Matchmaking - comment needed

slickdealer · July 11, 2019 2:52AM

How is having 100 fewer G12 and 20 fewer zetas working as intended?

This is no reason why this couldn’t be fixed from the last GAC.

Matchmaking on top 80 alone is monumentally ridiculous.

Nothingface · July 11, 2019 2:55AM

Hope you got a big mod advantage lol

CCyrilS · July 11, 2019 2:56AM

You've earned more lifetime points, so could it be you're better somehow?

Kilker · July 11, 2019 2:57AM

Just 1 shot all your opponents teams and you’ll be ok. Sarcasm. Obviously this matchmaking is just as flawed as the previous version based on pure gp. How bout we compare all characters on a roster over a certain power threshold based on your division.

Nikoms565 · July 11, 2019 2:58AM

7 v. 1 on g13 would seem to suggest that "top 80 GP" doesn't weigh g13 very heavily...if at all.

CCyrilS · July 11, 2019 3:00AM

Really though, he has his opponent beat in nearly every GA stat... maybe it's not so far off

slickdealer · July 11, 2019 3:18AM

Kilker wrote: »

Just 1 shot all your opponents teams and you’ll be ok. Sarcasm. Obviously this matchmaking is just as flawed as the previous version based on pure gp. How bout we compare all characters on a roster over a certain power threshold based on your division.

They need to split the difference.

Rather than matchmaking on 170 or 80, matchmake on 125.

There is zero point to running GAC with this algorithm.

Nikoms565 · July 11, 2019 3:19AM

CCyrilS wrote: »

Really though, he has his opponent beat in nearly every GA stat... maybe it's not so far off

Those stats are from the last round. I'm not understanding how they are even slightly relevant to the current match-up - or why you think they would be factored in at all?

Nikoms565 · July 11, 2019 4:47AM

Let's be honest. CG is not going to come here (or even the mighty. Reddit) and say "The good news is that the algorithm is working, the bad news is that we realized it's based on bad numbers that don't really make the" close matches" we were aiming for. "

More than likely, we'll get some heavy spin like, "We'll be looking at the match-up data and might be tweaking it after this GAC has run its course."

So they'll fix it 5-6 weeks from now....which will be sometime in....oh Lord.

Kalano · July 11, 2019 4:53AM

slickdealer wrote: »

...
There is zero point to running GAC with this algorithm.

I agree on that. It doesn’t even need an algorithm. Just match players within the same division randomly. That’s the only way to get the global leaderboard fairer

Neo2551 · July 11, 2019 5:22AM

This has to change. I am against a pool of players with 400k-600k more GP than I.

Rath_Tarr · July 11, 2019 5:48AM

My opponent has over 800k more GP than me! Totally unfair!!!!!

He has no chance.

Saada · July 11, 2019 5:59AM

Based on top 50 characters we are close but we need 60 since I am in div 4. He ishas in div 3, has 700k more character gp as well as 3po, padme, traya etc

Now I wouldn't mind if I was facing this guy in week 3 or 4 but my first week is exactly like this with all 7 opponents. If I manage to beat 3.5mil to 4mil accounts with my 2.5mil account I deserve more then just getting a 5 mil account in the second week.

They just need to make sure the match making is based on the amount of characters needed for the higher division. So div 3 need 80, so make sure if your matching them with div 4, make sure the top 80 are close in go not the other way round.

YaeVizsla · July 11, 2019 5:59AM

Well, OP, considering you and your opponent are within 100k GP top 60 toon power, you clearly have something to swing around of relevance.

Neo2551 wrote: »

This has to change. I am against a pool of players with 400k-600k more GP than I.

We just had months of learning why pure GP is a poor indicator of the quality of a matchup. That they're half a million GP more than you means little; it's how that's distributed that matters.

CHFC22 · July 11, 2019 6:02AM

New rule: come and complain after the GA, not before. You can't possibly say with certainty whether it's unfair or not until you've actually fought them.

Waqui · July 11, 2019 8:11AM

slickdealer wrote: »

How is having 100 fewer G12 and 20 fewer zetas working as intended?

Almost same 'relevant GP'. Working as it was announced.

Saada · July 11, 2019 8:20AM

Waqui wrote: »

slickdealer wrote: »

How is having 100 fewer G12 and 20 fewer zetas working as intended?

Almost same 'relevant GP'. Working as it was announced.

Lol that's not relevant with 100 more g12. It also means that they have probably been around longer so could have got padme and c3po first time around as well as have traya since they've been doing heroic sith raid for longer

Vinniarth · July 11, 2019 8:59AM

Have the same issue.
G13: 1 mine vs. 9 his
Malaks: 0 mine vs. g13 his
G12: 80 mine vs. 100 his.
He even has Shaak Ti.
What the hell?

CamaroAMF · July 11, 2019 9:58AM

Nikoms565 wrote: »

7 v. 1 on g13 would seem to suggest that "top 80 GP" doesn't weigh g13 very heavily...if at all.

The difference between G12+5 and G13 is 210 GP. GP is a useless stat.

leef · July 11, 2019 10:01AM

Waqui wrote: »

slickdealer wrote: »

How is having 100 fewer G12 and 20 fewer zetas working as intended?

Almost same 'relevant GP'. Working as it was announced.

"relevant GP" is what exactly though? An assumption made by the DSR bot makers of what GP counts towards matchmaking?

DarjeloSalas · July 11, 2019 10:03AM

It’s been tested - double the number of toons you’re setting on defence. You and your 8 bracket opponents will have very similar figures for this.

TVF · July 11, 2019 10:49AM

Vinniarth wrote: »

Have the same issue.
G13: 1 mine vs. 9 his
Malaks: 0 mine vs. g13 his
G12: 80 mine vs. 100 his.
He even has Shaak Ti.
What the hell?

Everyone had the chance to do the marquee.

No_Try · July 11, 2019 10:52AM

leef wrote: »

Waqui wrote: »

slickdealer wrote: »

How is having 100 fewer G12 and 20 fewer zetas working as intended?

Almost same 'relevant GP'. Working as it was announced.

"relevant GP" is what exactly though? An assumption made by the DSR bot makers of what GP counts towards matchmaking?

Let us know if you manage to come across a sample that doesn't fit with that assumption. There's also this where you can try out different assumptions. The assumption comes from the ingame notice as you know. What it means is somewhat vague, yet almost all such samples fit with all the MM I've seen so far.

https://swgohevents.com/gp-compare

leef · July 11, 2019 11:24AM

No_Try wrote: »

leef wrote: »

Waqui wrote: »

slickdealer wrote: »

How is having 100 fewer G12 and 20 fewer zetas working as intended?

Almost same 'relevant GP'. Working as it was announced.

"relevant GP" is what exactly though? An assumption made by the DSR bot makers of what GP counts towards matchmaking?

Let us know if you manage to come across a sample that doesn't fit with that assumption. There's also this where you can try out different assumptions. The assumption comes from the ingame notice as you know. What it means is somewhat vague, yet almost all such samples fit with all the MM I've seen so far.

https://swgohevents.com/gp-compare

It seems unlikely that the top 80 toons of a player with less than 80 G12's and less g13, g12 pieces and zetas than his opponent has roughly the same GP as the top80 of his opponent.
I can't base any definite conclusions on the printscreen posted by the OP because his top80 could be 1.649.999GP while his opponent's is 1.650.001 and it would be displayed as 1.6 vs 1.7.

No_Try · July 11, 2019 11:27AM

leef wrote: »

No_Try wrote: »

leef wrote: »

Waqui wrote: »

slickdealer wrote: »

How is having 100 fewer G12 and 20 fewer zetas working as intended?

Almost same 'relevant GP'. Working as it was announced.

"relevant GP" is what exactly though? An assumption made by the DSR bot makers of what GP counts towards matchmaking?

Let us know if you manage to come across a sample that doesn't fit with that assumption. There's also this where you can try out different assumptions. The assumption comes from the ingame notice as you know. What it means is somewhat vague, yet almost all such samples fit with all the MM I've seen so far.

https://swgohevents.com/gp-compare

It seems unlikely that the top 80 toons of a player with less than 80 G12's and less g13, g12 pieces and zetas than his opponent has roughly the same GP as the top80 of his opponent.
I can't base any definite conclusions on the printscreen posted by the OP because his top80 could be 1.649.999GP while his opponent's is 1.650.001 and it would be displayed as 1.6 vs 1.7.

I guess you're right. Merely the amount of g12+ difference should inflate top80 on one side at least a bit.

Waqui · July 11, 2019 11:31AM

Saada wrote: »

Waqui wrote: »

slickdealer wrote: »

How is having 100 fewer G12 and 20 fewer zetas working as intended?

Almost same 'relevant GP'. Working as it was announced.

Lol that's not relevant with 100 more g12.

Yet it's working as announced.

leef · July 11, 2019 11:31AM

I find it hard to trust other people on this forum since most clearly have an agenda and are willing to sway the truth at least a little to make their argument more compelling.
Unfortunately i can't test it for myself because my matchups have all been pretty close since the new matchmaking.

slickdealer · July 11, 2019 11:31AM

DarjeloSalas wrote: »

It’s been tested - double the number of toons you’re setting on defence. You and your 8 bracket opponents will have very similar figures for this.

Yes this has been mentioned multiple times in the thread.

This completely ignores the flexibility advantage the higher GP player has with having every counter readily available.

This completely ignores the flexibility of utilizing multiple teams to take down a stronger defense.

Waqui · July 11, 2019 11:32AM

leef wrote: »

Waqui wrote: »

slickdealer wrote: »

How is having 100 fewer G12 and 20 fewer zetas working as intended?

Almost same 'relevant GP'. Working as it was announced.

"relevant GP" is what exactly though? An assumption made by the DSR bot makers of what GP counts towards matchmaking?

I believe it was the words used by CG in one of their official announcements.

TVF · July 11, 2019 11:54AM

leef wrote: »

I find it hard to trust other people on this forum since most clearly have an agenda and are willing to sway the truth at least a little to make their argument more compelling.
Unfortunately i can't test it for myself because my matchups have all been pretty close since the new matchmaking.

@leef what's your agenda with this post?

New player? Check out the new player index

Matchmaking - comment needed

Replies

Howdy, Stranger!

Quick Links