The new ELO-based ranking system


  • @MrRoboto said in Proposal for a new, ELO-based, ranking system:

    A steady climb after his return brought him back to the top, although his rating seems to have plateaued around 2030

    @axis-dominion is so transcendent, he’s not just 1-2 moves ahead of you, he’s 6-7 years ahead of all of us.
    During 2024-2030, he wholly devoted his life to increasing his A&A skills, playing 500 more games.
    Word is, he just stepped out of his time machine and is about to share his extra knowledge with us poor slobs who plod on 1 day at a time. He’s only going to take the Axis, so it will finally, truly be…


    a time of the Axis Dominion



  • Most under-rated A&A league skill:
    Starting a game with a good player who’s going to drop all his games before yours is done

  • 2023 '22 '21 '20 '19 '18 '17 '16 '15

    For the OOB ranking most top players are not active. It should be a tool to filter away the non active players to achieve a current/up to date ranking of active players. How do ido that? If you were great in 2014, 16 and 17 and havent played a game since, then, I dont care what your ranking is

  • 2024 2023 '22 '21 '20

    Yeah, originally we were going to have a historic ELO for posterity and curious interest only and a second ELO for yearly playoffs then @MrRoboto explained we really don’t play enough games to have a yearly ELO.

  • 2023 '22 '21 '20 '19 '18 '17 '16 '15

    definately it should be possible to filter for active players. The current format is close to useless for browsing relative strength of active players

  • 2023 '22 '21 '20 '19 '18 '17 '16 '15

    I think there could be an historic rating like @AndrewAAGamer says and one showing active players. If you havent finished a game in 18?? months (or whatever) you should be moved to the historic sheet. If you re-appear you are moved back to active


  • @oysteilo said in The new ELO-based ranking system:

    definately it should be possible to filter for active players. The current format is close to useless for browsing relative strength of active players

    yeah, just make a filter where you can chose if you like all players or only active players


  • @oysteilo There already is one, but maybe you guys can’t use it right now.
    But maybe you can. Go to the filter views icon - is at top right next to the Sigma/summation symbol.

    If you can’t do that, you maybe don’t have the capability yet (I’m not sure if it’s married to edit game results capability)

    Have no fear, it’s a simple toggle.

    And as soon as I win or lose against you I will be active even though the other 56 games are all from many years ago

    Can you not see the tier color bands? It seems pretty easy to me to see the ranking of actives, but maybe you don’t have the same view.


  • And those top 11 inactives are pretty awesome players

  • 2023 '22 '21 '20 '19 '18 '17 '16 '15

    My bad, it works. I just did not understand/know. I am happy with this


  • It’s good to know everyone has that capability so thanks.
    Yeah I had to ask MrRoboto about it too - didn’t know what that box was

  • 2023 '22 '21 '20 '19 '18 '17 '16 '15

    maybe it is explained somewhere, but there are so many pages…Now I have a question about this formula

    090b963b-df43-4e58-bfa8-7e1b5ef0794d-image.png

    What is RB? Is it RBold? If it is RBold then what happens if two players with the same rating meet? Well then EA =1

    And putting EA = 1 here makes no sense for the winning person:

    efdef5ce-607f-40ba-ba78-ea70dfe9aa79-image.png

    So I am doing something incorrect. Maybe the skilled people here can show their work (just like we did in school) with a couple of examples


  • I don’t know the formula yet, but 2 new players have exactly 1500 each.
    With the current sensitivities, one goes up 55, the other goes down 55, if that helps.
    MrRoboto is the formula guy

  • 2024 2023

    @oysteilo said in The new ELO-based ranking system:

    maybe it is explained somewhere, but there are so many pages…Now I have a question about this formula

    090b963b-df43-4e58-bfa8-7e1b5ef0794d-image.png

    What is RB? Is it RBold? If it is RBold then what happens if two players with the same rating meet? Well then EA =1

    R(B) is only used in context with R(A) to find the Expected Outcome (should A or B win). So yes it is R(B(old)) if you want to think of it that way.

    And putting EA = 1 here makes no sense for the winning person:

    EA = Expected outcome for Player A

    If you have two players of equal rating, EA formula looks like this:

    EA = 1 / { 1 + 10^[(RB - RAold)/F] }

    RB - RAold = 0 since they are equal; formula then simplifies to:
    EA = 1 / { 1 + 10^(0/F) }
    10 ^ 0 = 1, so formula simplifies to:
    EA = 1 / { 1 + 1 }
    EA = 1/2

    EA is always 1/2 for equal ratings, this conceptually is that they both have a 50% chance of winning.

    EA will never equal 1, but it can get close to 1 if Player A is significantly higher rated than player B.

    As an example, if A is rated 2250 and B is rated 750, and F is 500; we have:

    EA = 1 / {1 + 10^[(750 - 2250)/F] }
    EA = 1 / {1 + 10^[-1500/500] }
    EA = 1 / {1 + 10^( -3) } = 1 / {1.001}
    EA = .999 or 99.9% expected chance of Player A winning

  • 2024 2023

    @oysteilo For another example, our comparative overall ELO is:

    Oysteilo - 1621 Player A
    Stucifer - 1549 Player B

    The expected outcome EA is:
    1/ {1+10^[ (RB - RAold = 1549 - 1621 = -72) / 500] }

    this piece 10^[ (RB - RAold = 1549 - 1621 = -72) / 500]
    is 10^(-72/500) = 10^(-0.144) = 0.71779

    EA = 1 / { 1 + 0.71779}

    EA = .58214 or 58% expected chance Oysteilo wins, 42% Stucifer wins

  • 2023 '22 '21 '20 '19 '18 '17 '16 '15

    @Stucifer Thank you! That explaines it. It is 10 exponetial. I read it as 10 multiplied with the fraction

    very helpfull

  • 2024 2023

    @oysteilo I made the same mistake the first time I was reading the formula too. :D

  • '19 '18

    Interesting that two people made the same mistake, despite the term being quite clearly above the 10.

    I just love that the last questions have been resolved so easily :D
    I wish all challenges worked like that


  • Fun to see this graph and read your narration of it, thanks for putting it together Roboto! yah i figured the few times where i was too addicted and needed to suddenly drop out for a while to regain balance in my life would take a toll on my record. I guess it didn’t seem as bad when at the time it was always resetting each year, but now with this new system it seems to matter more. Still debating if I’m ready to come back again!

    @MrRoboto said in The new ELO-based ranking system:

    And because it’s so much fun, some more data :-)

    Who has played the most BM4 games? That would be our beloved @simon33 who unfortunately went AWOL a couple of weeks ago.

    78a05bc7-7dac-4ddf-92b9-6ffcfc131977-image.png

    You can see clearly how he improved a lot, from his lowest point in March 2017 (1077 ELO) to his highest Rating in May 2019 (1693 ELO). Which is to be expected!

    Now from then on his rating gradually declined - slowly but steady. Did he become worse? Maybe.

    But remember, we are not measuring an absolute value. Your ELO rating is actually only a number that gives you a relation to the other players. So it’s not absolute, it’s relative to every other player.
    I’d interpret this graph differently: I think over the past ~4 years his skill level plateaued or stagnated but A) more and more other people did improve while he did not and B) more and more new people joined the league who are on average better than him.


    And now a completely different graph. The player with the 3rd most completed BM4-games, after @simon33 and @Giallo ?
    Our very own @axis-dominion

    Here is your graph, Mister:

    2db50aee-fe7c-41c0-ad67-f390dc53f18a-image.png

    A steady climb at the beginning where he chose mostly average players and won consistently almost all of those games.
    The opponents were weaker than him, but still reasonably close so the games were still worth something. The climb could have been a lot faster had he won against the top players from the get go - but he didn’t! The matches against already strong Adam514 back then were almost all losses.

    An unbelievable streak of 30 wins in 31 games between July 18 and March 19 (the only loss was against, of course, Adam514) rewarded him his all-time high of 2101.

    3 losses on 16 Apr 2019 (probably mass forfeit?) saw him drop a whopping 158 points before his break.
    A steady climb after his return brought him back to the top, although his rating seems to have plateaued around 2030


  • You’re so on to me gamer!!!

    btw bids are up in the 40 range for allies in that future, just fyi

    @gamerman01 said in The new ELO-based ranking system:

    @MrRoboto said in Proposal for a new, ELO-based, ranking system:

    A steady climb after his return brought him back to the top, although his rating seems to have plateaued around 2030

    @axis-dominion is so transcendent, he’s not just 1-2 moves ahead of you, he’s 6-7 years ahead of all of us.
    During 2024-2030, he wholly devoted his life to increasing his A&A skills, playing 500 more games.
    Word is, he just stepped out of his time machine and is about to share his extra knowledge with us poor slobs who plod on 1 day at a time. He’s only going to take the Axis, so it will finally, truly be…


    a time of the Axis Dominion


Suggested Topics

  • 25
  • 51
  • 44
  • 103
  • 136
  • 165
  • 163
  • 2.2k
Axis & Allies Boardgaming Custom Painted Miniatures

32

Online

17.4k

Users

39.9k

Topics

1.7m

Posts