Guild Wars 2

Meri.7512:

And before it starts: There are still bugs..
I q’ed to see the time and left the q. Started to play at the silverwastes and got a q-pop…

Hmm. Your leave queue message must have got lost. I’ll check on it tomorrow.

Looking at your game history, you have 3 desertions, which count as full loses. They have cost you a total of -4 points over the last 2 hours.

These desertions are not yet shown in the losses column, but that will most likely change in the near future.

Also, the website takes snapshots every hour, so it may be some time before they update.

ArrDee.2573:

Why is there such a large emphasis on effort! Why is it that someone who has 3000 games played with a 40% win rate can be higher on the leaderboards than someone who has 1000 games played and an 80% win rate! It doesn’t make any sense at all! It defeats the whole concept of a leaderboard!

Wahoo! Bye frands!

Well budzy, probably because the 1000 games weren’t during the ladder period?

You can’t compare account lifetime games played to ladder standing and assume it’ll match up.

Personally, I think the perception of an over-emphasis on effort is caused by the false start. I could be wrong though, so I’m looking forward to seeing how this plays out.

:)

Laserbolt.6731:

“don’t put me against that player because I’m a 2 star playing and shouldn’t be facing any 5 star players”

Well, yeah, but strict matchmaking does that already. Right?

If the star rating implied skill, matchmaking should already be doing that through MMR. Could probably correlate star rating to MMR, either validating it or pointing out issues. Could also be matched with professions to see which need looking at.

Like I said, would be a very interesting metric for us.

uberkingkong.8041:

I like shorter que’s because you cant really do anything in hotm anyways.
There should be a playerbase rating system added tho. Like if the entire losing team knows they lost because of that one or two players on the enemy team, they should be able to flag them or give them stars like 5 stars, indicating don’t put me against that player because I’m a 2 star playing and shouldn’t be facing any 5 star players.

This might be a very interesting metric for us (ArenaNet) to have. Maybe correlate perceptions with other pieces of data. I can’t imagine ever sharing that information in-game.

Something like “Community Awards” at the end of a season might be interesting though, assuming it wasn’t being gamed.

I’d rather see it be a generic vote, something like “How would you rate the other players on your team?”. We could still tease individual ratings out of that, without worrying about individual bias against particular players.

Skady.5916:

Current ladder is an utter joke (no offense meant)

Most of top players are just pvp bots with <50 games won and over 9000 games grinded.

Please don’t belittle other players like this.

Skady.5916:

Ladder should reflect skill (winning %, personal scores etc.) over time spent in pvp.

The goal of this scoring method is put emphasis on both skill and effort. Take Rank 1 in both EU and NA. EU has virtually the same score as NA, but had NA to play more than 150 games to achieve the same score. Both represent things that deserve credit: perseverance, consistency, and skill.

We know some players would like skill to be weighted higher than effort, all I can say as that we will review the data and be looking at feedback throughout the season, and especially at the end.

Also, do not forget that the current ladder data isn’t a good test of the system. We had matchmaking errors, configuration changes, etc, not to mention the ladder had a false start where data was being collected before people knew it would be visible.

mistsim.2748:

hey Justin. what makes the map not competitive is that spawn camping is possible and rampant! have you thought about adding multiple spawn points to the map?

it’s a fun map if youre with a team. consider adding more DM maps.

I joked with Hugh that we should put a big burning oil siege at the main exit.

We’re looking into serious ways to improve things, but it isn’t as high of a priority as the new game mode.

Exedore.6320:

In regards to map selection, I know the map selection on ranked is restricted to Temple, Khylo, Nifelhel, and Foefire. But can unranked play still get Courtyard, Skyhammer, and Spirit Watch?

They will remain in unranked. Sorry, I’ll go edit that other thread to include that information.

As some of you may have found by now, we’re shifting the rules for Courtyard a bit in hopes it improves the playing experience.

We’ve changed respawn time to 5 seconds, and the total time limit to 8 minutes.

Please let me know if that makes the map more competitive.

Also, as a reminder, this map, along with Skyhammer and Spirit Watch, will not be available during the beta ladder in ranked arena. These maps will still be available in unranked arena.

Archaon.9524:

If it’s gonna be like this noone would actually care about it anyway

Hater’s gonna hate.

In all seriousness though, we’ll be looking for feedback after this iteration.

The Baws.5164:

The current MMR system is pretty ridiculous. I just got stomped in a game from a big lead due to playing against CM Collero in a 3 man party…

I’m rank 22.

You just came off a 7 game winning streak, in the middle of the night in EU. There is only so much I can do. :/

The back-end servers have been updated in preparation for tomorrow’s build. This means we get all the matchmaking fixes early!

I’ll update the wiki page with the details shortly, but the tl;dr of it is this.

We are now targeting a 2 minute queue time average, with an 8 minute maximum. Queue times will vary around those numbers based on population, but please let me know if you are stuck in the queue for longer than 8 minutes.

Please let me know if you notice any changes in quality (good or bad).

Just as a reminder, the beta ladder starts tomorrow shortly after the new build goes live. Before the ladder starts we will disable progression and reset all ladder data except MMR.

We will not start the beta ladder immediately after the build, but soon after when we are sure everything is working correctly.

We will give a one hour notice before the new ladder will begin collecting data.

Good luck!

Courtyard, Spirit Watch, and Skyhammer will not be included in ranked arena once the beta ladder officially starts.

Wispy.8679:

I’ve been noticing since the last patch that disconnecting mid match will also not count as a win to your record and you also get no post match rewards, and you’re saying that it will now also give dishonour?

I’m not certain the no post-game rewards is accurate. I believe you still get those.

Wispy.8679:

I’m sorry but that is incredibly harsh. I played a courtyard game a couple days ago, dc’d mid match but reconnected. Team still won and I still scored personal points. It got counted as a loss, and I recieved no post-game rewards and now I will also get dishonoured for something completely out of my control? Was the first and only dc that day, had no way of knowing it would happen so it’s not like I just decide to not queue when internet is feeling sluggish.

Dishonor works on an exponential curve. The first time will have a small timeout. It’s repeat offenders who will be hit the hardest. Even a small disconnect can have a negative impact on the other players in the game.

Wispy.8679:

This isn’t even taking into account the dishonour you get for simply not accepting the queue pop (Which is utterly ridiculous and should never be implemented).

We felt a minor amount of dishonor (and queue dodging has the smallest of all types) was a good way to discourage trolling by constantly declining match pops.

Wispy.8679:

Due to the new system you’re already forced to sit in HoM, and now you can’t even afk for the 10-15minutes the queue may take to pop???

Aside from a bug that will be fixed on the Tuesday 16th, you shouldn’t be getting 10-15 minute pops. Let me know if you do get long queues after Tuesday.

Wispy.8679:

harsh dishonour systems will make people stop pvping.

We have different standards of what is harsh, as does nearly everyone on the forum. We can only try to implement systems that works for as many people as possible.

Here are a couple of links that include more details on how dishonor is implemented.
http://wiki.guildwars2.com/wiki/PvP_Matchmaking_Algorithm#Dishonor
https://www.guildwars2.com/en/news/go-with-the-flow/

uberkingkong.8041:

But the community must like seeing less information so it got removed.

We removed seeing other player’s rank to prevent harassment and the shaming of new players.

We don’t show MMR for the same reason.

Sevastianos.5672:

Our bunkers normally don’t enter the match as their bunker character, and also experience disconnect. We normally take care to make sure if someone disconnects they atleast earn points before the match ends but could there be abandonment penalties for swapping or DCing?

Yes, disconnecting or changing professions will count as desertion, and give that player dishonor. They will also be scored on the ladder as if they had a full loss, meaning as if their team had scored 0 points.

Sevastianos.5672:

These people made huge contributions to the outcome of the game but are actually losing points on the leaderboard even when we win.

They shouldn’t be loosing points for wins (unless they desert). We base the amount of ladder points awarded on team score, not personal score.

The rest of your post seems premised off the same idea, that the ladder is based on personal score instead of team score. Am I wrong? Or, are you suggesting we should base it on personal score?

If you want, I can review your account to see if there is anything odd going on.

Lotus.1682:

So I’m guessing the probabilities for winning are decided before the match itself begins? In that case, I have lost a few 4v5 matches. If our original probability of winning was high, it’d make sense that I lost so many points.

Yes. It is determined by starting values to prevent exploits.

Lotus.1682:

I know that leaderboards won’t be totally operational until the 16th so I honestly don’t care, but I was wondering in case something like this happens in the future. I’ve been constantly playing games, but my total number of points seem to have decreased. When I check at one point, I was at 7 points and then a few matches later I was at 5. I was under the impression that you cannot lose points? Is this a bug?

No, you can lose points.

There’s a graph on this page that will show you when you will gain or lose points.
https://www.guildwars2.com/en/news/finding-the-perfect-match/

There are more details here:
http://wiki.guildwars2.com/wiki/PvP_Matchmaking_Algorithm#Ladder

lilwizz.5783:

There is this bug where after the arena ends and I leave the map, the loading for Heart of the Mists is there but then there is another loading right after it that takes me back to the map I was in for the arena. It happened multiple times. I am unaware if someone already said something about this but it happened quite a few times.

This happens when you left before the match finished (in the background, not actual game.)

It’s annoying, but harmless. I’ll double check we have a bug for it.

How would we determine your league? By that, I mean, do we place you in a league based on the MMR at the start of the season? 10 days after the season starts? Or, do we let you move between leagues as your MMR goes up and down throughout the season? Do we let people pick their own league?

What sort of implications does this have for matchmaking? Do we only let people fight others in their league? If yes, this could cause long queue times, especially in the upper leagues. If we let people fight players from any league does the current odds of victory point handicapping still fit?

How does this affect rewards? If we rewarded the top 150 players, do we now only reward the top 50 in each league? If we split rewards by league, doesn’t that imply locking people into a league at the start?

I like this discussion, but keep in mind, I’m not a designer and be kind to me if I don’t know the answer to these sorts of questions.

Relentliss.2170:

Justin ODell.9517:

Silentshoes.1805:

Is it a coincidence that the PvP designers have relaxed the matchmaking, favoring team formation and strongly hurting the solo experience?

Yes. We do not intend to ruin the experience for people that queue solo. Matchmaking was relaxed overnight because I would not be available to monitor the servers to prevent queue times from growing too large.

This is NOT something any of your customers asked you to do. Just your small team forcing a particular playstyle on your players.

I’m confused. What isn’t something our customers asked us to do?

^__________________^

You just helped me find the bug!

Edit: Sorry, that was a bit unprofessional. I’ve just been trying to track it down for days.

Silentshoes.1805:

Is it a coincidence that the PvP designers have relaxed the matchmaking, favoring team formation and strongly hurting the solo experience?

Yes. We do not intend to ruin the experience for people that queue solo. Matchmaking was relaxed overnight because I would not be available to monitor the servers to prevent queue times from growing too large.

Belarorn.9062:

What the Heckz did you do?

I told it to match people as soon as it can.

Belarorn.9062:

When you gonna fix this Sir?

It get’s fixed permanently on the 16th, but I can ramp it up while I’m at work and this weekend when I have the time to monitor it.

Silentshoes.1805:

So, if you are given a match with essentially no chance of winning, you are docked a point for not getting at least 200 points against the uber team.

I’d do it differently.

So would we, sort of, which is why Courtyard won’t be in rotation once the season starts.

crunchyraisin.6054:

So… as much fun as it is to qq, complain and restate problems…I’m gonna keep trying to think of solutions and post those instead.

Please do! I’ve committed to not making decisions (for myself, it’s not my call, we’re a team and we also have designers get paid to think about this stuff, etc.) until I see the data after the season is over.

That said, been wracking my brain trying to think of potential solutions to the concerns brought up. Maybe even something as simple as winning always awards 1 bonus point might help.

Silentshoes.1805:

Both Ranked and Unranked are about 15 to 20 secs in Q.

Is there any selective matchmaking going on at all? I’ve had nothing but blowouts for two hours.

It’s filtering by MMR only, and it’s being generous about that.

HipposWild.7185:

These matches are so bad!!!! I don’t understand what happened but someone did a terrible god awful job implementing this. Every other match has a 300 point landslide with 5 pre-made dragons versus 5 solo rabbits…. This is the dumbest most enraging state of sPVP I’ve seen since launch. It’s borderline unbelievable.

That’d be me. Sorry. I basically gave the system a lobotomy until we can get in a fix to let it think smarters.

Mrbig.8019:

Without brackets it’s impossible for high rated teams ( or individuals) to get even matches, they’ll get stuck into an ELO hell they won’t be able to escape from unless playing 5x times more then X “random Joe” who’s nothing more than decent.

It’s more likely RandomJoe.5555 Will probably have to play more often to get the same points the higher rated teams will.

Mrbig.8019:

Facts to be told, if u win against a highly more rated team, you’ll get 3 points just like if you were facing a “slightly better” team.

Not sure what you’re saying here. You only get +3 points if you win when you’re expected to lose against by a wide margin. If you play a slightly better team, you’ll only get +2 points if you win.

Check our the table on this page:
https://www.guildwars2.com/en/news/finding-the-perfect-match/

The wiki will also have the details: http://wiki.guildwars2.com/wiki/PvP_Matchmaking_Algorithm#Ladder

Mrbig.8019:

We need a huge “point increase per win” and more “point loss” per loss.

Changing the scoring is certainly something that could be done. We’ll be looking at things like that, and for your opinions, at the end of the beta ladder season.

Mrbig.8019:

Moreover the best “point score” needs to be awarded ONLY ( and i mean ONLY) if you play against EQUAL or MORE RATED teams than yours ( or better team average)

That would suck for the more rated team, since you’re lower than they are they wouldn’t be getting points. All lose and no win doesn’t seem very fun, so I doubt they’d enjoy it. Could be wrong though.

Mrbig.8019:

No one cares about lamas.

:( I love my llama!

P.S. MMR != Ladder Points, we use Glicko2 for MMR.
http://en.wikipedia.org/wiki/Glicko_rating_system

Silentshoes.1805:

Had a couple of Courtyards where my team was crushed 500-0.

Say we had 1% chance of winning. Do we still take a point loss if we lose 500-0?

0 points always awards at least -1 points. It’s more if you’re expected to win.

The wiki page has the exact numbers:
http://wiki.guildwars2.com/wiki/PvP_Matchmaking_Algorithm#Ladder

My blog post has an easier to read table:
https://www.guildwars2.com/en/news/finding-the-perfect-match/

longbawl.5072:

Meaning what for solo q? Won’t we will always be kitten because of premades? Sounds like solo q is just disappearing then.

Meaning nothing for queuing solo. You can play the way you wish, we won’t stop you. I just meant we want people to form teams and play together.

Silentshoes.1805:

Are the top players farming now that they believe it’s the number of games you play, not your MMR?

They’re partially right. MMR is factored in and skill does matter.

Look at Rank 5 vs Rank 1 NA… Rank 1 has to play ~1.7 matches for every game Rank 5 plays just to keep their lead.

Also, when the ladder starts, the matchmaking system will be able to put more weight on ladder position, so they could end fighting each other more often.

Silentshoes.1805:

Maybe the lesson is “Do NOT play Ranked Q without a team and TS”.

It certainly helps, and it’s something we encourage.

Lopez.7369:

I honestly can’t tell a difference in the matchmaking compared to the previous patch — except now I notice that I fight teams of two, three, or four sometimes. Although that’s actually a good thing for me, since it means I can now queue with one or two people.

Yeah, I put min-potentials down all the way because people were getting long queues. I’ll ramp it back up once the next patch hits and it can scale down by itself.

So the good news is that you’ve got super fast queue, the bad news is that its with the first people we found.

Style.1092:

No comments on a thing that scoring doesn’t work properly ?

¹ Winning will always award a minimum of one point.

http://imgur.com/42a4imT

The leaderboard site was updated sooner than we planned. Since we didn’t think it was going to be visible, we were making changes to the ladder data after the 2nd which caused some players to have data that doesn’t line up properly. This won’t be an issue after the 16th when the beta ladder officially starts and all the data is reset.

Just to be safe though, I just took another look at the code and verified that it will give you at least 1 point for a win. I’ll also ask QA to run another pass on this as well.

Thanks. I already know about this one and I’m busy trying to track it down.

I’ve fixed half the problem already; with the next build you won’t be able to accept if you’re already in a match.

Now I just need to stop it from happening in the first place. -_-

It attempts to take a snapshot every hour. Sometimes it’s busy and doesn’t get one done until a bit later. It’s also not exactly an hour so they stagger a bit and don’t all try to do it at once.

Silentshoes.1805:

As has been mentioned, even having a group of 2 on a team, potentially in realtime communication on TeamSpeak, gives a huge coordination advantage,

We know this is the case.

Looking at the data, and running simulations where we take roster size into consideration during the filter part of matchmaking. Couple of ideas sound good to me at first glance. 1) Adjust their MMR upward so that if they do get paired against lower roster sizes they’ll be against more skilled opponents. 2) A hard filter that just prevented them from being considered at all (at least until a fail-safe kicks in.)

As for TeamSpeak, we can’t reliably detect that to make it a factor.

Sebastian.4610:

seems to me like pvp got the ol casual catering patch. #1 player on leaderboards is a “bad” player that most people think is a bot because of how mindless it runs around and doesnt help the team. #1 player also has a negative win rate .. yet is #1 .. seems flawed.

This logic is flawed. 47% win rate is within the range you’d expect to see if matchmaking is working well.

I’ve looked at their game history… most losses, when they lose, are narrow losses. Maye they’re in some matchmaking sweet spot where they are getting frequent good quality matches. It could also be simply that they’ve been taking ranked serious from day 1, where everyone else hasn’t.

-shrug-

Silentshoes.1805:

Yes, but there was a different system behind that button and it has likely been erased.

I wish! It’s still being used to host custom arenas and activities.

Silentshoes.1805:

If they do what I stated, you as a solo player will always play solo Qers only. So don’t worry about the button.

The new systems can do a solo queue, we just don’t feel it promoted our goals or fit with where we want to take the game.

That doesn’t mean we like people getting into situations like 5 vs 1+1+1+1+1, so we will fix this one way or another. In fact, I think I found the culprit so it may happen less far less often after the 16th. I say ‘may’ here only because there are still situations where it could happen. E.g. it’s 3am and they’ve been waiting in the queue a long time and you’re the only players near their skill level.

crunchyraisin.6054:

Top teams are missing out on points because they aren’t being matched up against equal opponents (because there isn’t one playing at the time).

I agree with this, and it’s something to worry about and must be included in any reasoning about how the ladder works.

It may end up balancing itself out because easy wins take less time. That’s negated right now by longer queue times, but the build on 16th should include a change that will keep queue times shorter by dynamically scaling min-potentials.

crunchyraisin.6054:

Top players tend to play on teams. With this matchmaking system teams tend to win a high percentage.

Ladder points factor in rating, rank, and roster size when predicting odds of victory. I don’t see how that’s a bad thing. If I went against Abjured or TCG and won (hah!), I should get more points than if they beat me.

Maybe the issue is people believe we are calculating odds too generously? That we’ll be giving out +3 points left and right?

Silentshoes.1805:

What puzzles me is that if top players are winning 80% of their matches, then their MMR is not high enough. Shouldn’t they typically be winning 60-65% if they are really good?

Yes and no. With ideal matchmaking, and a sufficient # of players available at any time, everyone should be close to 50%. With random matchmaking, the best players should definitely have a higher win %.

With real world matchmaking it’s somewhere between the two for any number of reasons, so it’s not uncommon to have people with higher than 50% averages.

Factoring in MMR somehow might be workable (though we are already doing that by changing the number of points you win or lose), but adding anything more might awkward since we don’t expose that information.

Using the number of wins, or the win/loss ratio, is asking for problems. The number of wins vs loss does not reflect your skill because it’s not from a large random sampling of matches, and does not include critical information, i.e. 1) who you played against, and 2) how close of a victory it was.

As far as people’s worries about it being number of games played that means you’re at the top, try to keep a few things in mind.

• Seasons will be short.
• A 500-50 record is more impressive (and more statistically significant) than 10-1 record.
• Eventually, we’ll show you the match prediction information and tell you how you need to perform to get X number of points.

The ladder points system already gives players what they are trying to find by looking at wins vs losses, but is also more accurate because it works around the non-random sample by carrying with it the information that is missing from wins/losses. (See 1 & 2 above.)

I’ve looked at the top people, they’re playing games against opponents that are challenging, and they are consistently winning or coming close to winning.

I think part of the issue is that since 1 & 2 are not immediately apparent (we don’t even show it in-game yet), we humans prone to assume it doesn’t exist, is trivial, or is random.

Aaron.2413:

TLDR Don’t make most games = Top, and a .5 w/l vs a .9 w/s lol.. you guys said the .5 was better because he was LOSING closer…. thats ridiculous, the .9 is face rolling… BECAUSE HE IS BETTER

Not necessarily true. The matchmaking could be flawed for any number of reasons. E.g. they tanked their MMR before the start of the season to get easy wins. We are underplaying the benefit of queuing as a group. They only play late at night when matchmaking quality is lower.

If we had completely random matching W/L might mean something, but as long as we filter by rating it’s tainted by selection bias.

Relentliss.2170:

…

No, we aren’t bringing solo queue back. Yes, I am working on making sure you’re less likely to be stacked against pre-mades.

The change I’m referring to benefits everyone because we’re purposely holding back the matchmaking system until this fix goes through. Without this fix, queue times grow during off-peak hours to levels we don’t find acceptable… which causes a feedback loop that increases queue times further as people leave the queue because of the wait.

It benefits outliers because there are far fewer players available at their level, but the system doesn’t (yet) compensate for that fact, so it simply won’t match them. There may be enough players in the queue to form a good match for them, but not enough to meet the amount we prefer for the common scenario. We want a larger pool of choices so we can be picky about thing like profession. So instead they sit in queue until the fail-safes kick in (which increases the range of possible matches by lowering standards) and the system gives them a lower quality match. In that way it benefits the more typical player by preventing the fail-safes from engaging in the first place.

lilwizz.5783:

…

I’ve got a fix for this that will let us ramp up min-potentials, and dynamically scale it back for off-peak hours or outliers.

Should be in for the 16th build unless something goes wrong.

Lordrosicky.5813:

This last part is vital to the games success. I posted this many times over the last year on the forums. Please don’t sacrifice this ideal because it is so important. Some people will QQ about it. But they have vested interests. It is so important that less skilled people who play alot are rewarded.

I think what people might be forgetting, is that we also have tournaments, ESL cups, etc. to cater to those who want skill to be the absolute determinant.

Lopez.7369:

Question: If someone starts the season with a high MMR, and only gets matched with similarly rated or lower-rated people, giving him a 40 to 100 percent of chance of winning, does that mean he’ll have to grind a lot of one-point games to get to the top of the ladder, even though his MMR is so high?

Possibly, if matchmaking was ideal. Ugg, I’d really hate to argue that matchmaking not being perfect was a positive thing. :P

If the beta ladder does show that sort of behavior, what would you guys like to see? The number of points adjusted by the rating relative to the default rating, instead of just relative to the players in the game?

Well, the new matchmaking shouldn’t be making one sided matches.

The exception being when they’re caused by a full party that the system is having trouble matching them. (Looking at you Booms.)

Let me know if you feel like you’re getting trapped and I’ll monitor your games.