I am matched up vs a player that has 400k more character GP than me. How the hell is this a fair matchup? At least before when we matched straight GP you can see why your matched up. This **** isn't even right. Everyone in my GA shard has 350k to 400k more GP than me. Have you guys thought of "If than" statements for your coding? There are tons of players it shouldn't be this hard to find matches. Statement one Find a player with Gp +/- 50k GP difference. Statement two If character GP isn't also within 50k GP difference find a new match. Shouldn't be this hard to make a matching algorithm. There has been tons of practice for regular GA plus all that practice form TW to give what 1 or 2 years of testing to make a decent matching algorithm?
8
Replies
The first match GAC match algorithm is horrid.
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*
Where do i make complain that actually developers going to hear?
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*
You win the medal OP!
He's tied with a similar post on megathread sent at the same minute.
there shouldn't be more than 100k disparity tbh never mind 500k+
Sorry OP. You lost the medal!
The algorithm has been identified.
GP of # of defence toons x 2 will be very close within your 8-bracket.
Stop mentioning total GP differences please.
You can set X defence teams, each containing 5 toons.
For me, it’s 8 teams, so 40 toons on defence.
The matchmaking is counting GP of my top 80. And matching me with people who have a similar GP for their top 80.
Regardless of what they say the algorithm is some of these matchups just dont add up.
Can you showcase us how it doesn't add up? You can use DSR bot which auto calculates necessary efficient GPs in comparison. Are there mismatches there? See mine:
Each GAC we all start back at Carbonite League so we are are going to see some of these crazy-looking matchups.
After the first GA of the GAC, some players will have advanced to Bronzium League and others will not. When that is factored in, matchmaking for the second GA should look a bit better.
By the third and fourth GAs of the GAC, matchmaking should look pretty interesting for most players
I am going against an opponent with similar GP, but he has JKR, DR, Traya and the other usual metas. I don’t have any Revan or Traya.
And don’t tell me it is fair because the opponent just made smarter choices which toons to farm. With that argument you could pair a level 25 player with a level 85 player and argue it is fair because the level 85 player was smarter spending more time leveling up.
Not until meta teams are considered in matchmaking it will be fair. (If there is a thing like “fair” in that context)
You still can. Compare your 'relevant GP' to your opponent's 'relevant GP'.
Total (character) GP is irrelevant.
So is "relevant GP" :
https://forums.galaxy-of-heroes.starwars.ea.com/discussion/208118/why-relevant-gp-doesnt-work-for-gac-mm/p1
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*
The problem is no matchmaking will ever be completely fair. Even if you were matched against your identical twin mods and gesr would be different. It's just not practical or possible. Just look at every major sport and how lopsided some of the round 1 matchup are. This is the way of things.
What we tell you doesn't matter. CG may keep changing matchmaking methods. Even if they try 100 different methods, they will never even try one that takes into account specifics toons. It will always stay as player advantage.
That's a horrible analogy - in sporting events the first round match-ups of tournaments are supposed to be lopsided. Lol. That's what the regular season is for. The top seeds play the bottom seeds.
CG has specifically stated, several times now, that GAC is supposed to consist of "close matches". Imho, and probably most who are trying to be non biased, a match-up having a person with 7 g13 including a g13 maxed Malak team, a g13 GG sep team and a maxed GR team pitted against a guy with no g13, no Malak and a 6* Padmé with 1 zeta isn't going to be "close" - not by a long shot.
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*
Its not a terrible analogy. The matches will get closer and closer (if the ladder works correctly).
There is absolutely zero chance that every player will get a close matchup every round. Zero. Chance. They can make it as close as possible and it will get better as the ladder progresses. But your expectations are unrealistic, IMHO.
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*
Except that not all the relevant gp is being weighed. It only takes into account defense teams ×2. Someone that has a more expansive roster will have more versatility to attack that isn't being counted. And pilots will weigh more heavily with less benefits- think g12+ biggs and ackbaar. Finally, the person with the overall significantly higher gp tends to have played longer which translates to having had more time to mod farm.
You can set X defence teams, each containing 5 toons.
For me, it’s 8 teams, so 40 toons on defence.
The matchmaking is counting GP of my top 80. And matching me with people who have a similar GP for their top 80. [/quote]
Regardless of what they say the algorithm is some of these matchups just dont add up.[/quote]
Can you showcase us how it doesn't add up? You can use DSR bot which auto calculates necessary efficient GPs in comparison. Are there mismatches there? See mine:
[/quote]
Sure - here is an example. Higher GP and better toons.
Regardless of what they say the algorithm is some of these matchups just dont add up.[/quote]
Can you showcase us how it doesn't add up? You can use DSR bot which auto calculates necessary efficient GPs in comparison. Are there mismatches there? See mine:
[/quote]
Sure - here is an example. Higher GP and better toons.
[/quote]
Nah, no can do. It shows nothing about the algo not working as intended. You need to show two -effective gp- figures not matching with eachother. You may not like how the current algo works, but I've yet to see an indication it doesn't work.