cool part here or there. for instance, "in which tournaments are held and data is collected based" was just awkwardly phrased. it could have been like "of tournaments, comparing and analyzing data pertaining to winners and competitors of each match and in total"..... anyway, it really ruins the feeling of "hey! i'm having fun while reading this!"

also, i would not make a big serious mini-essay and then end it with "here are some wiki links". in essay world, wiki is taboo. we all know that it IS very useful, so you probably for the validity of your mini-essay said something like "...wiki links, which -though wiki articles- i find quite useful in this matter".



this whole idea has already been widely regarded as true, particularly by players of lower-tiered characters (eg sheik and jigglypuff) or unpopular characters (eg ice climbers)... so i decided to rate your mini-essay. i like it . i just love phrases like "data collection" XD.

The current tier list doesn't take into account the fact that some characters are played by more people. The more folks play a given character, the more likelihood of that character being played by a SKILLED person. The more skilled people play a char, the more likely he or she or it is to win a tourney. A divisive factor (i.e. the number of current mainers for that character, or perhaps the percentage of smashers maining that character) would bring the results back to something realistic.

This might also put a bit of a damper on character over-play; with the current system, the more people play a character, the more people are GOING to play that character.

Example: Meta knight is intrinsically popular by the release of Brawl. MK attracts many good players in addition to the rest. MK wins more tourneys due to having more skilled mainers. MK tops the tier list. More people play MK so they, too, can win. More good players switch to MK along with all the others. MK wins MORE tournaments. MK STILL tops the tier list. Et cetera, et cetera, et cetera.

Cloud Cleaver said:

The current tier list doesn't take into account the fact that some characters are played by more people. The more folks play a given character, the more likelihood of that character being played by a SKILLED person. The more skilled people play a char, the more likely he or she or it is to win a tourney. A divisive factor (i.e. the number of current mainers for that character, or perhaps the percentage of smashers maining that character) would bring the results back to something realistic.

This might also put a bit of a damper on character over-play; with the current system, the more people play a character, the more people are GOING to play that character.

Example: Meta knight is intrinsically popular by the release of Brawl. MK attracts many good players in addition to the rest. MK wins more tourneys due to having more skilled mainers. MK tops the tier list. More people play MK so they, too, can win. More good players switch to MK along with all the others. MK wins MORE tournaments. MK STILL tops the tier list. Et cetera, et cetera, et cetera.

Click to expand...

other characters still have a chance. one extremely skilled player destroying a tourney or in some cases friendlies... or EVEN youtube videos... can impress middle-class players into getting deeply into that character, then after much practice attracting more players to that character by placing high in a tourney.

it's not a very good chance, but it's still a chance.

Cloud Cleaver said:

The current tier list doesn't take into account the fact that some characters are played by more people. The more folks play a given character, the more likelihood of that character being played by a SKILLED person. The more skilled people play a char, the more likely he or she or it is to win a tourney. A divisive factor (i.e. the number of current mainers for that character, or perhaps the percentage of smashers maining that character) would bring the results back to something realistic.

This might also put a bit of a damper on character over-play; with the current system, the more people play a character, the more people are GOING to play that character.

Example: Meta knight is intrinsically popular by the release of Brawl. MK attracts many good players in addition to the rest. MK wins more tourneys due to having more skilled mainers. MK tops the tier list. More people play MK so they, too, can win. More good players switch to MK along with all the others. MK wins MORE tournaments. MK STILL tops the tier list. Et cetera, et cetera, et cetera.

Click to expand...

i stopped reading at the red, please give me a reason to read on because i skimmed the rest of the post and saw meta knight as the 2-3rd words of the last paragraph.

we do the best we can. no1 has the time to unbiasedly test absolutyely everything. just a game guys

What.... um.............

Melee > brawl? lol jk

25%Cotton said:

i like the way you talk .




""and what kind of smart person links them talking about math stuff to WIKIPEDIA!!! send them to a math website...

did you just get out of a stat class or something and went on a power trip?""

""also, i would not make a big serious mini-essay and then end it with "here are some wiki links". in essay world, wiki is taboo. ""

Click to expand...

Despite your suggestion, that post was not intended to be an essay. I wrote it on the couch with my wife watching television. It was a post on a video game message board, thus the use of citing wikipedia is far more appropriate than citing peer reviewed articles. Furthermore the use of wiki-vandalism is generally removed from the wiki hard sciences and there is a large body of contributors that moderate and evaluate this information (http://en.wikipedia.org/wiki/Category:Wikipedians_with_PhD_degrees) . As for my writing style and my fluidity criticisms.. I’ll go with Mark Twain .. “I didn't have time to write a short letter, so I wrote a long one instead.””. @betterthanbonds, the reason you don’t understand the message is because you listen to your history and your English professor, as opposed to your statistics professor . There is no need to fuss over this. It was a suggestion for the community to embrace more valid approaches to tiering characters. It would be fairly simple as well— if tournaments would post their entries and results online uniformly somewhere. As for the powertrip, I am a graduate student in applied research methods, forgive me.

Well put, but I don't agree with some things.

A tier list is a general belief, not fact. But is a belief backed by data accurate enough to make it worth believing that MetaKnight is better than Captain Falcon.

betterthanbonds9 said:

i stopped reading at the red, please give me a reason to read on because i skimmed the rest of the post and saw meta knight as the 2-3rd words of the last paragraph.

Spoiler

SPOILER: there is no brawl tier list yet

Click to expand...

No there isn't, but there's only been like a billion tentative ones posted on this forum.
OP makes a good point; while it is obvious from playing them that some characters are indeed worse than others, taking data off of tournaments without considering all the factors will lead to skewed results.

vyse said:

Well put, but I don't agree with some things.

A tier list is a general belief, not fact. But is a belief backed by data accurate enough to make it worth believing that MetaKnight is better than Captain Falcon.

Click to expand...

I disagree that the data is accurate enough to make an appropriate estimation on the inherit advantages or disadvantages of a character. Tiers, as they are currently devised, measure more popularity and skilled player character selection, as opposed to inherent character atributes. Despite the value of thoughtful observations and “educated opinions”, the use of appropriate statistical techniques and sound methodological technology will more closely estimate the inherent advantages or disadvantages of a character. Note; I am not suggesting that random error accounts for the misinformed tier lists— I am suggesting that systematic error accounts for this. It is not a huge undertaking to account for some of these variables, as I have noted before. This could be done using open source software such as R (http://www.r-project.org/), and the data collection would be relatively easy if the community would commit to it. Once again, information originated from methodologically sound data analysis is not for everyone; it was simply a suggestion. For those that dislike this idea there is always “intuitive based” or “general opinions” to guide their information.

An example was requested: In the 1960’s our intelligence tests and “general opinion” as well as our data showed that African Americans are stupider, worse behaved, and more likely to be incarcerated than whites and latinos. So the “general opinion” was that African Americans are worse than white people. However if you take a population estimated sample, particularly in the east coast, and add socioeconomic status as a predictor variable the significance of these other predictors (bad behavior, incarceration, and low intelligence) becomes nill. So in essence it was the SES that predicted a large portion of these behaviors, not the race. This is a very basic example but hopefully some will benefit from it.

interesting that u should use that example...making Brawl tier lists almost seems like stereotyping

i think the result of this should be that ppl need to stop obssessing over tier lists or make more scientifically accurate ones

TL;DR version:

Tier lists are more of a popularity contest than an anctual ranking of character statistics.

RoboticTuna said:

No there isn't, but there's only been like a billion tentative ones posted on this forum.
OP makes a good point; while it is obvious from playing them that some characters are indeed worse than others, taking data off of tournaments without considering all the factors will lead to skewed results.

Click to expand...

The backroom makes the tier list...all the tier lists posted here aren't legit.

Here's the thing that everyone is forgetting here. The tier list isn't exactly a reflection of tournaments, it is a list of who can be played the best at the highest level humanly possible (this is inhumanly possible). It isn't meant to be absolute and is subject to many changes (melee had 7 [or was it 8?] tier lists).

Plus, is anybody in this thread in the BR? no

I so gotta back this guy up.

1. He talks funny because he just got out of freakin GRADUATE SCHOOL. Do you know what those scary places are like?
2. In ALL areas of life, statistics are proven to be more accurate than subjective opinions. Not spot on, but always more accurate. Watch TV. There's an election coming up. Do you see the people on TV saying: "here's the guys from the CNN Back Room. They really really know there stuff and know who's going to win."
Very rarely something like that happens, but 9 times as often, they tell us what the most recent poll says. Polls polls polls. Because polls are statistics and statistics are more accurate.
3. He's right, the hard science portions of wikipedia are spam free. Plus i bet he's looked over those pages himself and knows they are right. And wikipedia is free, i think statistics journals cost money.
4. The data collection would be easier than all the subjective argument that goes on in these forums. The data collection is the only hard part, the number crunching would just take someone's Saturday afternoon, at most.
5. We should do this guys.

popsofctown said:

In ALL areas of life, statistics are proven to be more accurate than subjective opinions. Not spot on, but always more accurate.

Click to expand...

That is, of course, assuming that your statistics are for something that can be proven statisically. For exampe, say I asked the world's population what the best color is. Since "best color" is a subjective topic, I couldn't use this poll to say that the color with the most votes is the statistically best color. I could say, however, that this poll shows the statistically most liked color.

You know, using good science to determine a tier list would actually take a LOT less time. No more of this "move so-and-so up" or "man, he's too high" BS.

And anyone too lazy to read a thorough counterargument has no reason to be debating anything in the first place.

Cloud Cleaver said:

You know, using good science to determine a tier list would actually take a LOT less time. No more of this "move so-and-so up" or "man, he's too high" BS.

And anyone too lazy to read a thorough counterargument has no reason to be debating anything in the first place.

Click to expand...

--you sure seem to merely blank post people's arguments arguing that we either didn't 1) read his argument, i read the wiki articles too btw. 2) are dumb (without giving a reason, merely stating it) or 3) arguing that there is a current tier list for brawl

DOES ANYBODY IN THIS GOD FORSAKEN TOPIC KNOW WHAT THE SBR IS? I mean really, you have not seen what they do, you have not read their stuff, you do not play this game at their level.

I love this quote from OP:

Pilgrim said:

I disagree that the data is accurate enough to make an appropriate estimation on the inherit advantages or disadvantages of a character. Tiers, as they are currently devised, measure more popularity and skilled player character selection, as opposed to inherent character atributes. Despite the value of thoughtful observations and “educated opinions”

Click to expand...

Here's a hypothetical situation and i have a feeling that you're into these, because I am too. You have just finished writing a paper on Euler's method (I use him as the example because i love his name) and i turn it in. Now, who would i have had revise my paper? Would i choose a. a random hobo b. my classmate or c. a mathematician? I would choose him because he is an expert in this matter (math). For this very reason the SBR works to revise each other statements much as my mathematician friend would because they are EXPERTS. Now, do you REALLY believe that players dont choose their mains based on "inherent character atributes" and "thoughtful observation?" I know that i tried wolf, but i found him too heavy for me. I also know that i observed videos of Darkmusician's Zelda, who is now currently my secondary. These skilled players that you bash for choosing "popular" characters forget the main logic of your first post: tier lists cause people to choose good characters. Meanwhile you forget entirely that as it is now there is no tier list but you see popular characters. This is because people enjoy using good characters because people enjoy winning. Remember this, skilled players choose good characters and win tournies, these good characters become popular because people like to win, tier lists come out and reflect that the character can be used to a high potential. Fun Fact, Shiek used to be top tier.

ps. I wish somebody here would tell me HOW this would make a difference (would pichu be the worst or would M2 remain the lowest?). Or even a manner of showing this stuff in action with the current stats. I could care less if i have to run all these characters by a formula if I get the SAME fricken answer.

tl;dr: show me when this would matter and come back to me

EDIT: you do know that there are too many variables to account for in brawl to ever come up with a true formula, dont you? I mean, just for SEM you'd need to have a matrice of each characters moves (idk how you could POSSIBLY rank them) then you'd have to combine that with usage (which would be a rough estimate) in order to see who can be used as a counter. THEN you'd have to calculate the user of the character which requires more specific analysis on the effectivenes of each move and so on and so on. In the end all you'd do is find a tier list for an individual player or at best a tier list of the players and their best character, which nobody cares about because we can generally tell who is better than who just by looking at them than creating the hypothesis that "Ken is the best at melee" and then confirm it using pages and pages of notes. Honestly, by the time you proved that Ken was the best, Brawl would've been out and everyone would already agree with you, but on the bright side, it could make for an interesting analysis of Ken's game and maybe a good read if you could find a publisher.

Same pretty much goes for multilevel modelling, they use set data to compare the effectiveness (the article mentions education stats, which are AT BEST shallow compared to the necesary stuff you wish to prove in brawl). Using multilevel modelling would create different tier lists for different levels of skill. So maybe newbs in melee were best with Roy, casuals were best with shiek, and pros were best with fox. What would this prove, we only care about the highest level of game so we only require one level. Nobody cares if your little brother has a mean M2, we care more about whether ken would **** more with fox or falco. And yes that is what you would test because it would be comparing different classes (ie skill levels).

logistic regression requires a logistic function and you should know that you need to have a set formula that would in lamemens terms have an input (x) and come out as f(x). For the sake of tier lists this would require an absurd statement like metaknight>snake. Most would argue that snake has the advantage in this matchup but in order to mathematically prove it without a doubt would be no less than impossible due to the fact that MKs have beaten Snakes and it goes to show that you have to account for a skilled player's skill. In other words there are far too many variables for one convienient formula to get and input (matchup) and come out with a 100% accurate output (winner)

Don't get the idea that i hate proven fact, but do understand that facts that are impossible to prove are not in my liking.

speaking of which, when are tier lists usually released?

Pilgrim said:

Despite your suggestion, that post was not intended to be an essay. I wrote it on the couch with my wife watching television. It was a post on a video game message board, thus the use of citing wikipedia is far more appropriate than citing peer reviewed articles. Furthermore the use of wiki-vandalism is generally removed from the wiki hard sciences and there is a large body of contributors that moderate and evaluate this information (http://en.wikipedia.org/wiki/Category:Wikipedians_with_PhD_degrees) . As for my writing style and my fluidity criticisms.. I’ll go with Mark Twain .. “I didn't have time to write a short letter, so I wrote a long one instead.””. @betterthanbonds, the reason you don’t understand the message is because you listen to your history and your English professor, as opposed to your statistics professor . There is no need to fuss over this. It was a suggestion for the community to embrace more valid approaches to tiering characters. It would be fairly simple as well— if tournaments would post their entries and results online uniformly somewhere. As for the powertrip, I am a graduate student in applied research methods, forgive me.

Click to expand...

uh... you DO realize that i wasn't the one with the "power trip" quote... and i also didn't have any trouble understanding your post >_>....

i understood your post and liked it while doing so.

betterthanbonds9 said:

"""you do know that there are too many variables to account for in brawl to ever come up with a true formula, dont you? I mean, just for SEM you'd need to have a matrice of each characters moves (idk how you could POSSIBLY rank them) then you'd have to combine that with usage (which would be a rough estimate) in order to see who can be used as a counter. """"

Click to expand...

You can use logistic regression in SEM where your outcome variable is a dichotomy (one of two outcomes [i.e. win or lose], trichotomy etc. etc.). A model that specifies move sets would be absurd and data collection would too rigorous. However using logistic regression in a multilevel model could provide a much more accurate estimate of wins and losses, particularly for character and skill level match ups (as determined by tournament outcomes). What you are referring to would be labeled as an advance survival analysis which would require millions of observations to have an adequately powered and unbiased estimate. You would never run something like that on video game statistics. You are not trying to come up with a “True Formula”— which is more like science fiction rather than science— Just a feasible, more accurately estimated, character advantage estimate.

However, sports psychologist use the statistics described at the onset of this post very often. Most professional athletes (e.g. Angelo Taylor along with many other Olympic competitors), use data based decision making for techniques, gear and outcome prediction— as opposed “intuitive knowledge” or “backroom lore”.

Other games such as popular competitive MMO’s already use this technology (I know because I have used it before). This is a very feasible (i.e. easy) alternative to the current methods of tier selection which are laden with threats to accuracy and validity.


EDIT:
@bonds -- I applaud your attempt at taking a wiki crash course on applied statistics. You either 1) have confirmed your assumption that wikipedia is horrible at explaining things or misinformed, or 2) read a few snipits of information and tried to pretend like you understood what you were talking about. Logistic functions can be used in both SEM and MLM, and the outcome variables can be categorical as opposed to continuous.

Pilgrim said:

"""you do know that there are too many variables to account for in brawl to ever come up with a true formula, dont you? I mean, just for SEM you'd need to have a matrice of each characters moves (idk how you could POSSIBLY rank them) then you'd have to combine that with usage (which would be a rough estimate) in order to see who can be used as a counter. """"

Click to expand...

dude, use the handy [ quote ] [/ quote] stuff

Pilgrim said:

You can use logistic regression in SEM where your outcome variable is a dichotomy (one of two outcomes [i.e. win or lose], trichotomy etc. etc.). A model that specifies move sets would be absurd and data collection would too rigorous. However using logistic regression in a multilevel model could provide a much more accurate estimate of wins and losses, particularly for character and skill level match ups (as determined by tournament outcomes). What you are referring to would be labeled as an advance survival analysis which would require millions of observations to have an adequately powered and unbiased estimate. You would never run something like that on video game statistics. You are not trying to come up with a “True Formula”— which is more like science fiction rather than science— Just a feasible, more accurately estimated, character advantage estimate.

Click to expand...

CAN YOU PLEASE SHOW ME HOW THIS AWESOMENESS WOULD CHANGE THE MELEE TIERS? Or at least show an example of this used in melee (just use critical hit 3 as the subject and use whoever won [i think it was azen] as the hypothesis: [winner] will win. I like accuracy, but if the answer is the same and more work is needed...there's no point.

And correct me if I'm wrong about this argument but arguing that there is no true formula just means that there is no true impact of using your methods versus the SBRs. Claim: science helps warrent: [methods] impact: still inaccurate and arguably the same because you have yet to show where it makes a DIFFERENCE.

Pilgrim said:

However, sports psychologist use the statistics described at the onset of this post very often. Most professional athletes (e.g. Angelo Taylor along with many other Olympic competitors), use data based decision making for techniques, gear and outcome prediction— as opposed “intuitive knowledge” or “backroom lore”.

Click to expand...

Facepalm, i know michael phelps and all the other swimmers have that technology to measure their speed/effectiveness. But for god's sakes, how the crap are you going to measure such movements on the characters at set times?

Pilgrim said:

Other games such as popular competitive MMO’s already use this technology (I know because I have used it before). This is a very feasible (i.e. easy) alternative to the current methods of tier selection which are laden with threats to accuracy and validity.

Click to expand...

Which MMO? That way when i edit this post i will know for sure to make an argument versus runescape/ WoW.

cjrocker said:

speaking of which, when are tier lists usually released?

Click to expand...

first melee tier list came out about 10 months after melee was released in NA. So we could have a nice christmas present if we're lucky

betterthanbonds9 said:

first melee tier list came out about 10 months after melee was released in NA. So we could have a nice christmas present if we're lucky

Click to expand...

And it will be methodological invalid unless more appropriate methods of tier selection are used

Still lol'ing at bonds attempt to pretend like he had a valid counter to why we shouldn't use more appropriate and potentially easier method of estimating character win variance !

betterthanbonds9 said:

dude, use the handy [ quote ] stuff



CAN YOU PLEASE SHOW ME HOW THIS AWESOMENESS WOULD CHANGE THE MELEE TIERS?

Click to expand...

For one, the only extra work in this type of data collection would be recording the characters used in tournaments, their match ups, the outcome (win, loss etc.) and the specific tournament EDIT: as well as the player entrants outcomes (duh). Two: I am not saying I know the difference between the current melee tier and the actual-- as I don't have a data set with the information required. However, you will ALWAYS get a more valid estimate of a statistically derived outcome if the model is specified more appropriately. ALWAYS-- assuming your error is controlled by an adequately powered experiment (which the sample size would be adequately powered if tournament play continues).

So put simply: The utilization of more appropriately modeled tier selection will always provide a more accurate estimate of reality as opposed to a less or not appropriately derived tier selection.

I used SEM, logistic regression and survival analysis in DAOC.

betterthanbonds9 said:

And correct me if I'm wrong about this argument but arguing that there is no true formula just means that there is no true impact of using your methods versus the SBRs.

Click to expand...

Okay, You are wrong. The same logic would go as follows: We are back in 1960: Despite more appropriately modeled statistical analyses being available let us still assume that blacks are all around worse people than whites because of their skin— Which is what the backroom of a large body of educated circles thought. Veto brown vs. Board of Education! So data showed us: despite majority opinion blacks genetics do not determine their likelihood of incarceration, mental ***********, or maladaptive behavior. This was found by not trusting mean differences and using a more appropriately specified model.


All course jokes aside: The point is you are currently using an underspecified statistical model (or not specified at all ). If you use a better fit you will always have a more accurate prediction. Who knows, it could be a fluke that your tiers appropriately estimated reality, but not likely given the number of alternative tiers (you said 8 or 9) and the amount of dissension over these numbers.

lets not make this a race thing...

More of a history thing than a race thing, actually...

And yes, when I see someone write "tl;dr", I IMMEDIATELY assume that they didn't read the post in question, since the literal translation is "too long; DIDN'T READ."

...wow....

Alright. So some characters will be ranked lower because they aren't as "popular" and don't have good people maining them? Amazing. We knew that. Marth was lower on the tier list until Ken made him popular. It doesn't matter. Tiers are based off of what we think NOW. We have no way of knowing what techs and strategies will be developed in the future. If the tier list is accurate due to CURRENT standings it is for all intents and purposes correct.

Terios the Hedgehog said:

If the tier list is accurate due to CURRENT standings it is for all intents and purposes correct.

Click to expand...

If you are more interested in obtaining popularity data, character selection data, tournament characteristics, then yes. If you wanted this you could simply exam the post counts on each of the brawl character discussion threads and name who is the best character by the post count. If you are interested in data based decisions and suggestions regarding match ups and tiers, then no. Anyways, it was a suggestion-- it sounds like some are interested and some are ignorant or not interested. My 2 cents are here peace, 0.o.

Pilgrim said:

If you wanted this you could simply exam the post counts on each of the brawl character discussion threads and name who is the best character by the post count.

Click to expand...

By these results, Sonic would be king, and Jiggly would be dead last. That's kind of interesting.

Gory snake said:

Since when did super computers frequent Smashboards?

Click to expand...

I lol'd

10lols