• This forum is strictly intended to be used by members of the VS Battles wiki. Please only register if you have an autoconfirmed account there, as otherwise your registration will be rejected. If you have already registered once, do not do so again, and contact Antvasima if you encounter any problems.

    For instructions regarding the exact procedure to sign up to this forum, please click here.
  • We need Patreon donations for this forum to have all of its running costs financially secured.

    Community members who help us out will receive badges that give them several different benefits, including the removal of all advertisements in this forum, but donations from non-members are also extremely appreciated.

    Please click here for further information, or here to directly visit our Patreon donations page.
  • Please click here for information about a large petition to help children in need.

Outlier Revision

Status
Not open for further replies.
Kulf_Boba

Kulf_Boba

866
564
Currently the wiki has standards and rules for all kinds of feats and declarations, but something as important as the Outlier does not have a guide to help identify them, so I propose the methods to identify them mentioned in this blog.

They consist of:

INTERQUARTILE RANGE (OPTIONAL):​

One of the most commonly used methods in statistics to detect outliers is the elaboration of dispersion measures following the criteria of Tukey's test. The basic idea is to elaborate a box plot using the median as the central axis where the difference between the third and the first quartile (i.e. the difference between 25% and 75%) forms an interquartile range, of which the values that exceed or fall below this range would be considered outliers. Putting it graphically:

+−−−−−+−+
* |−−−−−−−−−−−| | |−−−−−−−−−−−|
+−−−−−+−+

+−−−+−−−+−−−+−−−+−−−+−−−+−−−+−−−+−−−+−−−+−−−+−−−+ feat level
0 1 2 3 4 5 6 7 8 9 10 11 12

Note that only values falling below 1.5 of the interquartile range of the first quartile or exceeding 1.5 of the interquartile range of the third quartile could be considered outliers (in this case the asterisk). Ideally, we will calculate the median with the most outstanding feats of a character, but if he does not have many, we can use all of them. For greater consistency of results, I recommend excluding the feats performed by the character in a very casual way, except those that there is a suspicion that may exceed the third quartile.

If you have doubts about how to calculate an interquartile range, this guide can help you.
Click to expand...
Note: In this blog you will find an example of how to use the interquantile range to identify an outlier in fiction.

And the second thing that should definitely be on the outliers page are the following guidelines for identifying one.
vsbattles.fandom.com

Antvasima/Sandbox

vsbattles.fandom.com vsbattles.fandom.com

Guidelines​

Here are some guidelines for how to recognise outliers. Fulfilling all 5 of the requirements at once should not be necessary in order to qualify, and our members should try to use their common sense regarding what seems reasonable in each specific case, depending on to how extreme degree the respective categories have been fulfilled in combination.

1) Is it a big jump or drop in power? If a character with several planetary feats shows themselves capable of destroying a star, we cannot necessarily consider it as an outlier, for the reason that the jump between tiers is not extreme enough to be so, despite the jump between energy values we attribute to them. If the character has very few feats, we can likely ignore this point.

2) Is it a unique or exceptional incident? If incidents of a similar level are repeated consistently over time, they are unlikely to be outliers. Usually, from the third incident onwards. If the character has very few feats, we can also likely skip this point.

3) Is the event unexplained and unjustified? If an extreme incident is not accompanied by any kind of explanation that justifies it, it is probably an outlier. But if it can be explained by means of some power-up, vulnerability or limiter, it is likely not. If Superman is wounded by a bullet, it's probably an outlier. But if Superman was under the influence of red sunlight or was previously weakened with Kryptonite, it is not.

4) Does the event break the previously established power-scaling? Here we must take into account a number of factors, such as comparable characters possessing feats or statements of a similar level to the hypothetical outlier, the outlier not being supported by fights that might suggest a similar level, or subsequent events or statements that contradict it in some way.

5) Does the event break with the narrative of the work? Many times an outlier breaks with what has been previously established or shown in a work, creating inconsistencies that are difficult to resolve unless we invalidate one of the two events. If, for example, a character claims not to be able to dodge bullets, even though he was previously seen dodging them explicitly, we are faced with a contradiction where we will probably have to resort to using the more reliable evidence.
Click to expand...
 
Last edited:
Guidelines look good. Not sure about sliding Inter quartile range so I'm neutral towards that.
 
Planck69 said:
Guidelines look good. Not sure about sliding Inter quartile range so I'm neutral towards that.
Click to expand...
By the way, I don't think that the Interquartile Range should always be used as it is not necessary, but I consider that it should be used for feats that are in doubt if the jump between levels is high and for characters that have too many feats.
 
Ah one thing that should be changed is the example that the jump from 5-B to 4-C cannot be considered Outlier, since the difference between these levels if large 2,289,103,337 times to be exact.

And one very important thing, if a feat covers two stats for example speed and AP, but the AP in this case is inconsistent, then the AP would be Outlier and the speed would not.
 
Last edited:
I'm confused on the last comment; why can't "A jump from 5-B to 4-C" be considered outlier? And because it's that large? It's a very case by case thing; a jump from 9-B to Tier 2 from one feat can still be consistent if context is in its favor. Likewise, it's also possible from a jump from Low 7-B to 7-A to be an outlier despite more than one feats if context is against the favor or if it comes from feats that create too many loopholes.

Also, we have advised in the past that if a specific feat is an AP, speed, and range feat; that making something an outlier in terms of AP but accepting the range or speed as consistent is against the rules.
 
As I said before, I like the five guidelines being proposed, but I'm not in favor of the interquartile range stuff.
 
DarkDragonMedeus said:
I'm confused on the last comment; why can't "A jump from 5-B to 4-C" be considered outlier?
Click to expand...
I say that part should be corrected, as the jump is very large and if there is no justification it is obviously an outlier.

It's a very case by case thing; a jump from 9-B to Tier 2 from one feat can still be consistent if context is in its favor. Likewise, it's also possible from a jump from Low 7-B to 7-A to be an outlier despite more than one feats if context is against the favor or if it comes from feats that create too many loopholes.
Click to expand...
A slight jump between levels cannot be considered Outlier unless proven otherwise.

I think question 1 could be changed this way:

Is it a big jump or drop in power? If a character with several city destruction feats is shown to be able to destroy a mountain, we cannot necessarily consider it as an outlier, for the reason that the jump between tiers is not extreme enough to be so, despite the jump between energy values we attribute to them. If the character has very few feats, we can likely ignore this point.

Also, we have advised in the past that if a specific feat is an AP, speed, and range feat; that making something an outlier in terms of AP but accepting the range or speed as consistent is against the rules.
Click to expand...
What rules?
 
Last edited:
Damage3245 said:
As I said before, I like the five guidelines being proposed, but I'm not in favor of the interquartile range stuff.
Click to expand...
Do you agree to use it in cases like the ones I have mentioned?
but I consider that it should be used for feats that are in doubt if the jump between levels is high and for characters that have too many feats.
Click to expand...
If you disagree, could you explain why not?
 
I don't think the interquartile range thing works for our purposes. It works in statistics because one uses large datasets of data that is supposed to be taken under equal conditions. (Well, as far equal as one can control)
The same doesn't apply for us. We rarely have enough data points to justify any statistical analysis. Additionally, the data points aren't equal. They are done with various degrees of ease, with various techniques and at different points in time. 1 million wall level feats wouldn't make a planet level feat an outlier if the wall level feats were all irrelevant side damage from casual combat, that required no effort and caused no damage to the character.

The other 5 guidelines are ok, I guess. One guideline I'm missing is something like "is contradicted by later feats". Often one can explain an outlier by saying that the character simply got more powerful than in earlier feats, so having contradictory showings later is more noteworthy.
 
Along with everyone else, I agree with the Guidelines and disagree with the interquartile range stuff
 
i share the same opinion with everyone here
agree with the guidelines, disagree with the interquartile range stuff
 
Elizhaa said:
The guidelines look good. Interquartile ranges, I don't think would necessarily work for reasons such as those stated by DontTalkDT.
Click to expand...
i'll take this moment to plug and tell you to check my latest comment in the ygo thread
cristiano-ronaldo-drinking.gif
 
Kulf_Boba said:
I say that part should be corrected, as the jump is very large and if there is no justification it is obviously an outlier.
Click to expand...
I wouldn't say "Obvious Outlier". I also had plans to add protagonists archetypes that help determine outliers; I'd agree that a big jump coming from a Type I or Type II protagonist would be a likely outlier without further context, but not entirely if it's a Type III or Type IV. I'll define those protagonists archetypes later for each section in more elaborate details, but here's the more simplified descriptions
  1. Type I: The Everyman, also nicknamed the Underdog or Average Joe or Pawn of Duty. Protagonists that are typically within the same ballpark as their piers and typically demonstrate barebones superhuman feats if any. Most of these characters should be scaled from their own destruction feats and scaling from characters much stronger than they are or some notable lucky durability feats may be prone to outlier territory even if done frequently.
  2. Type II: The Fluctuative Hero, also nicknamed the Neutral Dynamical protagonist or Adaptive Protagonist. Basically protagonists that are hardest to determine due to how back and forth the character is portrayed and especially if they star in verses that tend to be all over the place such as Marvel/DC Comics. Some may qualify for variable tiers, but others simply cannot seem to decide if they're a Type 1 or Type 4, so most of them may be mid-balled if there's a lack of better options
  3. Type III: The Constructive Hero, also nicknamed the Evolutionary Protagonist. Characters whose primary trope is that they get stronger after every in verse chapter or fight they enter and survive. They basically grow from Type 1 to Type 4 over time. This is basically a given for various Shounen protagonists and nearly all RPG protagonists.
  4. Type IV: The Dominant Hero, also known as the Top Dog, Final or Ultimate protagonist. This is where the main protagonist is not only one of the strongest but often times the strongest character in their own series, and are often the case by far. Characters in question also often display little to know weaknesses and are commonly believed to be invincible within the showings of their own franchise. Prominent examples are Saitama from One Punch Man and Bobobo from the series of the same name.
Kulf_Boba said:
A slight jump between levels cannot be considered Outlier unless proven otherwise.

I think question 1 could be changed this way:

Is it a big jump or drop in power? If a character with several city destruction feats is shown to be able to destroy a mountain, we cannot necessarily consider it as an outlier, for the reason that the jump between tiers is not extreme enough to be so, despite the jump between energy values we attribute to them. If the character has very few feats, we can likely ignore this point.
Click to expand...
Actually, I can think of examples of small gaps still being outliers. For example, I have seen plenty of cases where a 9-B character survives 9-A explosions but the context was a "They were hospitalized but they survived". Or weaker characters basically chip damaging stronger characters isn't always enough evidence for scaling and especially if the context is using a sharp weapon to stab their eye.

Also, they aren't outliers anymore after long discussions. But it was agreed that Roshi's Moon busting feat that if we're going to accept Moon level, we also have to accept Planetary range and Relativistic speed. And not excepting the speed would likewise reject the AP and range feat.
 
Last edited:
Count me in for agreeing with the guidelines and disagreeing with the interquartile range stuff for DT's reasonings.
 
DontTalkDT said:
I don't think the interquartile range thing works for our purposes. It works in statistics because one uses large datasets of data that is supposed to be taken under equal conditions. (Well, as far equal as one can control)
The same doesn't apply for us. We rarely have enough data points to justify any statistical analysis. Additionally, the data points aren't equal. They are done with various degrees of ease, with various techniques and at different points in time. 1 million wall level feats wouldn't make a planet level feat an outlier if the wall level feats were all irrelevant side damage from casual combat, that required no effort and caused no damage to the character.
Click to expand...
The idea is to do an intelligent analysis of feats, preferably collecting feats that are not done too casually or where the character holds back too much, or else you will get even 5-B feats for Thor as Outliers.
DontTalkDT said:
One guideline I'm missing is something like "is contradicted by later feats". Often one can explain an outlier by saying that the character simply got more powerful than in earlier feats, so having contradictory showings later is more noteworthy.
Click to expand...
This would already be supported by question 1.
 
DarkDragonMedeus said:
I wouldn't say "Obvious Outlier". I also had plans to add protagonists archetypes that help determine outliers; I'd agree that a big jump coming from a Type I or Type II protagonist would be a likely outlier without further context, but not entirely if it's a Type III or Type IV. I'll define those protagonists archetypes later for each section in more elaborate details, but here's the more simplified descriptions
  1. Type I: The Everyman, also nicknamed the Underdog or Average Joe or Pawn of Duty. Protagonists that are typically within the same ballpark as their piers and typically demonstrate barebones superhuman feats if any. Most of these characters should be scaled from their own destruction feats and scaling from characters much stronger than they are or some notable lucky durability feats may be prone to outlier territory even if done frequently.
  2. Type II: The Fluctuative Hero, also nicknamed the Neutral Dynamical protagonist or Adaptive Protagonist. Basically protagonists that are hardest to determine due to how back and forth the character is portrayed and especially if they star in verses that tend to be all over the place such as Marvel/DC Comics. Some may qualify for variable tiers, but others simply cannot seem to decide if they're a Type 1 or Type 4, so most of them may be mid-balled if there's a lack of better options
  3. Type III: The Constructive Hero, also nicknamed the Evolutionary Protagonist. Characters whose primary trope is that they get stronger after every in verse chapter or fight they enter and survive. They basically grow from Type 1 to Type 4 over time. This is basically a given for various Shounen protagonists and nearly all RPG protagonists.
  4. Type IV: The Dominant Hero, also known as the Top Dog, Final or Ultimate protagonist. This is where the main protagonist is not only one of the strongest but often times the strongest character in their own series, and are often the case by far. Characters in question also often display little to know weaknesses and are commonly believed to be invincible within the showings of their own franchise. Prominent examples are Saitama from One Punch Man and Bobobo from the series of the same name.
Click to expand...
Type III and IV characters have a justification for the power jump.
DarkDragonMedeus said:
Actually, I can think of examples of small gaps still being outliers. For example, I have seen plenty of cases where a 9-B character survives 9-A explosions but the context was a "They were hospitalized but they survived". Or weaker characters basically chip damaging stronger characters isn't always enough evidence for scaling and especially if the context is using a sharp weapon to stab their eye.
Click to expand...
9-B characters can quietly survive a 9-A blast as they commonly don't get all the energy from it.

If a high-end 9-B character jumps to 9-A it cannot be considered Outlier unless there are contradictions, an outlier is something out of the ordinary.
DarkDragonMedeus said:
Also, they aren't outliers anymore after long discussions. But it was agreed that Roshi's Moon busting feat that if we're going to accept Moon level, we also have to accept Planetary range and Relativistic speed. And not excepting the speed would likewise reject the AP and range feat.
Click to expand...
And why not, if we are already discarding something that is in the same work and that for the author always happened, why can't we consider only one statistic as Outlier being that it really is since it is inconsistent while another one is not?
 
Last edited:
Well if the interquartile range is not accepted no problem let's go with the guidelines,@Antvasima could change question "1" in this way as it makes more sense:
1) Is it a big jump or drop in power? If a character with several city destruction feats is shown to be able to destroy a mountain, we cannot necessarily consider it as an outlier, for the reason that the jump between tiers is not extreme enough to be so, despite the jump between energy values we attribute to them. If the character has very few feats, we can likely ignore this point.
Click to expand...
 
Thank you for the reply.

I would prefer a few more confirmations before I go ahead though.
 
Okay. Singular seems mismatched with the naming conventions for our categories and other pages though.
 
Can you inform me of relevant examples please?
 
Thank you for the list.

What do the rest of you think? Should we keep those page titles in singular or change them to plural instead?
 
Status
Not open for further replies.

Similar threads

Vzearr
  • Locked
Using Guns
2
Replies
72
Views
3K
Planck69
Planck69
LordGriffin1000
  • Locked
Danny Phantom A Glitch in Time - Potential Immeasurable Speed Upgrade (Or just Dimensional Travel)
2
Replies
50
Views
2K
LordGriffin1000
LordGriffin1000
Fwyzzverse
Timeframe | Creation Feats (Revision)
Replies
33
Views
2K
Qawsedf234
Qawsedf234
Fezzih_007
Time travel range
Replies
0
Views
348
Fezzih_007
Fezzih_007
KingogKings777
Mom's achievements carrying the daughter
Replies
29
Views
967
KingogKings777
KingogKings777
Back
Top