Season 3 - Match Up Discussion

Discussion in 'Wars' started by RedHotVixen_MaCHiNE, Jan 22, 2014.

  1. Everyone complaining should sit out s3, then grow to the size of the ones yall are complaining about
     
  2. u forgot about the number of no matches
     
  3. number of no matches just proves our point and its showing that more and more ee clans are sitting out or going with the stacked roster... and of them few stacked rosters even match each other and when they do... one side complains the other has "too many gh" or "how did we get a match with them they are too big". im happy there are high ammount of no matches means clans are listening and / or going through the stacking system either way... proof is in the increased number of no match/low hit ratio and the fact there there are less clans warring.

    Still support op.
     
  4. Theres a lot of clans that are sabotaging their rosters with statless accounts that end up inactive .So basically destroying good chances at war matchups in order to get a better matchup.


    How i know this? Im looking at all rosters. Look for yourselves and see where the problem lies already i have over 40 names of inactives in 10 wars
     
  5. there called plunder busters... waste ur spies and atk's on a drone that will make no gold all war... lol the 30k ones crack me up... giving the bigs another 3/4m stats to work with...
     
  6. I agree a 100% with OP as well as with her previous thread, can't war with your family anymore coz o the current trend, heck have an 8m cs minimum cap for these EEs.
     
  7. I really am glad this has been brought up by war hardened warriors like KOTFE, as I thought only rather new casual EE family clans had the problem of getting hammered with a normal type of roster.

    Warring with an alt in Vikings in Pyjamas, I had a look at the list of published roster strength after our matchup in war 1. We were stuck in between Wingless Angels and Rising Hawks, while having rather low BFE and BFA. We probably were lucky to get smashed by Wingless Angels instead of RH...

    I fear that EE will slowly die out and the only remaining participants will be two dozens of specialized EE clans with stacked roster if nothing changes to the mechanism. The barrier to new entrants is extremely high and frustrating the few remaining non stacked clans will not help.

    Honestly, as the matching mechanism seems totally off anyway, why not make it fully random? At least noob clans would have a slight chance to get a similar opponent. Why not include (individual?) prestige in matchups?
     
  8. I'm not so sure the match ups are that bad, there's only so many clans warring so there will always be a biased one way or the other.

    The problem as mentioned is pay, if the pay is fixed then mid size builds will become more common, improving the match ups, hopefully more would war to increasing the number of matching options.

    Overhauling the plunder structure is a bit extreme though, it works fine outside of EE. Also I'm not sure but introducing a entirely different plunder structure just for EE would be quite hard to code I think. Thing is the original war system may have the answer, introduce a war tax, but rather than the tax being a percentage of what you earn make it a flat rate, say 50m for every successful hit/steal and use the war tax to work out the winners. Given this system is already built don't think it would do any harm to try it out for a few wars!
     
  9. Agree totally with vixen.
    EE wars should not be monopolised by the few 150 lb players and 13-18 sh /gh per war. It's the best part of the game. Should be for everybody who is prepared to make an effort of a decent war build.
    At the moment cs, bfa and bfe overall taken into account. This relies on all clans who are warring having a fair mix of all builds. An analogy being a 'normal/ Gaussian' distribution in cs. In truth there is a bimodal distribution of cs in these war clans. Peaks at two ends of cs spectrum to allow matching a clan with a normal cs spread.
    A simple statistical test could be matching a bimodal distributed clan with another of similar distribution. A number of statistical tests are available to detect these. Therefore these clans are always non matched or matched with each other.
    This does not require a big change in kaw mechanics. But does avoid an un-winnable war.
    Most builds are by definition mid builds. We need them to be interested in war. At the moment there is no interest in most clans to allow these players to war. Hopefully the above change will allow such builds to be looked again favourably.
     
  10. Looks like boycotting season 3 is getting alot of support and still no comment from the devs.
     
  11. Mr Singh, I like your comparison to statistical distribution models when looking at roster stats. I am pretty sure one could draw interesting conclusions if analysing rosters and individual builds in such a way.
     
  12. Recap of Season 3 week 1 :

    - trying to recruit unknown SH from wc that may be inactive or a mole;
    - spending half a day getting a 'perfect' unbalanced roster and getting a no match;
    - signing up without 10SH/2LB and getting a huge mismatch;
    - telling 3 year old builds to sit out wars so that their place can be taken by 3 month old accounts;
    - mercing on wc but getting zero responses because I'm neither LB or SH;
    - reading in forums devs won't even test anything relating to the matchup system until after season 3...

    I dont know about the rest of you but this will be a looooong 7 weeks

    Support to Vixen and this thread
     
  13. Support

    Fed up of matching LB stacked clans with an average roster. Devs are taking all the fun out. Season 3 is a complete failure so far: no reward worth the pain and ugly mismatches.
    Not worth trying.
     
  14. Support to Vixen 

    Something needs to give. The devs tell us to look at the clans 1 above and 1 below to try and match rosters. We got a match last night with a clan 11 spots above us, whose top 14 players were all bigger than our #2 player. Seriously? Lol! 
     
  15. A total of 6 matchups were made for war #15.  6. But don't worry guys... The devs have big plans for Season 4 I'm sure. Pathetic. 
     
  16. Ty DN_Kassio. I think an additional statistical test on clans warring will not delay any war matching or times. A relatively easy solution to a major problem.
     
  17. The solution requires people with skills to analyze statistical data and draw inferences from it. Then tweak the algorithm and the structure to bake that in.

    If you look at the way the Devs post those graphs, you can clearly tell they have no one at the ATA HQ who has a clue about number crunching, data analysis and maths. If they did this situation would never have occurred.

    Great minds think out of the box. With every update to a system, mechs, stats you do a test analysis to see beyond the update and check for all contingencies that could make or break the system. They released lands, building, equips, etc. But at no point in time did they test for future stats and how these builds will impact the game and players.

    Well I could go on and on but it makes no difference till the Devs cannot hire a couple of decently skilled Apes to fix the system. And frankly, it will take a herculean task because they did a pitch work issue after issue. GH issue being a point in the recent mix of things. They did not eradicate the problem, they made it humongous with the patch work.
     
  18. I see you watching Kaw_Admin...please you need to fix this.
     
  19. Not sure why u guys even make these posts. If u don't understand that the devs are too stubborn to admit that changing from the s1 algorithm to this new one has ruined ee, u haven't played enough. And also I blame most of you guys for this as well. Yeah we know u wanted to copy the original exploiters and have all these gh's, but u are a major cause of this yourselves. Either war or don't but if u actually think that the devs care 1 way or the other u are on crack.

    They won't change squat. Accept it and move on.