GAC matchmaking - painful, but not broken
I’m struggling with the continuous negative comments about how the new matchmaking is creating broken match-ups.
I think the biggest problem is with people’s understanding (or expectations) about what is really driving things. So here is my 2 pence to add to the discussion.
- Eventually everyone will “find their level”. This might be higher or lower than the division where you were first put.
- When you’ve found your level, expect to win about half of your battles. (If you don’t, you haven’t found your level yet!)
- In a game where GLs (and Executor) are hard to counter with non-meta teams, your level will probably look fairly similar to you in terms of GL count. BUT…Good players may compensate with off-meta counters, clever strategy or roster depth, and be able to punch up a bit
- Every time you get a new GL (or Executor) you have a chance to float up to a higher level. But remember everyone else is trying to do the same
- a whale who auto-deploys and doesn’t attack will lose every time, even if it isn’t much fun to fight them. A strong roster who only engages with GAC sporadically will move about a lot and be annoying when they drop down and then climb up again. We have to get used to this, but is this really that common?
- it looks like CG is keeping a limit on the size of Kyber 1 (top 15k players?) and resetting the boundaries to keep it like that. It’s nothing personal. Keep fighting hard and you’ll get to the level you deserve.
TLDR: we still have good players climbing up, and bad or uninterested players falling down. This will take a bit of time everyone to settle, so please don’t complain about it just yet
My last round had me facing 7 opponents with Executors (which I don’t have) and more GLs than me. But I know why I was facing them, I know why I lost all three matches, and I know what I’m working on to address things.
End of rant, thanks for listening :-)
I think the biggest problem is with people’s understanding (or expectations) about what is really driving things. So here is my 2 pence to add to the discussion.
- Eventually everyone will “find their level”. This might be higher or lower than the division where you were first put.
- When you’ve found your level, expect to win about half of your battles. (If you don’t, you haven’t found your level yet!)
- In a game where GLs (and Executor) are hard to counter with non-meta teams, your level will probably look fairly similar to you in terms of GL count. BUT…Good players may compensate with off-meta counters, clever strategy or roster depth, and be able to punch up a bit
- Every time you get a new GL (or Executor) you have a chance to float up to a higher level. But remember everyone else is trying to do the same
- a whale who auto-deploys and doesn’t attack will lose every time, even if it isn’t much fun to fight them. A strong roster who only engages with GAC sporadically will move about a lot and be annoying when they drop down and then climb up again. We have to get used to this, but is this really that common?
- it looks like CG is keeping a limit on the size of Kyber 1 (top 15k players?) and resetting the boundaries to keep it like that. It’s nothing personal. Keep fighting hard and you’ll get to the level you deserve.
TLDR: we still have good players climbing up, and bad or uninterested players falling down. This will take a bit of time everyone to settle, so please don’t complain about it just yet
My last round had me facing 7 opponents with Executors (which I don’t have) and more GLs than me. But I know why I was facing them, I know why I lost all three matches, and I know what I’m working on to address things.
End of rant, thanks for listening :-)
9
- 60.9K General Discussion
- 9 Developer Q&A Archive
- 2.6K New Players
- 30.5K Arena and Character Strategy
- 4K Individual Events
- 4.6K Guild Events
- 4.6K PVP and Rivalry
- 50.1K Guild Recruitment
- 731 Art and Media
- 4.1K Off Topic
Howdy, Stranger!
Replies
Then having to squish so soon after just shows how poorly thought out it was.
No, no, no. Not everyone is understanding this, far from it. There’s a million of painful post about this out there proving my point.
This is a lazy, cliched take on matchmaking that is also wrong.
In the old system there was no guarantee that the best performers would ever be drawn against one another, and in many cases the top of Kyber could be secured by someone who got 12 consecutive mismatches in their favour.
In the new system the people at the top of kyber are guaranteed to face each other as the GAC progresses, so the ones who end up at the top have genuinely beaten all the best players and deserve their title.
If you like, I can post you a dozen ally codes of players who spend a fortune at the game but suck at GAC?
People that make lazy, cliched posts don't want actual facts or data.
I recognize my circumstances are unique to me but so far these GAC changes and MM have been really good.
Won one bad matchup because the opponent didn’t care (autodeploy set four GLs in the top row, that was scary, but their attacks couldn’t beat my g11 defenses). Lost the second. Lost the fantastic matchup, but it was such a fun loss because it was against someone with a roster built similarly as mine and defenses placed almost exactly as mine! They earned their win because they had a better Executor, but it was refreshing to play against someone with a similar strategy both in terms of defense and roster construction.
tl;dr—it’s a rocky rollout. But it’s also self-correcting.
- You had a go and were successful (most whiners resign themselves to defeat because of the output from a discord bot)
- When you did lose, you recognised something you could have done differently (most whiners insist that every defeat is the result of the unfair MM, and only their victories are the result of any skill)
Top of Kyber 1 equals good player but high GP does not equal best player.
This ^^
There are probably better players but you don't see them because the leaderboards are broken. The only reason Kyber1 worked last season is because everyone started with the same score and there isn't a higher division full of low-skilled, high GP players for the "leaders" to be matched with.
CG needs duplicate the season1 Kyber1 leaderboard success in the other divisions/leagues. To fix the leaderboards by using W-L record instead of Skill Rating (SR should still be used for matchmaking and rewards)
There are likely players from Kyber2 last season that are undefeated, but unable to rise to the top of the rankings because they started with a lower SR.
Just... no. Maybe we don't agree on what makes someone the "best". I highly doubt there's a scrappy 5M player out there that could be handed the roster and mods of the guys at the top and compete at the top. There is SO MUCH to learn. From watching Ahnald's and Fatal's streams the last couple months, there is so much difference to the offense/defense meta, mod meta, counter meta than most have even thought about let alone mastered.
I actually watched a bit of an Ahnald stream recently. It was quite entertaining to see him make a rookie mistake against an Enfys Nest.
Again, I am not saying that the top players in Kyber 1 are not good players - they would have to be in order to beat their peers - just that they are not necessarily the best players. There are a lot of good players scattered throughout the leagues who are now buried in the rankings and are unlikely ever to be matched against the highest GP players.
This put me into Kyber 1. The adjustment put me down to K2, esp after facing opponents with 5 GLs (I beat one of them). (Yes, I understand IRT and logits for the 50:50 win:loss rate.)
I'm now facing a group that is much more similar to my rosta and I feel happier about it, all for a cost of 20 crystals per day.
I don't, nor did I, dispute that.
For sure. He's human like the rest of us. He makes mistakes. There are also a few things he does frequently that I think are sub-optimal. But if others watched you play the game, even if you weren't trying to entertain a stream, I guarantee you, they'd find mistakes. Heck, record yourself and rewatch it. I guarantee you'll find things you could do better everytime.
I disagree. I think they are the best players. Sure, maybe there's a smart guy with a smaller roster that could do more than them with the same tools. But he doesn't have those tools, so he hasn't done so, so I don't care.
There may be someone out there that could run a new WR in the mile if only he had been identified early, trained his entire life for the event, and had everything come together for a perfect race. But that's not how we measure accomplishments.
I knew someone would recognize my talent on the track, even if only on a Star Wars forum.
Agree the top of kyber1 is good player + good bank account/credit card.
A C1 player might not have to bother with all the mechanics of the latest toons (because he simply does not have/face them yet), but he still need to worry about the zone placement and counter management as much as a K1 does. Sometimes you could say even more because he does not have so many resources at his disposal to do multi-shots. It’s often one-shot or lose.
Conclusion: try to say there might be a “best player” in lower leagues is nonsense. It is equally nonsense to say the best player is somewhere in the top50 of K1. Every league is a different game. There are some skill factors you can benchmark but some definitely not.
One ship / character shouldn't be deciding as many matches as I feel like executor is deciding. That's from other players matches I'm watching/reviewing, not just my own. I bet the data is going to show that as well. (Not to mention Razorcrest being the "weak zombie" of fleets due to speed differential and speed order)
There is some RNG involved, but there are strategies with which you can beat 7* Executors MOST of the time. If you're not winning the Exec counter battle like 75% of the time, then that's on you. Also, if people really put garbage defenses outside of Exec, then you can just undersize all the other battles and can sometimes even afford to 2-shot the Exec and still win the match. Sure, Executors do decide many matches, but it's more a reason of people not using the counter properly rather than bad RNG. And if you're really afraid of bad RNG then just keep your Exec for offense and just oneshot theirs, while also benefitting from their weak ground defense. It's pretty straightforward honestly.
75 percent? Puh-lease. I haven't seen a single guild in TW (395M guild) get over 50% on an Executor wall. Most land in the 30-40% range.
https://swgoh.gg/p/396682921/
Our guild (400+) has done 60% a couple of times, otherwise mostly 50%ish.
75% is either a low GP player against G12 ships or a huge exaggeration.
Yes this is me. Thanks for proving my point though. You must have misread the GAC history. The most recent week I literally went 2/2 with the Exec counter:
And this is just GAC. I'm easily above 75% in TW and probably around 75% in Fleet Arena, mostly against maxed out fleets. And this is strictly with Mal. When I have leftover battles I sometimes practice the Mal counter in the arena as well.
It's obviously difficult to get 75% as a whole guild, because it's 50 people and not everyone is equally good or has the same rng. Just because 400 mil guilds can't get a 75% average, doesn't mean individual people can't have 75+%. Even on my alt (that has mediocre Geos at best) it's still like 50%. If you have a good Mal fleet and 100% know what you're doing, then you should at the very least be hitting 65%. If you don't then you're doing something wrong.
You do realize that g12 ships is an impossible thing to go against outside of RC and XB? And most that have a 7* Exec have JMK anyway so Cad Bane is also r5 at the very least. So besides the occasional g12 BAM, all the ships you go up against are decently relic'd no matter what.
These are not good Executor fleets. I can even tell non-BH were reinforced.
I checked your GAC back quite a ways. You have nothing to offer anyone about Mal vs Executor based on your GAC history. Maybe you do really well in TW, I dunno. But you have some proving up to do.
Do you realize that sometimes reinforcements like Scimitar actually make it harder for the Mal counter? I've seen some high end players and guilds use it on defense in TW and GAC as well. It doesn't ALWAYS work out for them of course, since the AI can derp in many ways, but it's a valid counter to the Mal counter. Even Lando's falcon is extremely dangerous if you don't finish the battle quick enough, as he can just wipe out your entire team (or what's left of it). Pure BH lineups on defense might be a bit more consistent, but certain other ships can mess the Mal counter up massively since they're unpredicatable. For Exec MIRRORS, a pure BH lineup usually does much better on defense, but for non-Exec fleets "random" ships can really mess up the battle sometimes.