Guild Wars 2

Yes, a necro seems more “team-dependent” than most.

I am a pretty good terror necro player, as long as I am on a team. But boy, when you are placed with new players as teammates who are running everywhere alone, you just get destroyed trying to move anywhere on the map alone.

Just sayin’

Burn.5401:

Party Size

Since playing with your friends gives you an advantage, we’ll try to keep things fair by matching you with other parties of the same size. Playing with a full team? We’ll prefer matches with other full teams. Playing alone? We’ll prefer to match you with other lone wolves. Even if we can’t find another party of the same size within your skill level, we’ll work to get it as close as possible..

This was taken out of one of the site news regarding the new system. Why am I not seeing it in-game?
It’s the same as if I queued alone for the team ladder during the “previous” system.

The key word in there was try.

If a whole bunch of high-skilled players are in the Q in groups that cannot be split up, the only possible games at the moment will be blowouts.

Just a thought, what if the matcher scoring also looked closely at four players in a potential team: two weakest MMRs on each team?

This would be in addition to the team average MMR. But if the two weakest players on one team are really weak, then reduce the score for that potential matchup a lot.

Xista.7391:

convinced mmr is broken

I have questions about it, too.

MMR is derived from your team’s expected performance in matches that are not necessarily as tough or as easy for you as the matcher predicts—- because the matches are created by possibly inaccurate MMRs!

Circular errors perhaps.

Kharr.5746:

I’ve been playing a whole lot of unranked arena solo, and I have to say that the matches are way better. I normally play around server reset times, and the queue time is 20-30 seconds most times, while the matches never feel like a landslide.

If the problem with matchmaking is during off-peak hours, is it possible to dynamically adjust matchmaking based on server population? Or, maybe the player base from ranked and unranked could be merged, but player rank would only be affected from a win/loss if you entered the match via ranked queue.

I suspect they are looking at doing just that.

Are you in Ranked Q?

I suspect that with the low number of people queuing in your “bracket” there are not enough for good comps as well as a good, competitive match.

Sounds like not enough top players in ranked for a good game with good compositions.

Are you still playing unranked, Lopez?

butch.8136:

Well, the ladder will reward you for trying and performing even against stacked teams.

I just hope it would tell you how much of an underdog your team is before the match starts. Tell you how many points your team needs to score to get +1 in your ladder rank.

Then you have a realistic goal. I would like that.

Butch, I tried ranked and saw it was at 4 minutes average. I didn’t stay in Q, cause I figured I am not skilled enough yet for the new ranked Q.

Acandis.3250:

Laserbolt.6731:

Under this new system, these 3 players in top 1% realize…. I could either take my chances with the MM algos, or i could just hit up the other 2 homies and farm the ez wins. This is exactly what is happening now, and will continue to happen in a system where teams and solo players get dumped in the same low population bucket.

In Ranked Q if you do this now, you just won’t get a match if there are no other teams of your skill to play against.

That’s what the long Q times are about today.

They’ll get a pop… after 30 minutes. So instead of waiting forever, they’re just gonna go do unranked and curb stomp nabs there. Back to issue in OP, this is boring for both teams.

I think most of the mature top tier players have more empathy for the new/average player than to do that mindless, boring stomping any longer.

Never had a bad match today yet. Most all were close and could have gone either way till the end, and turned on a few key decisions in the last few minutes.

I am playing in Unranked, at an average MMR.

I suspect the top tier players got tired of stomping 500-0 in Unranked and are playing Ranked Q for a challenge at the moment. Or they are not playing at all right now.

Under this new system, these 3 players in top 1% realize…. I could either take my chances with the MM algos, or i could just hit up the other 2 homies and farm the ez wins. This is exactly what is happening now, and will continue to happen in a system where teams and solo players get dumped in the same low population bucket.

In Ranked Q if you do this now, you just won’t get a match if there are no other teams of your skill to play against.

That’s what the long Q times are about today.

“You beat some good players Cruuk and Raplion, just goes to show how much of an impact having TS/Mumble/Ventrillo has on the game.”

+100

“Like I said before, if you guys didn’t party together, you’d have more good matches. But since all the good players play together, you don’t have any good matches, or hardly any good matches.”

Yup. Well said. Good insight.

But wouldn’t you say overall more players are having fun today, than they were when they were being crushed 500-0 by a few teams of high skilled players that they were fed to?

I think so. And that is great for the future of the game.

I believe Ranked Arena refuses to put you with lower-skilled players now.

So apparently there are not enough skilled players in Ranked right now to find a real match for you.

As an “average” skilled player, I have had much better matching today.

Whatever changed in the matcher, it’s a good thing.

More close games. I like that. Haven’t faced any teams of top skilled players blowing us away 500 to 50, etc.

I think the matcher is refusing to put average players up against top-skilled players in the ranked queue.

The high-level players have to wait longer because there are not enough of them to play each other, and throwing newbies against them is bad for both sides. No challenge.

Justin,

Has it been decided that after X minutes a very low-MMR-average teams will simply be put into a match with a very high MMR team if that’s the only set of 10 players available?

If so, is that better than no match at all? Or is it actually worse, since players just stop playing. Or they AFK until the match is over. So what has really been accomplished?

People who cannot get an arena can still do Practice. They can join a custom arena. They can duel.

Skady.5916:

The worst thing about this thread is that developer genuinely asking how we get to know if we facing premade or not.

That means that they are completely out of touch with their own game. The option to know if your teammates/enemies are in the group or not was there since release and was widely known and used by pvp community.

Literally if you follow any tPvP stream for at least 10 minutes you will get to know that.

The game was out for 2 years… I’m facepalming…

I think the developer showed a record that proved that the perception was wrong that the other team was a premade. He wanted to know what made the complaining person think that it was a premade.

I think it’s more that the higher skill people who have been winning are all playing and they group up with other high-skilled players they are used to teaming with.

They are the ones winning all the games. For some reason the matching system is putting less-skilled people up against them.

My experience as a pure soloer last couple of days in unranked. Not enjoyable. But I did it to test. Can’t take it much longer though.

You know, putting people up against much more skilled players is NO DIFFERENT than a gear-based MMO when you are up against people with better gear.

It feels exactly the same when you get “spawn camped” from higher skill and knowledge players as it does from higher geared players with stronger weapons.

And since GW2 prides itself on not being gear-based…it should never be done.

Well, what good is a 500-0 game that is absolutely effortless, and turns off 5 other players maybe for good, just to shorten the wait?

Sometimes there is just nobody at your level to play with, as I see it.

Having people that are high skilled wait a very long time is about as bad as feeding 100s of average and new players as cannon fodder to the veterans. I think 20-25 minutes is about as often as I would feed them an unchallenging blowout that demoralizes the folks who are learning and never had a chance.

Every 20-25 minutes is long enough to let the average player have some fun before the demoralizing stomp. It reduces the stomping and keeps the majority happy.

I’ve been as high as 200 or so. But it easily dropped. I never Q’d with anyone I knew.

But if the matcher has mostly mid to low rated people, THEY should get the matches and never be fed to high MMR. What’s the point of that? We’ve been seeing it all day and it isn’t pretty. It lowers the player base.

I just got two matches of competent players who knew what they were doing. We also won, and the 2 were both close games that went back and forth.

The Q was still under a minute.

So is it worth it to allow 500-0 blowouts so that the top people can play?

Isn’t it better to let people play at their own levels? If there are more average players queuing, why feed them to top level people? Let them majority have a reasonable time. Majority rule. Better PvP scene. More encouraging.

Firebird.8324:

here’s another http://i.imgur.com/aUZgC2x.jpg 500-0

Let’s think about what would cause the matcher to let this match happen?

Shouldn’t it have rejected it?

Shouldn’t it make Firebird’s team wait for a ~50% chance match?

Or is it perfectly OK with 99.99% win chances. I.E., blowouts?

Queues are still VERY fast 10:25 CST

Lordrosicky that makes sense.

Justin,
Q times are instant ATM.

Clearly there are many good players playing, but now many of the solo and average players may be staying away until they have a chance for a reasonable fun game in their bracket.

I just wonder why the high MMRs aren’t just waiting longer in queue and then play against their own level.

I imagine my MMR is plummeting in unranked from all the losses. This happened on the LB a year ago. Once you were low, you got worse teammates, as I recall.

Avg Q time = 22 secs. Not many choices for the matcher would do that. Also explains my 4×5 just now.

I suspect the number queuing is LOW. Looking at the Mists tonight it look like about 1/2 of what it was last night.

Just had a 4×5. We lost with with our four, 500 to 429. Should of won if we had 5.

My theory is that just by BEING IN A GROUP of any size means you are a more cooperative, social player who is more likely to work with others on a team strategy. I suspect too many solo players tend to be freelance roamers without a team plan. Just sayin’.

If many solo players have stopped queuing, then there are mainly groups queuing… and solo players “fill-in” as opposing teams for them.

Can you try setting it so that teams with at least one group must face teams with at least one group?

Thanks, Justin. First match seemed better.

I also suspect they are hoping that the team vs solo problem will urge people to group up with other solo players who want to work together but have been a little “shy” about asking other people to group.

Lordrosicky.5813:

I would like to advocate rolling back the matchmaking changes. It really was fine beforehand and the only people who moaned about “3 thieves on my team” were idiots. The current system is just making the game unplayable.

I respectfully have to disagree. The system has been rewritten and a lot of great things are possible with this new, more flexible system.

They can re-load in everyone’s established MMR from the old system (and I believe they are planning on that right away).

They can adjust the “dials” to mimic the old “match selector” if it comes to that.

I think the allowing teams mixed with solo players, and resetting of MMRs are the main reasons for the current pain.

Just knowing that your team was officially outgunned would greatly lessen the “sting” of a loss.

It would be reassuring when it says: “Your team did not lose by as many points as predicted. Your ladder rank has been increased by 1"

Terrorsquad.2349:

If you want my honesty, since the patch December 2, I’ve won all except one match. (was vs CM; 4x cele eles)
So yea :p

Do you only Q solo since Dec 2? Or are you grouped with other good players?

I look at my Unranked by Profession stats: Played 22 won 7.

And I remember the last match where 3 of my team fought at the mine in Niflhel the entire match and we lost 500 to 49. It’s interesting to see the stats, but I have to remind myself that they do not necessarily mean that I played badly.

GhOst.4019:

Evan Lesh.3295:

This is definitely something we’re interested in doing with the new system. I’m imagining something liked dotabuff where you can see each match, score, duration, profession makeup, builds, etc.

Not sure if I am reading all this wrong but I think you are talking about 2 completely different things. I believe Evan is talking about adding in something to game that allows us to look back on recent matches and see the scores, personal stats…etc.. (Which would be EPIC).

I believe OP wants the total data from all matches (Including excessive data) released to the public so they can help fix the matchmaking system. So a 500-0 game…what resulted in that being a" legitimate" matchup and how can “what went wrong” be fixed.

Maybe I am all backwards on this whole post lol.

GhOst, you are right.

I was talking about using more eyes (and maybe data analysis in spreadsheets) to see patterns and reveal odd anomalies that might help improve the matchmaking settings.

It may be getting worse if only the winning players are queuing now, and the rest of the persistent soloers are cannon fodder.