adumbrodeus
Smash Legend
Link to original post: [drupal=1609]Banning Metaknight: Pro-ban's "To-Do List" to Show that Metaknight warrants a Ban[/drupal]
1. Introduction
Once again, there's a should "Metaknight be Banned?" thread in Brawl tactical discussions, and once again we have the general festival of unreasoned arguments and half-truths from both sides of the debate.
Pro-ban is intent on proving that Metaknight is so far beyond the rest of the cast that's he needs a ban, or the fact that he has no counters warrants a ban, or whatever the argument of the moment is.
Anti-ban however, seems intent on proving that MK is actually quite weak, and why they acknowledge that MK is the best character in the game, they seem to be willing to put forth any character imaginable as an MK counter. Furthermore, it seems to be getting to the Boards, with most boards acknowledging better match-ups against MK then they probably deserve, and the fact that Marth (who is basically a weaker MK, with slightly more emphasis on defense) renders more characters nonviable according the the current match-up consensus seems to illustrate this.
So, rather then resorting to hyperbole, it's time to take a step back and look at how a character is proven to fit the ban criteria and apply it to Metaknight.
2. The Ban Standard
First, understand where this is coming from. The standard being used is overcentralization, and this standard is derived from Sirlin's standard of banning described in the Book, Playing to Win. David Sirlin is a former competitive Street Fighter player (actually, an amazing one), a current game designer, and probably the most respected competitive gaming theorist on the planet.
His Ban criteria is simple, and described here (part 1) and here (part 2). It requires that bans are Enforceable (in other words, judges are able to practically distinguish between what is and is not banned), Discrete (in other words, the tactic cannot "sort of be used", a lesser form of the technique cannot be exchanged that becomes the new best move), and Warranted (in other words, the whatever actually deserves to be banned). Since it's obvious that a character ban easily fits into both discrete and enforceable, this blog will concentrate on "Warranted".
Now, what exactly makes something warranted? Sirlin's explanation is actually pretty explicit on this point. 50%, straight-out with a mature community (in other words, metagame development is over because knowledge of the characters is as advanced as possible). It has to render that number of characters nonviable.
3. What makes a Character Nonviable?
In order to make a character non-viable, a character has to do this to another character:
1. A hard counter to that character at least (80-20 min would probably be more proper).
2. Has to be either:
a. Omni-present enough in the metagame that not being eliminated by this character is near impossible given the hard counter status (even if you're a superior player).
OR
b. the match-up is easy enough that an opponent can virtually pick up the character to beat you.
3. No other character fits these attributes.
4. This pattern has to be backed up by tournament data.
The reasoning should be pretty obvious, the counter has to be good enough to beat the opponent even with a significant skill gap, and has to be common enough that getting through with the character made nonviable is nearly impossible or you can use your counterpick to destroy the character (King DDD vs. DK is an example of the latter). At the same time, if another character fits those attributes, what good would it be to ban the first character as far as the nonviable character is concerned, they'd still be nonviable.
Tournament data is required as always, as a self-check to make sure an error was not made in the match-ups.
4. Why Use Sirlin's Standards as a Basis?
Next another common point should be addressed, why Sirlin?
Ultimately, this standard works because of practicality, the fundamental idea is not clogging up the metagame with unnecessary rules which makes it hard to play. At the same time, it recognizes that there are times when bans have to occur.
Firstly, one should not ban based on low standards because in order to be consistent if you ban something, you need to ban something else which reaches that threshold. For example in MvC2, 4 characters basically dominated the metagame, however if they were banned, the next few would dominate the metagame, and so on, eventually resulting in you needing to ban everyone except the bottom 3, resulting in the same amount of tournament diversity.
And what about arbitrary bans? Ultimately, they prevent advancements in the metagame, because people never know if their tactic will be the next to be banned by the community. This prevents continual improvement in the metagame and ultimately stagnates it.
And what about not banning at all? In some cases a single tactic is overpowering enough that it will totally dominate the metagame. This was seen in the case of all infinite stalls which are universally banned. If for whatever reason, you can permanently lock up the opponent, because of the timer rule, you've won. Super Turbo's Akuma (and almost certainly HD remix's Akuma) are perfect examples of characters that do the same.
Therefore, we need a harsh standard for banning, but one that does not ban what is needed, and this standard serves on both counts.
5. Pro-Ban's To Do List
Based on those assumptions, I came up with this list of things that need to occur in order for Pro-ban to prove it's point that MK is bannable. This entire thing would be useful for the community, but proving that points 2 and 5 do not match up with pro-bans assertions would also prove positively that MK does not warrant a ban. Furthermore, various aspects of this would be useful information for the community as a whole.
1. Make a better theoretical model for match-ups. The current one really doesn't suffice because it neglects to take into account human error as far as I've observed, possibly a great deal more. I've attempted to help with my thread on "Mindgames Potential", though we still need a concrete understanding of how often a player should fall victim to mindgames, and to what degree before it can be finished.
2. Using the new theoretical model, make accurate match-ups.
3. Illustrate based on the match-ups, that MK makes 50% of the cast +1 nonviable, independent of any other characters.
4. Figure out a reasonable margin of error for tournament results to vary from the theoretical match-ups.
5. Illustrate that empirical results are the same as your match-ups would suggest or within the pre-determined margin of error.
6. Explanations
1. The current theoretical model we have is broken, I have noticed the lack of attention to human error as one example, but others have expressed many others. What is needed is a community project to ascertain what the correct theoretical model for match-ups actually is.
2. Once the model is established, match-ups must be discussed, this is required to determine whether or not MK makes enough characters nonviable to be bannable.
3. Taken directly from Sirlin's standards, except one point. Since 50% is "barely" and our community is relatively young, we'll need it firmly within the realm of bannable, which means a consensus should be used, hence 50%+1.
4. Tournament data is a good self-check, but unfortunately concentrations of good players in certain characters often throw off tournament results relative to actual match-ups. For that reason, tournament results MUST be interpreted in light of that fact, therefore we need to establish a margin of error. Remember, Pit is almost top tier in Japan because of a few amazing players.
5. Just a presentation of data in light of point 4.
7. Conclusion
Hopefully this should provide some insight into why I maintain an anti-ban position, even though I personally dislike MK's influence on the metagame because he negates the metagame influence of my favorite character, Marth.
I do not claim to speak for all of anti-ban, however, there does seem to be a reasonable number of people who agree with me, so this idea is not without influence. In order to positively prove MK is banworthy, you need some reference to an outside standard instead of creating your own, and hopefully this standard will be universally accepted to allow for more organization and (god forbid) actual good faith attempts to truly figure out whether or not MK is bannable.
1. Introduction
Once again, there's a should "Metaknight be Banned?" thread in Brawl tactical discussions, and once again we have the general festival of unreasoned arguments and half-truths from both sides of the debate.
Pro-ban is intent on proving that Metaknight is so far beyond the rest of the cast that's he needs a ban, or the fact that he has no counters warrants a ban, or whatever the argument of the moment is.
Anti-ban however, seems intent on proving that MK is actually quite weak, and why they acknowledge that MK is the best character in the game, they seem to be willing to put forth any character imaginable as an MK counter. Furthermore, it seems to be getting to the Boards, with most boards acknowledging better match-ups against MK then they probably deserve, and the fact that Marth (who is basically a weaker MK, with slightly more emphasis on defense) renders more characters nonviable according the the current match-up consensus seems to illustrate this.
So, rather then resorting to hyperbole, it's time to take a step back and look at how a character is proven to fit the ban criteria and apply it to Metaknight.
2. The Ban Standard
First, understand where this is coming from. The standard being used is overcentralization, and this standard is derived from Sirlin's standard of banning described in the Book, Playing to Win. David Sirlin is a former competitive Street Fighter player (actually, an amazing one), a current game designer, and probably the most respected competitive gaming theorist on the planet.
His Ban criteria is simple, and described here (part 1) and here (part 2). It requires that bans are Enforceable (in other words, judges are able to practically distinguish between what is and is not banned), Discrete (in other words, the tactic cannot "sort of be used", a lesser form of the technique cannot be exchanged that becomes the new best move), and Warranted (in other words, the whatever actually deserves to be banned). Since it's obvious that a character ban easily fits into both discrete and enforceable, this blog will concentrate on "Warranted".
Now, what exactly makes something warranted? Sirlin's explanation is actually pretty explicit on this point. 50%, straight-out with a mature community (in other words, metagame development is over because knowledge of the characters is as advanced as possible). It has to render that number of characters nonviable.
3. What makes a Character Nonviable?
In order to make a character non-viable, a character has to do this to another character:
1. A hard counter to that character at least (80-20 min would probably be more proper).
2. Has to be either:
a. Omni-present enough in the metagame that not being eliminated by this character is near impossible given the hard counter status (even if you're a superior player).
OR
b. the match-up is easy enough that an opponent can virtually pick up the character to beat you.
3. No other character fits these attributes.
4. This pattern has to be backed up by tournament data.
The reasoning should be pretty obvious, the counter has to be good enough to beat the opponent even with a significant skill gap, and has to be common enough that getting through with the character made nonviable is nearly impossible or you can use your counterpick to destroy the character (King DDD vs. DK is an example of the latter). At the same time, if another character fits those attributes, what good would it be to ban the first character as far as the nonviable character is concerned, they'd still be nonviable.
Tournament data is required as always, as a self-check to make sure an error was not made in the match-ups.
4. Why Use Sirlin's Standards as a Basis?
Next another common point should be addressed, why Sirlin?
Ultimately, this standard works because of practicality, the fundamental idea is not clogging up the metagame with unnecessary rules which makes it hard to play. At the same time, it recognizes that there are times when bans have to occur.
Firstly, one should not ban based on low standards because in order to be consistent if you ban something, you need to ban something else which reaches that threshold. For example in MvC2, 4 characters basically dominated the metagame, however if they were banned, the next few would dominate the metagame, and so on, eventually resulting in you needing to ban everyone except the bottom 3, resulting in the same amount of tournament diversity.
And what about arbitrary bans? Ultimately, they prevent advancements in the metagame, because people never know if their tactic will be the next to be banned by the community. This prevents continual improvement in the metagame and ultimately stagnates it.
And what about not banning at all? In some cases a single tactic is overpowering enough that it will totally dominate the metagame. This was seen in the case of all infinite stalls which are universally banned. If for whatever reason, you can permanently lock up the opponent, because of the timer rule, you've won. Super Turbo's Akuma (and almost certainly HD remix's Akuma) are perfect examples of characters that do the same.
Therefore, we need a harsh standard for banning, but one that does not ban what is needed, and this standard serves on both counts.
5. Pro-Ban's To Do List
Based on those assumptions, I came up with this list of things that need to occur in order for Pro-ban to prove it's point that MK is bannable. This entire thing would be useful for the community, but proving that points 2 and 5 do not match up with pro-bans assertions would also prove positively that MK does not warrant a ban. Furthermore, various aspects of this would be useful information for the community as a whole.
1. Make a better theoretical model for match-ups. The current one really doesn't suffice because it neglects to take into account human error as far as I've observed, possibly a great deal more. I've attempted to help with my thread on "Mindgames Potential", though we still need a concrete understanding of how often a player should fall victim to mindgames, and to what degree before it can be finished.
2. Using the new theoretical model, make accurate match-ups.
3. Illustrate based on the match-ups, that MK makes 50% of the cast +1 nonviable, independent of any other characters.
4. Figure out a reasonable margin of error for tournament results to vary from the theoretical match-ups.
5. Illustrate that empirical results are the same as your match-ups would suggest or within the pre-determined margin of error.
6. Explanations
1. The current theoretical model we have is broken, I have noticed the lack of attention to human error as one example, but others have expressed many others. What is needed is a community project to ascertain what the correct theoretical model for match-ups actually is.
2. Once the model is established, match-ups must be discussed, this is required to determine whether or not MK makes enough characters nonviable to be bannable.
3. Taken directly from Sirlin's standards, except one point. Since 50% is "barely" and our community is relatively young, we'll need it firmly within the realm of bannable, which means a consensus should be used, hence 50%+1.
4. Tournament data is a good self-check, but unfortunately concentrations of good players in certain characters often throw off tournament results relative to actual match-ups. For that reason, tournament results MUST be interpreted in light of that fact, therefore we need to establish a margin of error. Remember, Pit is almost top tier in Japan because of a few amazing players.
5. Just a presentation of data in light of point 4.
7. Conclusion
Hopefully this should provide some insight into why I maintain an anti-ban position, even though I personally dislike MK's influence on the metagame because he negates the metagame influence of my favorite character, Marth.
I do not claim to speak for all of anti-ban, however, there does seem to be a reasonable number of people who agree with me, so this idea is not without influence. In order to positively prove MK is banworthy, you need some reference to an outside standard instead of creating your own, and hopefully this standard will be universally accepted to allow for more organization and (god forbid) actual good faith attempts to truly figure out whether or not MK is bannable.
