Territory War Matchmaker

Replies

  • DarjeloSalas
    9944 posts Member
    edited November 2018
    Bigby wrote: »
    Although we don't get GP mismatches as such, our GP is quite high for the TW readiness. So we keep getting beat badly. In fact as a guild we are considering giving up and not even putting defensive teams down. Just take the rewards from the loss and not waste time fighting battles.

    Can the matchmaking algorithm not go beyond GP and look at true TW power such as G12, legendary and zetas?

    @Bigby just to save you from future heartache, if any guild member scores 0 banners total, for example by not setting any defensive teams and not winning any battles on attack - they will NOT earn any rewards at all.

  • Mhmacleod wrote: »
    Our guild: Glorious Geeks. 89mil GP. ~48 members at time of TW (we don't play with roster numbers for TW). 40 participating.
    Their guild: 90mil GP. No problem. Except they had 39 members. From the way the War is turning out, it's like all 39 were participating.

    Invite-only guild. Back to 48 members right now.

    Before:
    lyngi9ufocl9.png

    After:
    q2m3a7wft9i9.png

    I should probably say more, but still too triggered with TW and ships combined to compose myself tonight.

    Hello there. I’m in the same alliance as the guild you faced. While it may seem like a sandbag, was truly not. Players from that guild came to my current guild and another guild merged in. Because moves can’t be made during these modes it has to be before or after.

    Sorry for the confusion.
  • Condenserhead
    4 posts Member
    edited November 2018
    CLONES UNLEASHED TERRITORY WAR MATCHMAKING ISSUE 11/16-11/18
    General Information:
    • Clones Unleashed (Us) v. Marianas Trench
    • Date of Territory War: 11/16/2018
    • Clones Unleashed GP: 112 Million (2.1mGP Avg.)
    • Marianas Trench GP: 159 Million (3.2mGP Avg.)
    • Clones Unleashed TW Record Pre-100 Million GP: 31-5
    Territory-War Specific Information:
    • Clones Unleashed Members Joined: 50 (112 Million GP Joined)
    • Marianas Trench Members Joined: 43 ((Approximately 138 Million GP Joined, Based off of actual reported GP on swgoh.gg (31 listed accounts at 100.99, total) + an averaged GP generated from the remainder of the unreported accounts against the total remaining GP (12 unlisted accounts, for a total of 36.63 Million, or an average of 3.05 per member))
    • Clones Unleashed Zetas on Defense: 281
    • Marianas Trench Zetas on Defense: 401
    • Territory Size: 22 Teams
    • Clones Unleashed Per-Team Efficiency: It took 1.8 Battles to defeat an opponent’s team on defense (164 Battles to Destroy 4 Territories)
    • Marianas Trench Per-Team Efficiency: It took 1.4 Battles to defeat one of our defensive teams (217 Battles to Destroy 7 Territories)
    • Final Score: 11,407 – 15,372

    In sum, the (active/joined) GP disparity between our guild and our opponents not only means that they are in an entirely different reward bracket than us, but it also means that they could field approximately 260 teams of 100,000 GP over our entire active roster. The only logical conclusions based upon these numbers are (A) the matchmaking algorithm is flawed and/or using inaccurate inputs to determine opponents, and is thus matching up teams across tier brackets; or (B) the matchmaking algorithm is being compromised by savvy guilds who have found ways to circumvent the system in place. Regardless of the reason, the foregoing data proves that the current matchmaking system harms guilds by matching them against unbeatable opponents, and rewards guilds (at a higher reward tier, no less) that exploit the current system by providing them with greater gear rewards, zetas, and currency.
    This issue first started becoming an insurmountable problem once we eclipsed 100mGP. Until then, we were able to overcome opponents through discipline and strategy. Now, that is no longer the case. While the matchmaking system quantifies extreme mismatches based upon the number of members who join territory war amongst guilds, a 30 Million GP difference and a 1 Million GP average roster difference is too great a burden to overcome. Even though we have more people playing in the battle (conceivably), the teams we are up against have determinably higher gear level, a higher average roster level, and far more zeta-level abilities per team assigned. Because battles don’t operate on refresh (and TM is carried over per battle, meaning that each subsequent loss results in a head start for the defensive team), G12 and zeta-heavy defensive lines almost certainly outperform offensive fronts (keeping in mind that those defensive teams will be of a higher quality simply because the “active” rosters used by the matchmaking algorithm are approximately 1 ½ to 2 times more powerful than our average roster). This is the third TW in a row where we have faced an overwhelmingly difficult opponent.
    Based upon the forgoing, your matchmaking algorithm merely punishes guilds consisting of 50 dedicated, active players, while rewarding guilds consisting of casual ones (or guilds organized around half that number of dedicated players offset by alternate accounts and ticket hounds). As such, the current matchmaking setup for TW harms the entertainment and playability (and further, the monetarization) for a far greater number of players than it actually rewards, irrespective of pay-to-play status.
    I sincerely hope that this data can be used to better the current Territory War environment and fix a significant flaw that harms the entertainment value of your game. I would be open to discuss any further information (our guild documented just about every conceivable factor in the past TW), and look forward to your response.

    Thanks,Condenserhead,Clones Unleashed
    https://swgoh.gg/g/35094/clones-unleashed/
    i2pl06w9csek.png
    5xd4vp5q01km.png
    Post edited by Condenserhead on
  • and still no single reply from CG... really sad. There is a lot of people spending time to provide a lot of data and they are not even getting a statement or anything. Not even - thanks for your input, we are really sorry. This is indeed not how the match should work, we are looking into it and thinking about how to compensate you for the trouble...

    Sad performance CG
  • @CG_SBCrumb is there ANY update here?
  • Yes, it would be good to know if there is any progress. Our 98M GP guild just got matched with a 121M GP guild in the 80M-89.9M tier.I don't think it will be pretty.
  • 110 vs 140... wont post more details as it dosn't change anything.
    this is getting boring - we are not setting a dev this time.
  • abc127
    19 posts Member
    edited November 2018
    Speaking of: here is our current match

    Our guild
    46 joined
    Active gp 60M
    Total gp 64M
    Avg gp 1.3M

    Their guild
    31-32 joined (16 spots/sector)
    Total gp 124M
    Avg gp 2.5M

    cj4ptk324m1v.jpg
    rh0r0rymxcm7.jpg
  • Hey guys, what kind of bot you're all using to compare the stats of both gilds? I would be delighted if you can give me a link. Thanks in advance.
  • @waxweazle01 heres a link to read up on
    https://www.reddit.com/r/SWGalaxyOfHeroes/comments/9eu30e/dsr_bot_check_for_your_hstr_readiness_tw_matchup/

    and theres an update to Matchmaking just been announced starting next TE 29/11/18 just been announced ingame message
  • 1dcp2a7ctlhf.jpg

    Best news I've heard in a while!!
  • Date of TW: 27 Nov 2018

    Your Guild’s GP: 178m GP
    https://swgoh.gg/g/429/thailaxy/

    Opposing team’s GP: 201m GP
    https://swgoh.gg/g/39390/dark-lords-of-p/

    # of my guild members who joined: 49/50
    # of opposing guild members who joined: Don't know. But we have 24 team set on Defense

    wthHtpd.jpg
  • Date of TW: 27/11/2018
    Your Guild’s GP: 116mil
    https://swgoh.gg/g/17336/iron-dogs/
    Opposing team’s GP: 168mil
    https://swgoh.gg/g/2664/danish-pride/
    # of my guild members who joined: 50
    # of opposing guild members who joined: not sure
    b7wuwaagfvgo.png
  • Giodude2000
    240 posts Member
    edited November 2018
    My guild: II ANZGC II
    Opponent's guild: Braveheartz

    Date of TW: 27/11/18 or 11/27/18
    Your guild's GP: 135M
    Opposing team's GP: 198M
    # of my guild members who joined: 50
    # of opposing guild members who joined: unknown. There are 22 placements so I assume 44.

    9olep6auw6da.png
  • Giodude2000
    240 posts Member
    edited November 2018
    On another account:
    My guild: ANZGC
    Opponent guild: Dark Łords
    Date of TW: 27/11/18 or 11/27/18
    Your guild's GP: 123M
    Opposing team's GP: 138M
    # of my guild members who joined: 49
    # of opposing guild members who joined: unknown. 25 placements per territory.

    tu3uprjjkmja.jpg
  • Date of TW:27/11/2018
    Your Guild’s GP: 62,921,006
    3cfcctrr5bzb.jpg
    Opposing team’s GP:107,591,111
    vwm4dvs28k1i.jpg
    # of my guild members who joined:48/50
    # of opposing guild members who joined: ?/50

    We've been asked to fill up 18 teams in the regions.

    I don't think you need screenshots about the end of the TW

    Also the same thing happened with that guild :

    https://swgoh.gg/g/18349/legends-of-light/

    Fix that...thing... There is no point for us to keep playing in Territory Wars. And like us there are thousands of guilds with the same issues.
  • Hey Devs. I am the guild leader of Ascended Masters, and I have posted before.
    We are facing another guild that is almost 20 Million GP stronger than our guild.
    opposing guild: StarFrenchWar 83Mn4026u851lhf.jpg

    Our guild : Ascended Masters. 65M

    Please, please fix this.... We used to be excited, now we wonder what the GP gap is going to be....
    5m544pygn4yj.jpg
  • GenBonesworth
    35 posts Member
    edited November 2018
    zidcsr41pghc.png
    v1dpxq1p0d5w.png

    So we barely got past a zone and the other group has a full clear.

    Date: 11/27
    Us: 136M
    Them: 176M
    Joined Us: 47
    Joined Them: 42-44


  • Just a random thought on TW. If you are going to buff the heck out of a character for TW, how about we don't make it a jedi, thus reducing TW to "whichever team has the most number of Revan to pair with Mace wins."

    Again, crazy, I know - asking the developers to actually think through the implications of crazy buffs that ruin yet another game mode. :/

    Can we get a developer to respond to...well, pretty much anything. One quick question - do you guys intentionally try to make the game a worse experience or do you not even care?
    In game name: Lucas Gregory FORMER PLAYER - - - -"Whale blah grump poooop." - Ouchie

    In game guild: TNR Uprising
    I beat the REAL T7 Yoda (not the nerfed one) and did so before mods were there to help
    *This space left intentionally blank*
  • Date of TW: 16th-18th November 2018
    Your Guild’s GP: 89,073,033
    Opposing team’s GP: 133,871,904
    # of my guild members who joined: 42
    # of opposing guild members who joined: Unknown

    43+ million GP difference. Average GP 1 million higher per member.

    ey5ey9x0fkzp.png
    ajoxitnk3s9f.png
    0hhlbqprpjx4.png
    Date of TW: 16th-18th November 2018
    Your Guild’s GP: 89,073,033
    Opposing team’s GP: 133,871,904
    # of my guild members who joined: 42
    # of opposing guild members who joined: Unknown

    43+ million GP difference. Average GP 1 million higher per member.

    ey5ey9x0fkzp.png
    ajoxitnk3s9f.png
    0hhlbqprpjx4.png

    nqiy55ooeudh.jpg

    we had a similar mismatch to the same guild about a month ago we were 93 million they were 127million, but we wiped the floor with them. GP has little to do with your chances. it's about skill and strategy
  • Kol8Ajawa wrote: »
    I'm not seeing this update to matchmaking unless it was to make it worse...

    Lets see how the battle goes before calling out the devs on the new algorithm. We actually got a fair matchup when comparing what each guild as a whole has to bring to the battle.

    I ran your guild and your enemies guild in DSRbot to check the "quality" of the GP between you and your enemy. From what I can see, I have to say I think your guild has a far better chance of kicking their **** then they do yours lol. As long as you have a half decent strategy you should have a good matchup. I highly doubt they have their full roster available for the war. If they sandbagged you should be able to punish them for it.

    In fact when I run my guild vs the guild we're fighting we have a slightly higher overall GP (106m vs 103m), but the toons each side has at their disposable is pretty **** even.

    That said, i have seen some horrible matchups in another thread after the change. Clearly sandbagging still works for some guilds
  • DuneSeaFarmer
    3525 posts Member
    edited December 2018
    Looking at teams with Revan... yeah this is gonna be a lot of no fun Rumor has it whales wanted easier matches. Guess they got their wish.
  • 21/1/18 match up
    Blood Saber Legion vs Empire Crits back
    93m vs 112m
    48/50 vs ??/50 (23 per territory for both)

    5 in 6 TW we've been 20m disadvantaged. down 20% to your opponent at the start is starting to get real old real fast.
  • Legendaries really need to be taken into consideration in the matchmaking. We just beat our first HSTR and we’re playing against a guild full of Traya and some Revan. Simply putting GP vs GP just doesn’t give a balanced experience. Their mods and gear will be inherently better due to the rewards they’ve been accumulating theoughout their many HSTR successes that we’ve only just begun earning. 40 of their players will still have a huge advantage over 50 of ours. 1e89f4f9gigu.jpeg
  • Erebus2017
    50 posts Member
    edited December 2018
    New “fixes” don’t seem to have addressed anything:

    TW DATE 11/29
    Force Maestros
    Our GP 85M 46/50 participated 82 M GP enrolled
    Fist Full of Wookie
    Their GP 112M unknown on participation but we only deployed 17 per zone.

  • you guys will never be happy, just more excuses
  • Nevermind sand bagging guilds in TW, people need to stop sand bagging their TW match up complaints.

    The number of defensive teams you set is determined by the guild with the lowest number participating. It's half of those active members.
  • Hey @CG_SBCrumb do you think this thread should be closed, and a new one for the new algorithm be started?
Sign In or Register to comment.