TW matchmaking [MERGE]

Replies

  • I don't think "perfectly dead even" matches are realistic either, but a 9-10M GP gap? There are thousands of guilds so I find it hard to believe matchups cannot be avoided.
  • It's all about the fair match. They provided us with no information of our opponent except their name, no number of joined person, nothing.
  • _ho0ligan_
    31 posts Member
    edited December 2017
    path to contentment- expect less and accept more ;)
    find joy in the challenge and understand that one never loses, they either win or they learn. If you feel your always losing then your choosing to ignore the lesson before you.
    I learned after months of playing that there are team comps and strat involved in 'winning'
    I run an arena team that is 10K-25K score lower than the rest of my shard and still hold my relative position.
    So take some time away from playing and do some research into the work that many others have put into the game and farm certain toons that make up 'winning' comps.
    Understand each toons kit and how to mod them so they ehance or generate synergy.
    As far as our first TW went, we were 25 memebers up against a guild with a higher GP than us and we set up our defense well and by the end of the match we had destroy 5 zones punching a hole through the middle to the back column and had 1 team left in a 2nd zone to in their back column. It came down to team comps that we set up on defense and strategic use of attacks on offence. Unfortunately we missed the second TW by not meeting the 25 min.
  • NicWester wrote: »
    It’s only 8m. That’s a 14% increase. High end guilds can tie easily.

    Lol, are you listening to yourself, 8 million could equal to tons of extra teams
  • You can figure out the number of people from the lower registered guild by number of defensive slots. Defensive slots are 50% of the lower number of registered players.

    If you have 50 players in TW and the defensive slots show 13 it means they have 25-26 players and you need to stack heavier on defense, and/or just be resigned to not winning.

    Basically, they match based on GP of registered players regardless of how many. So if they have 25 players with a total of 40m GP and you have 50 players with a total of 40m GP, you will be completely out matched... Your saving Grace is the number of squads and the ability to just absorb their attackers... That being said, it's possible to wind up with 25 players at 1.6m GP with 70-80 good chars vs 50 players at 500k-600k with ~20-25 good chars... So even the.largee number of players doesn't offset it.
  • NicWester wrote: »
    It’s only 8m. That’s a 14% increase. High end guilds can tie easily.

    If a top arena team is between 105k to 90k power and if we were to say that a top team is on average 100k which on the high end. Then do the math. 8m/100k is 80 top arena teams. That is enough to fill up more than 3 full territories...

    8 million gp is a massive difference in terms of power and the guild with that kind of advantage will win 100% of the time. I don't have the slightest doubt about that.
  • Staubie12 wrote: »
    My guild is 30 gp so 9 gp more is significant. I don’t see your premise as a problem.

    That's because you fail at math.
  • Kyno
    32087 posts Moderator
    B0untryHR7 wrote: »
    NicWester wrote: »
    It’s only 8m. That’s a 14% increase. High end guilds can tie easily.

    If a top arena team is between 105k to 90k power and if we were to say that a top team is on average 100k which on the high end. Then do the math. 8m/100k is 80 top arena teams. That is enough to fill up more than 3 full territories...

    8 million gp is a massive difference in terms of power and the guild with that kind of advantage will win 100% of the time. I don't have the slightest doubt about that.

    not all teams are created equal, a more powerful clone team is still just a clone team (under AI control). there are limits to what a team can do no matter the number associated with it. 8M-10M is really not that much.
  • Galahax wrote: »
    Similar story of Lukes Tauntaun...
    Our guild had 48/50 players signed into the TW. We have a GP of 79 Mil (so thats with everyone mind you...). We got steam rolled in the first few hours. Took a look at opposing guild on swgoh.gg and found they had 94 Mil GP. Thats TWO different brackets of GP that somehow the computer deemed as "appropriately matched". I can understand a +/- 5 Mil difference, but 15??? Come on! Matching making to be balanced my foot!

    It's simple. Only 42/50 signed up from their guild, giving them an EFFECTIVE GP of 79m. It was an 79m vs 79m matchup. Each of.their players was just proportionately stronger.
  • phatphil3 wrote: »
    Territory wars is showing us how our guild matches up against the rest of game. I think there are some "broken" mechanics but a lot of it comes down to strategy. The guild we are currently facing has a lot of zBarris teams. They put them all on the first two territories and we have yet to pass through one. Kudos to them. Lesson learned for us.

    Agreed entirely, except that they are matching players with hugely disparate GPs because of the crapass algorithm.
  • Our first match we were 12 million higher GP than our opponent, end result was a tie. This last match we were about 5 million less than our opponent, end result was also a tie... as in arena those power levels do not mean all. It's all in how you use what you have... not to mention there is a lot of "padding" done these days to up guild GP. I guess my point would be those differences are mostly in your head. I wouldn't lose sleep over second place rewards either way. An extra zeta is really the main difference...
  • 501stsquad wrote: »
    That's just happened to my guild. We're about 75M gp with 100% participation matched up against a 100M gp guild. The battle was pretty close, but in the end, even with less participation the amount of g12, gk, zaris, chase in the other guild created a great advantage against us.
    I think not only the gp should be considered for match making but also the number of participants.

    That's exactly what I am saying. Right now the number of participants DOES matter, but it's in favor of the lower participant group. You had 50 people, they only had 37 people. They win.

    Simple...
  • There seems to be guilds which are intentionally benching 10 players each tw for gaming the matchmaking system. This pretty much guarantees a win because of their deeper rosters possesed by each individual members. And everyone gets more rewards after a cycle of 5 tw. For addressing this exploit matchmaking system could give more weight to standard deviation between guilmembers gp's and average gp of a guild. I am no specialist in designing matchmaking algorithms. I am being very generous when I am calling this a strategy. Guild alliances engaging in such practise are big name alliances, which makes it even worse. I am unaware which moderator or dev should be tagged here. I hope this issue gets the attention it deserves.
  • Agreed entirely. I have another thread about exactly this problem, and the number of apologists or people who can't do math and don't understand why this is a problem is absolutely shocking
  • They should have just went with total guild gp. Not participation. I personally haven't ran into this but I thought this would be a problem even before they started. Guilds wouldn't boot players to do this.
  • ProximaB1_ wrote: »
    They should have just went with total guild gp. Not participation. I personally haven't ran into this but I thought this would be a problem even before they started. Guilds wouldn't boot players to do this.

    Going with total gp can also be gamed. They can just drop 10 members before each tb registration and bring them on board after registration is done. They might miss out 1 hpit if that guild is a 30k ticket daily guild, but with the current model such exploit is remains beneficial. Either the matchmaking algorithm has to be reworked or they need to introduce structural barriers to hinder such behaviour(such as registering 2 tw at once/ increasing guild joining time after leaving guild etc.)
  • @Blizzisme I just went through your thread. Major facepalm.
  • Guilds are exploiting the participation option. If you leave out a handful of players your gp drops substantially. But your squads are better overall giving you an easy win. With the tie issue for top tier guilds this could become a big issue. Why tie when you can win.
  • @Blizzisme I just went through your thread. Major facepalm.

    Yeah, people can't do math...
  • ProximaB1_ wrote: »
    They should have just went with total guild gp. Not participation. I personally haven't ran into this but I thought this would be a problem even before they started. Guilds wouldn't boot players to do this.

    Going with total gp can also be gamed. They can just drop 10 members before each tb registration and bring them on board after registration is done. They might miss out 1 hpit if that guild is a 30k ticket daily guild, but with the current model such exploit is remains beneficial. Either the matchmaking algorithm has to be reworked or they need to introduce structural barriers to hinder such behaviour(such as registering 2 tw at once/ increasing guild joining time after leaving guild etc.)

    Another problem is the tie at high end gp. Guilds that think they will tie will go for the win. The easiest way to do that is threw lowering participation.
  • Similar thread in flight on Reddit. I see this becoming a concern. Especially to those in the 60-90M GP range. It is looking like a a gameable strategy.

    https://www.reddit.com/r/SWGalaxyOfHeroes/comments/7hmtfr/are_tw_pairings_bunk_and_gameable/?st=JAT7ZPOG&sh=ba848742
    Former Star Wars Uprising General Multiple SB Winner + Cartel Master Of The Nightsisters // Guild Master Of "Exiles Of Dathomir"
  • How are you determining the other guild's GP?
  • Peer
    299 posts Member
    Blizzisme wrote: »
    @Blizzisme I just went through your thread. Major facepalm.

    Yeah, people can't do math...

    Maybe this is because your initial calculation is wrong? Or because you imply that reducing the number of active players leads to victory automatically? This "strategy" you talk about only works if a guild actually has enough members with the desirable distribution of GP over players/toons.
  • Mattikin wrote: »
    How are you determining the other guild's GP?

    You can look it up via guild search. Yes you can get to a tie against a guild with 10 million or GP but you have to accept that you can't win so the best you can try to achieve is to burn through their defense with strong offense teams and let the match end in a tie.

    And yes 8 to 13 million GP is a HUGE difference...
  • SentinelBorg
    60 posts Member
    edited December 2017
    As the matchmaking seems to work insides those GP brackets there are also guilds that try to stay exactly below the threshold for the next bracket. Our last enemy for example was at over 120M GP before the TW, then during the preview phase (they registered with 50 members) they dropped to 119.7M GP and now a day after they are back at 120.5M GP again. ^^ During the fighting we saw for example G11/12 GK with zero mods standing around, which leads me to the conclusion, that this was on purpose.
  • Corvus
    190 posts Member
    edited December 2017
    Peer wrote: »
    Blizzisme wrote: »
    @Blizzisme I just went through your thread. Major facepalm.

    Yeah, people can't do math...

    Maybe this is because your initial calculation is wrong? Or because you imply that reducing the number of active players leads to victory automatically? This "strategy" you talk about only works if a guild actually has enough members with the desirable distribution of GP over players/toons.

    It will work for every guild with 10 to 20 strong players with lots of g12 toons. it is an exploit to the matchmaking system itself but I doubt CG can do much about that. The only way to get rid of this is to calculate the number of g12, g11, g10 toons and find a match based on this. But I dunno if that is doable and it might also be flawed in some way...

    This will be even a bigger problem once you get banner for succesfull defenses...
  • If you bench 10 players though, won't they miss out on the rewards?
    If my guild asked me to sit out the TW and miss out on rewards, I wouldn't be with them for very long!
  • Ariella wrote: »
    If you bench 10 players though, won't they miss out on the rewards?
    If my guild asked me to sit out the TW and miss out on rewards, I wouldn't be with them for very long!

    Yes they will but if you rotate them it might work out posivtive in the end, and you could of course always leave your weaklings out...
  • We are a 50/50 member guild, we lost our first TW but won our second TW. In the first one we had 41 members play and in the 2nd we had 37. Not intentional as the way i see it i would like all members to play which isn't always possible.

    I wouldnt want our guild to win and then be accused of not fielding a full 50 team. We generally get between 35-45 members participate and raids and TB, we do try and remove and replace inactive members.

    However i don't think its underhanded tactics and intentional for all guilds!
Sign In or Register to comment.