1. New to the forums? Check out our Mentorship Program!
    Our mentors will answer your questions and help you become a part of the community!
  2. Welcome to Smogon Forums! Please take a minute to read the rules.

Measure of centralisation in a metagame

Discussion in 'Pokémetrics' started by X-Act, Dec 10, 2008.

Thread Status:
Not open for further replies.
  1. X-Act

    X-Act np: Biffy Clyro - Shock Shock
    is a Site Staff Alumnusis a Programmer Alumnusis a Smogon IRC SOp Alumnusis a Researcher Alumnusis a CAP Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis an Administrator Alumnus

    Joined:
    Feb 17, 2006
    Messages:
    4,675
    Here is a simple way to have a clue as to how centralised is a particular metagame.

    As you noticed from my recent stats, I wasn't just providing the list of Pokemon that were OU, but also a list of Pokemon that were 'frequently used'. Both of these are used to check for the centralisation of a metagame.

    To remind you, a Pokemon is OU if its usage is greater than the sum of all usages divided by 176.1407. A Pokemon is frequently used if its usage is greater than the sum of all usages divided by 37.7113.

    Count the number of Pokemon that are OU; this is equal to O. Then count the number of Pokemon that are frequently used; this is equal to F.

    The measure of centralisation is then this very simple formula:

    Code:
    Central = floor(500 / (O - F))
    Here is the centralisation of various metagames from September onwards:

    September Standard Ladder: 11
    October Standard Ladder: 11
    November Standard Ladder: 11

    September UU Ladder: 9
    October UU Ladder: 10
    November UU Ladder: 10

    September Uber Ladder: 26
    October Uber Ladder: 27
    November Uber Ladder: 35

    September Suspect Ladder: 12
    October Suspect Ladder: 22 (!)
    November Suspect Ladder: N/A

    It looks like the OU metagame and the UU metagame have low centralisation, with UU having slightly less centralisation than OU. Uber has a lot of centralisation. But this is something that you all know about. :)

    What is interesting is the Suspect ladder metagame centralisation. For September, it was only slightly more centralised than for OU, while for October it was much more centralised. The reasons for this are not entirely clear to me.

    I thought to share this piece of information with all of you. :)
  2. Matt_Lane

    Matt_Lane

    Joined:
    Sep 9, 2007
    Messages:
    556
    Thanks - yet again - X-Act!
  3. cim

    cim happiness is such hard work
    is a Smogon Media Contributor Alumnus

    Joined:
    Jun 3, 2007
    Messages:
    5,412
    I think that, although useful, we don't know how to apply the word "centralized" to an Uber argument. Most people use "how much impact does it have on my team" as "centralization", and I don't know if this numerical definition or the subjective one should have more weight.

    This is part of why I wanted Rayquaza test, so we can see these numbers in an obviously "one Uber" metagame.
  4. Ancien Régime

    Ancien Régime capitalism delenda est
    is a Team Rater Alumnusis a Battle Server Moderator Alumnus

    Joined:
    Jul 21, 2007
    Messages:
    1,453
    Ubers isn't supposed to be balanced so "Centralization" is wholly irrelevant.
  5. Pombo

    Pombo

    Joined:
    Feb 29, 2008
    Messages:
    701
    Thanks again, but... tell me something: whats the university you made? You got some talent/good teachers/both.
  6. X-Act

    X-Act np: Biffy Clyro - Shock Shock
    is a Site Staff Alumnusis a Programmer Alumnusis a Smogon IRC SOp Alumnusis a Researcher Alumnusis a CAP Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis an Administrator Alumnus

    Joined:
    Feb 17, 2006
    Messages:
    4,675
    Well, if a metagame containing "OU + Rayquaza" would have more centralisation than "OU only", then this formula should show that. Whether or not a centralising Pokemon condemns it to uber or not is a matter of debate, though.

    Of course, but knowing that the uber metagame is very centralised is a reason why it's a good test case to show that the formula works.
  7. Caelum

    Caelum qibz official stalker
    is a Site Staff Alumnusis a Smogon IRC AOp Alumnusis a Forum Moderator Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis a Battle Server Moderator Alumnus

    Joined:
    Apr 5, 2008
    Messages:
    1,656
    Where did the 500 come from?
  8. cim

    cim happiness is such hard work
    is a Smogon Media Contributor Alumnus

    Joined:
    Jun 3, 2007
    Messages:
    5,412
    I'm talking about Suspects, not the Uber tier.
  9. AeroGP

    AeroGP

    Joined:
    Aug 4, 2006
    Messages:
    133
    493 + 5 formes of Rotom + Shaymin-S + Giratina Origin = 500.
  10. devoninja

    devoninja

    Joined:
    Nov 8, 2007
    Messages:
    254
    + Deoxys forms =/= 500

    More likely it's an arbitrary number which only affects how big the final numbers are without having any statistical meaning itself.

    If you had 5000 instead all the results would be 10x bigger but would mean the same thing.

    For clarification, in this instance 'OU' does not mean "banned from UU but not from Ubers" rather it refers to pokemon which are Overused within their tier, such that there would be a number of 'OU' pokemon for the UU tier, correct?
  11. Evil Hamster

    Evil Hamster

    Joined:
    Sep 7, 2007
    Messages:
    579
    It if was that, 500 would only apply to the uber tier, it would have to be reduced for the other tiers to accomodate all the Pokemon banned in the tier. It's probably just an arbitrary number to make the statistics more sensible numbers as opposed to tiny fractions.
  12. CardsOfTheHeart

    CardsOfTheHeart

    Joined:
    Aug 21, 2007
    Messages:
    1,970
    The actual number I think we're looking for is 505, but we've already established that it's not important. The focus is on the relative centralization between the tiers--particularly between Uber and the other tiers.

    EDIT: 505 = 493 + 3 alternate Deoxys forms + 2 alternate Wormadam forms + 5 alternate Rotom forms + 1 Giratina-O + 1 Shaymin-S
  13. X-Act

    X-Act np: Biffy Clyro - Shock Shock
    is a Site Staff Alumnusis a Programmer Alumnusis a Smogon IRC SOp Alumnusis a Researcher Alumnusis a CAP Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis an Administrator Alumnus

    Joined:
    Feb 17, 2006
    Messages:
    4,675
    To answer how the 500 came to be, I need to explain how the whole formula came to be. Feel free to skip reading this post if you're not interesting in how the formula was created, as it is rather long.

    For me, it seemed obvious enough that a centralised metagame should have a high F and a low O, while a metagame that is not very centralised should have a low F and a high O. (Remember: O is the number of OU Pokemon while F is the number of frequently used Pokemon.) For example, in one case, a particular Standard metagame had F=6 and O=50, while a particular uber metagame had F=14 and O=28. As you can see, the Standard metagame has a low F and a high O, while the uber metagame has a high F and a low O. However, intuition isn't always right, so I kept my intuition to myself until I could prove that it is correct.

    To test my intuition, I decided to find the number of Pokemon that are used in 1 out of every T = 1, 2, 3, 4, etc. teams, up to 20. (Remember: a Pokemon is frequently used if it is in one out of every 4 teams, while a Pokemon is OU if it is in one out of every 20 teams.) Then I plotted these numbers against T and compared different graphs obtained from different metagames.

    I noticed that the graphs for the standard and UU metagames were very similar except that the UU graph was slightly steeper, while the uber metagame graphs were much less steep. Since 'steepness' is determined by gradient, I thought "so centralisation is related to the gradient of this graph; the higher the gradient, the less centralised the metagame is".

    So what I did at first is to plot a trendline of these points and find the gradient of this trendline. However, this whole process seemed to be way too complicated for the average Smogoner to understand, and the teacher in me decided that this process, although sound, needs to be simplified. :)

    Then I remembered that I had already mentioned that frequently-used Pokemon are those used in 1 out of every 4 teams (i.e. the point where T=4). So I connected a line between this point and the 'OU' point (i.e. the point where T=20) and found that the gradient of this line is very near that of the trendline. Thus I found a much simpler way of finding a very good approximation of the trendline gradient.

    There was, however, a problem: the higher the gradient, the LESS is the centralisation, and vice-versa... which means that the gradient is inversely proportional to the centralisation. This problem was solved easily by taking the reciprocal of the gradient as being the measure I wanted. Alternatively, you could argue that I found the difference in x divided by the difference in y instead of the other way round.

    Okay, so I had a point at (4,F) and another one at (20,O). The difference in x is 16, while difference in y is (O-F). Thus, my initial measure of centralisation was 16 / (O-F). I soon realised that this number is too small for practical purposes, however, so I just multiplied the number by a constant. After testing a few numbers, I settled for a constant equal to around 30 to 32, and since 30x16 = 480 and 32x16 = 512, I decided to use 500 as the constant. Finally, I just removed the numbers after the decimal point so that the measure is a neater number, and the formula floor(500 / (O-F)) resulted.

    The fact that (O-F) is in the final formula confirms that a centralised metagame has a low O and a high F, while a non-centralised metagame has a high O and a low F. If O is low and F is high, then O-F is a small number, and hence 500 / (O-F) is high. If O is high and F is low, the O-F is a large number, and hence 500 / (O-F) is small.
  14. Caelum

    Caelum qibz official stalker
    is a Site Staff Alumnusis a Smogon IRC AOp Alumnusis a Forum Moderator Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis a Battle Server Moderator Alumnus

    Joined:
    Apr 5, 2008
    Messages:
    1,656
    Thanks X-Act. I initially assumed you did a method similar to the one you described (the form of the equation was a bit of a tip off) but I just couldn't figure out how you'd get 500, didn't think about multiplying by a constant :nerd:. Nice work as always.



    (btw; for American high school students here; Gradient = Slope).
  15. Chou Toshio

    Chou Toshio @Fighting Necktie
    is a Forum Moderatoris a Community Contributoris an Artist Alumnusis a Contributor Alumnusis a Smogon Media Contributor Alumnusis a Battle Server Moderator Alumnus
    Moderator

    Joined:
    Aug 16, 2007
    Messages:
    8,242
    Nice Job X-Act. If you have a measure of centralization, I forgot that if you track it month by month we can see how stable things are/aren't and that goes a long way towards making an opinion about the centralization. Awesome Job. This is greatly appreciated.

    We can see here that despite all the weird tests we do, OU is pretty damn solid. UU varies in centralization levels more! Then again there are fewer players (less data) so that makes sense.
  16. Bourbon

    Bourbon

    Joined:
    Sep 10, 2007
    Messages:
    3,047
    Can you do this for pre september, like for all the months when Garchomp was unbanned, so we can see what sort of effect he had compared to now?
  17. cim

    cim happiness is such hard work
    is a Smogon Media Contributor Alumnus

    Joined:
    Jun 3, 2007
    Messages:
    5,412
    Less than you'll think, Skiddle. He didn't really limit the pool of usable POkémon, just changed it. If I recall correctly.
  18. wildfire393

    wildfire393

    Joined:
    Oct 3, 2008
    Messages:
    648
    It would be interesting to see some variations on that formula to check for few different things. For example, testing for weighting on O and testing for weighting on F. Like 500/(2O-F) and 500/(F-2O). This could show us whether the standard metagame has been centralized "the same amount" due to having a near identical number of Overused and Frequently Use Pokemon in the metagames or some other factor.
  19. jamespicone

    jamespicone

    Joined:
    Oct 12, 2008
    Messages:
    139
    I'd like to see the metagame where all frequently used pokemon are OU - it's infinitely centralised!
  20. Pisti

    Pisti formerly Diploman

    Joined:
    Jun 14, 2008
    Messages:
    1,548
    It would be 507 in that case, the 2 alternate Burmy forms.
  21. Mario With Lasers

    Mario With Lasers Self-proclaimed DEAD king
    is a Forum Moderator Alumnusis a CAP Contributor Alumnus

    Joined:
    Nov 15, 2006
    Messages:
    7,244
    Actually, one should expect to have more centralization in UU with fewer players, just like Suspect, no?
  22. CardsOfTheHeart

    CardsOfTheHeart

    Joined:
    Aug 21, 2007
    Messages:
    1,970
    Wrong. The only difference between the Burmy forms is sprites; there's nothing functionally different about them in battle, thus they are still considered to be one Pokemon. Besides, if we counted them, then we would also have to count Shellos's and Gastrodon's East Sea form.

    *end off-topic chatter*

    I just realized I haven't complemented you for this. Great work as always, X-Act!
  23. X-Act

    X-Act np: Biffy Clyro - Shock Shock
    is a Site Staff Alumnusis a Programmer Alumnusis a Smogon IRC SOp Alumnusis a Researcher Alumnusis a CAP Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis an Administrator Alumnus

    Joined:
    Feb 17, 2006
    Messages:
    4,675
    No. The amount of players in a metagame does not affect the centralisation of that metagame in any way.

    What affects the centralisation of the metagame is what the players use. If there are 1 million players using exactly the same team, then obviously that metagame is ultra-centralised (F=O=6 in that case, so Central would be infinity, lol). Conversely, if there are 100 players using teams having very little in common, then the metagame is very non-centralised (F would be close to 0 and O would be a large number, hence 500 / (O-F) would be very small).

    Suspect is inherently more centralised than Standard because there are many players who try new things on the Standard ladder, which decentralises the ladder. In Suspect, people have much less time to do this, as they're vying for a 1655/65 rating, so they tend to use their tried-and-tested teams instead of innovative teams.
  24. jamespicone

    jamespicone

    Joined:
    Oct 12, 2008
    Messages:
    139
    Extremely small numbers of people could influence centralisation, though - if only two people play UU, and don't change their team mid-month, only 12 pokemon at max have been used in the ladder - and all 12 of them are OU and frequently used, because they're in half of the teams.

    Basically, the less players there are, the less scope there is for different pokemon to be used - but that effect, I imagine, only exists for very small playerbases.
  25. eric the espeon

    eric the espeon maybe I just misunderstood
    is a Forum Moderator Alumnusis a Researcher Alumnusis a CAP Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnus

    Joined:
    Aug 7, 2007
    Messages:
    3,694
    That effect can be seen quite clearly in Septembers LC stats which had 11 Frequently used Pokemon but only 32 OUs. This was probably at least partly due to the fact that only 321 battles took place in that time.
Thread Status:
Not open for further replies.

Users Viewing Thread (Users: 0, Guests: 0)