3v3 Grand Arena Matchup Algorithm Needs to be "tighter" than 5v5
I am in a bracket of 8 where 2 guys have DR+Malak and 6 don't. No, this isn't a complaint - I (without) beat one of the ones who does. But there is over 100k difference in GP between several players. In 5v5 that difference is manageable. But in 3v3 that's almost 2 entire powerful teams (including DR/Malak).
My point is simply that when you have smaller numbers on teams, the spread in GP has to be lower, in order to keep things competitive.
That's all - just tweak the numbers so that 3v3 GP's are closer together than the acceptable range for 5v5. This seemed to be an issue last 3v3 as well.
My point is simply that when you have smaller numbers on teams, the spread in GP has to be lower, in order to keep things competitive.
That's all - just tweak the numbers so that 3v3 GP's are closer together than the acceptable range for 5v5. This seemed to be an issue last 3v3 as well.
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*
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*
3
Replies
Tighter matchmaking would be nice though.
Character as measured by the Discord bot. The other DR/Malak guy I'm facing this round has double the number of 6 dot mods and 24 speed mods 20+ to my 10.
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*
Oh no, I get that - especially in arena. It is my understanding (and at least implied) that GA is supposed to be more "evenly" matched. Squad arena is "keep up with Jones'" - and I totally get that. I don't think I'm entitled to have the same thing the DR/Malak guys do (I'll still try!)
I'm just saying if the goal of GA is supposed to be more even matchups, they need to tighten it up a bit for 3v3. Honestly, 5v5 is fantastic. Some real great GA matchups the last few times. Down to the wire.
3v3 seems a bit "off" is all.
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 may be better off looking in game.
I've never seen a gap of more than 10k, and rarely more than 2k, at the start of a GA round.
Just curious why you are looking at the discord bot instead of in the game? My understanding is that the discord bot sometimes hasn't updated for certain people for a while. And did you look when the GA started? It's also possible some people have done a lot of gearing in the last few days, but that won't impact the GA.
It seems odd to me because everybody in my grouping is within a few thousand difference in character GP. I could see there being such a big difference at the top GP levels because there might not be enough people to group together too closely, but otherwise that sounds fairly unusual.
For sure. I do appreciate 3v3 just for the sake of splitting it up. And I think that’s what they intended for a more even match up. Some squads are easier or beatable with out all their parts.
I’m 4.7 gp and some of these teams are out right dirty with speed but I do love when I see new configurations and clever teams.
My record is crap but I appreciate it
That's a fair point. The in game ones are closer - but there has also been debate about whether or not the in game displays (and even mods) are accurately "locked" at the start of GA as they are intended.
Part of my point was not simply overall GP - but those other factors as well (including DR/Malak). Just as an example (and why I prefer the Discord comparison) - here is what it lists for my current match (I'm the one on the left)
G12+5 :: 7 vs 17
6 dot Mods :: 56 vs 107
Speed+10 :: 160 vs 262
Speed+15 :: 45 vs 91
Speed+20 :: 10 vs 24
Off+100 :: 21 vs 33
While I wasn't trying to turn this into a complaint thread (more of a request to simply tighten it up in 3v3) - clearly, the person with 14 more 20+ speed mods has a big advantage. In 3v3 that's 4 entire team's worth. And 50 more 6 dot mods is enough to completely mod out 3 more teams. That's all I'm suggesting. Those differences in 5v5 are at least a little easier to overcome, as they tend to be spread out more and can be "strategized" around through good synergy/planning better.
Again, all of this operates under the assumption that GA is intended to be a relatively even match. The numbers above - along with the addition of DR+Malak (and the characters required to get it) suggest that if that is indeed the intention, its execution is a little off - and specifically only in 3v3. In 5v5, it's all good.
The reason for the thread is simply to suggest that CG look into tweaking the 3v3 algorithm, if all they did was "copy/paste" the 5v5 one into a different format.
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*
Depending how you access an individual character's detail screen within the GA you could be looking at either its current state or its locked-in state - https://forums.galaxy-of-heroes.starwars.ea.com/discussion/200268/viewing-roster-details-in-locked-down-ga-is-broken
I think they're trying to make it "fair" from the perspective of resources and general strength. It's not perfect. And it's a tough task. I still haven't seen a reason why a more complex algorithm wouldn't just make the complaints different a la TW.
See, I don't think so. Partly because that analogy falls apart if you're the guy or guys with DR/Malak who never even faced each other. That's also what arena is for - both of them. The rewards in GA aren't worth money. Crystals in the two arenas are. That's why I am offering the feedback - because I do think the design intent is for GA to be even.
I'm not saying it's easy. But it wouldn't be that difficult to take even just the data from the Discord script above and put acceptable ranges that are closer together than that.
Or a even a very rudimentary roster comparison. As a JKR owner, it wouldn't be fun for me to face someone who doesn't have him - but even more in 3v3 because having that 1 or 2 extra very powerful characters IS practically an extra team. And often the difference between a win and a loss.
Again, I'm not ranting or raging. It's just a suggestion to improve a game mode I really enjoy. The 5v5 version is a blast. Imho, the 3v3 just needs to be fine tuned a bit more.
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*
Agreed, especially since 3v3 breaks up some squads synergy enough that they go from competitive to just meh. That puts a lot more pressure on who has the best individual meta toons and mods since you can't count on better squad synergy to compensate.
I've seen this myself with the murder bears. In full 5v5 squads they have enough going on that they can hit decently above their weight class. 3v3 though they've been lackluster (for me at least).
I highly doubt this.
In a character only GA players are matched by character GP at roster lock - not total GP. I guess you're comparing total GP.
i don't know if GA championships are completely replacing regular GA, but if so that would solve the issue you're having. For me it will mean i'll start losing, hehe.
If player passed ladder level n-1 they get access to ladder level n. Requirement would be to get 3/3 victories on ladder lvl n-1 for example
For those getting hung up on the GP differences, where I got them from (Discord or in game), whether or not they included fleet (which they didn't) - you're kind of missing the entire point. GP being used (or at least, GP being given to much weight) is likely part of the weakness of the algorithm.
In fact, even if the GP were identical, you can obviously see by the above quote regarding high level mods and gear, that the match up algorithm is flawed if the goal of GA is to get even matches.
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*
Honestly, if they just compared the numbers I quoted from the Discord app (g12+ gear, 6 dot mods and 20+ speed mods) - I think opponents would be much more evenly matched. I can out-mod a slow 5* Malak. But if his whole DR team has 20+ speed mods, and I only have 10 on my whole roster, no chance.
That's part of the reason for this thread. The game is more about mods (especially 6 dot and 20+ speed) and g12+ gear than specific characters or how many 7 star rebels you have.
The match up algorithm should be more heavily weighted for those things that actually have influence on the outcome of the battles.
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*
Hiding a fleet behind a drevan/revan/cls territory
The point of your first post was, that the 8 players in a 3v3 GA should be within the same tight range of GP as in 5v5 GAs. I still don't believe there's any difference between 3v3 and 5v5 in that regard. I don't believe there's over 100k difference between players. However, feel free to post your evidence.
The goal of GA in its current design, is clearly not to create even matches with regards to mods and gear. Hence, there's no flaw in that regard. Since we all battle for the exact same rewards, creating even matches would be unfair and silly in several regards.
Why do you insist that it's not the goal of GA to create even matches? And at the same time insist that they must be even based on GP? Those two points are completely contradictory from a logical perspective.
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*
Did you ever experience an even matchup where all 8 players were even in all aspects and not just even in GP? Did you ever see an official announcement that players would be matched evenly in any other aspect than GP? No? Well, there you go.
That's a fact given by the official announcements back when GA was introduced. It's not something that "I insist".
Your logic is flawed.
You are matched with players with approximately the same total/character GP as you, but your rosters may vary greatly in other aspects like amount of high gear characters or high speed mods. What's so contradictory here? Do you even understand the meaning of that word?
First off, it is abundantly clear that you are more interested in "winning" a discussion rather than having one.
Second, if GP is even, "but your rosters may vary greatly in other aspects like amount of high gear characters or high speed mods." (per your quote) than what exactly is the purpose of GP? Apparently it measures nothing that actually has any real impact on the outcomes of the battles. Anyone that has played the game for a while realizes that stars and number of characters is not nearly as important as those with 6 dot mods or g12+ gear. Even CG has stated that g12 gear gives a significant advantage.
That's the entire point of this feedback. Which is to say, that IF GP is the sole factor in determining match ups, but GP doesn't accurately reflect roster strength, then the algorithm needs to be tweaked - especially in 3v3 where such large differences in "actual" roster strength (mods and g12+ gear - which 4 and 5 also increase soeed) are difficult to overcome.
I'm honestly not understanding why this is so confusing.
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*
I might steal this idea