Page 1 of 4 123 ... LastLast
Results 1 to 10 of 31

Thread: Eight queens puzzle in CWL MMA

  1. #1
    Senior Member
    Join Date
    Oct 2017
    Posts
    417

    Eight queens puzzle in CWL MMA

    I believe current CWL MMA is related to 8 queens alg.
    If you don't know what 8q is , please reference:
    https://en.wikipedia.org/wiki/Eight_queens_puzzle

    In each tier, there might be < 12 match models...,
    All players will be filled into this model from their tier, if full, then go.
    What do you think and how SC implement CWL MMA in 8 queens ?
    Last edited by frainlll; October 31st, 2018 at 12:34 AM.

  2. #2
    Millennial Club
    Join Date
    Apr 2016
    Posts
    1,415
    Quote Originally Posted by frainlll View Post
    I believe current CWL MMA is related to 8 queens alg.
    If you don't know what 8q is , please reference:
    https://en.wikipedia.org/wiki/Eight_queens_puzzle

    In each tier, there might be < 12 match models...,
    All players will be filled into this model from their tier, if full, then go.
    What do you think and how SC implement CWL MMA in 8 queens ?

    Interesting view, but I dont thin SC is using those kind of algorithm, because

    1. Mismatch is not as big as an issue to SC eyes compared to normal war
    2. Forcing all clan to only 15 size greatly increase the pool
    3. There is no need to check whether each of 8 clans in a group has to fill certain roles or constrain
    4. SC certainly want anything as simple as possible as it work faster and easier to patch
    https://www.clashofstats.com/clans/a...VLPP80/members

    Engineered clan focused on crafting optimized bases

  3. #3
    Pro Member nobeone's Avatar
    Join Date
    May 2015
    Location
    Deepest darkest Devon
    Posts
    696
    I don't quite understand, I think they used something like normal war weight to put you in a tier, then just bunged clans in that tier together as they hit search? In future they will just bung clans in the same tier together as they hit search? I don't think it is any more complex than this:

    a) If you have not cwl before: what is the war weight for top 15 bases, fit that into a tier in some way
    b) Take your tier, wait until there are 7 other clans searching, chuck you all in together
    c) matched ...

  4. #4
    Forum Veteran
    Join Date
    Jul 2017
    Location
    NE England, UK.
    Posts
    1,892
    Quote Originally Posted by nobeone View Post
    I don't quite understand, I think they used something like normal war weight to put you in a tier, then just bunged clans in that tier together as they hit search? In future they will just bung clans in the same tier together as they hit search? I don't think it is any more complex than this:

    a) If you have not cwl before: what is the war weight for top 15 bases, fit that into a tier in some way
    b) Take your tier, wait until there are 7 other clans searching, chuck you all in together
    c) matched ...
    Forgive me for being naive (not that I suggest Nobe is lol), but I tend to agree fully with him.
    I just cant see any reason for SC even wanting to try for anything more than tolerable initial matching, when the system is unweighted once a clan is in the said system.
    Basically, search as if doing a 15v with your top bases, then let the clans duke it out going forward.
    I applaud the cleverness of some of the guys here. I read a lot of stuff I never even heard about before. But in this instance, I just cant see any reason to do much at all.

  5. #5
    Millennial Club Thegreatpuma's Avatar
    Join Date
    Mar 2016
    Location
    Druidia
    Posts
    1,235
    yeah agree with the last two posts. why waste the time to code some sort of algorithm when random matching within tiers is perfectly effective?
    Last edited by Thegreatpuma; October 31st, 2018 at 12:44 PM.

  6. #6
    Senior Member
    Join Date
    Oct 2017
    Posts
    417
    I think in same tier, matchmaking still works for competition issue.
    It's not simply same tier then match.
    Let's see it in next season.

  7. #7
    Quote Originally Posted by frainlll View Post
    I think in same tier, matchmaking still works for competition issue.
    It's not simply same tier then match.
    Let's see it in next season.
    SC has made it very clear there is no matchmaking within the tier. The top 15 weight was used to place clans into a tier, but there is no matchmaking within the tier itself. Not only have they said it specifically, the results of matches prove it to be true.
    Contact SC here. Click here to see how trophies are calculated. My tips/thoughts on Legends here. My legends hope: 14 hours daily search limit to reduce advantage of account sharing; black clouds screen; Legends3 5000-5499 cups, Legends2 5500-5999, Legends1 6000+, while in clouds let us watch war, donate, do FCs, base design. Expand trophy range for matching down by 500 cups, reduce matching up by 500. Other suggested changes found here.

  8. #8
    Forum Veteran MajorJohnson's Avatar
    Join Date
    Apr 2017
    Posts
    1,509
    My impression is that the matchmaking for placement in tiers is based solely on defense weight rather than taking account of offense weight.

    For us, we were placed in masters 3 and I had my 1 cannon wonder in the lineup but isn't ranked in the top 15 but use it because of its extra firepower.

    Our matches have been pretty even, some have more TH12s than us but weaker attackers overall, some have a little less 12s but better attack/strategy.

    This has been a good spot for us, we will either rank #2 or #3 overall but we don't want to move up so we have been discussing throwing a war if #2 means we will move up.

  9. #9
    Forum Veteran
    Join Date
    Jul 2017
    Location
    NE England, UK.
    Posts
    1,892
    Quote Originally Posted by MajorJohnson View Post
    My impression is that the matchmaking for placement in tiers is based solely on defense weight rather than taking account of offense weight.

    For us, we were placed in masters 3 and I had my 1 cannon wonder in the lineup but isn't ranked in the top 15 but use it because of its extra firepower.

    Our matches have been pretty even, some have more TH12s than us but weaker attackers overall, some have a little less 12s but better attack/strategy.

    This has been a good spot for us, we will either rank #2 or #3 overall but we don't want to move up so we have been discussing throwing a war if #2 means we will move up.
    But if they used purely def weight, then would you not see more parity in the hall numbers of the top 15. I feel they have simply used the normal MM for matching. I cant see why they would change things for a one off spin, other than if they simply tried to match hall colour as closely as possible. They seem to have taken the easy route with cwl.
    Or are your accounts engineered, which would distort match results, so far as hall colour is concerned, to be thinking of defensive matching?

    One other thing occurred to me. SC said that engineering would be at a disadvantage. Without going into the pros and cons of inclusion, why would they say that, then start the first event using a MM that actually favours them?. It was a def MMA that gave rise to lop siding, and crippled the war scene. I dont expect they would risk cwl failing at season 1, for the same reason mate.
    Last edited by joshsgrandad; November 1st, 2018 at 12:17 AM.

  10. #10
    Forum Veteran MajorJohnson's Avatar
    Join Date
    Apr 2017
    Posts
    1,509
    Quote Originally Posted by joshsgrandad View Post
    But if they used purely def weight, then would you not see more parity in the hall numbers of the top 15. I feel they have simply used the normal MM for matching. I cant see why they would change things for a one off spin, other than if they simply tried to match hall colour as closely as possible. They seem to have taken the easy route with cwl.
    Or are your accounts engineered, which would distort match results, so far as hall colour is concerned, to be thinking of defensive matching?

    One other thing occurred to me. SC said that engineering would be at a disadvantage. Without going into the pros and cons of inclusion, why would they say that, then start the first event using a MM that actually favours them?. It was a def MMA that gave rise to lop siding, and crippled the war scene. I dont expect they would risk cwl failing at season 1, for the same reason mate.
    Pretty much all our bases our engineered, we are warring in our sister with our 12s and heavier style 9.11s. Of our top we have 5 th12s, 1 eagle TH11, 2 TH10.5s, 6 TH9.11s, and a 1 cannon wonder. While i may disgress that it may not rely fully on defense, it does seem to put more emphasis on it than regular matching. In our 7 wars we tend to be down more eagles and TH12s but the initial matching is pretty even.

    We also have a max 10 and some 9.5s in the lineup but left them out in favor of some more offense over defense.

    We will place 2nd overall winning 5 of 7 wars but of the 2 we lost, the one against the top clan was close and it was the most maxxerish clan, the other one we lost due to a member missing an attack.
    Last edited by MajorJohnson; November 1st, 2018 at 05:18 PM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •