Guild Wars 2

Sure Glicko2 is wonderful for accurately rating a chess player, and in other single person games.

And it can work for teams of random players if there is a sufficiently large number of people queuing for teams at all times, as in League of Legends.

But in GW2 at the moment, we have a very small pool of players, with skill ability skewed towards low level.

I just don’t see how Glicko2 can accurately judge your ability to win by assigning you to teams from a small pool of available players, many of whom are less skilled.

One very high MMR may mathematically balance 4 low MMRS…but that does not translate to an equivalent team "win probability " in SoloQ. It’s probably not a linear scale, but rather is a “discrete step function” on skill level. Meaning, sometimes a low MMR player cannot really contribute ANYTHING…but the math assumes they contribute a bit less than 1/5 of the win capability of the team.

Also, class types in a team are important for winning a match, but they are random for matches you are expected to win… to keep or raise your MMR rating.

There may be so much “noise” in the system that the true rating will not emerge for thousands of matches of the same set of people. New players introduce new “noise” in the data, skewing it, and keep the ratings inaccurate much longer still, I suspect.

Thoughts?

Justin,

I think you and the team have done the very best you could, with the situation you all are faced with. I am a supporter! And I had high hopes. Yet, in playing and thinking about it, I now see the issues and so I posted them for discussion.

One thing the current leaderboard absolutely requires is that all the more skilled players play constantly. They are in shorter supply, and when a similarly skilled player joins SoloQ, she or he needs to play against a team near her level, with teammates who are similar in level.

When the skill levels diverge, (and the class balance is off) the “step function” comes into play and it all falls apart. Yet the algorithm dutifully adjusts your MMR as if the match had been balanced reasonably.

SO if people get into the top 100 and stop playing, they tend to stay there too long. They are not exposed to the same forces that randomly force some losses.

I suspect that the people in the Top 100 are mostly good players who have been given reasonable matches where they did not have to “carry” more than is possible.

I suspect the “percentages” area has a smaller number of good players, who “fell into the pit” with bad teams and bad team compositions.

There are a number of problems with assuming that there’s a magical solution to finding everyone’s perfect MMR in SoloQ:

1) Every player has variance in their ability depending on time of day/fatigue/health/how drunk they are.

2) Player skill and knowledge is constantly changing. The skill of inherently good, but new, players will go up each match as they learn new tricks from watching others/dying to them (e.g. terrain use/positioning on maps — you only have to see it once to know that you can provide pressure onto a point from certain areas better than others)

3) Players are not equally skilled at all professions, yet they sometimes swap what they play on-demand, or just for fun. Also, professions aren’t necessarily balanced. This adds even more noise.

Given these problems and the massive amount of day-to-day and match-to-match variance in player ability, the current system does an amazing job balancing the teams in SoloQ.

Justin ODell.9517:

Laserbolt.6731:

SO if people get into the top 100 and stop playing, they tend to stay there too long. They are not exposed to the same forces that randomly force some losses.

Actually, your ratings deviation increases due to inactivity, so the longer you have been inactive the wider and more random your matches are likely to become. This was masked previously by the flaw in how players were matched.

Interesting.

Still, given the challenges of matching people into teams, I would expect that the top 100 would be less “static” then they currently seem to be.

I would expect them to get teams of low MMR teammates, on a regular basis, and go down rather quickly once they play again regularly.

Kharr.5746:

There are a number of problems with assuming that there’s a magical solution to finding everyone’s perfect MMR in SoloQ:

1) Every player has variance in their ability depending on time of day/fatigue/health/how drunk they are.

2) Player skill and knowledge is constantly changing. The skill of inherently good, but new, players will go up each match as they learn new tricks from watching others/dying to them (e.g. terrain use/positioning on maps — you only have to see it once to know that you can provide pressure onto a point from certain areas better than others)

3) Players are not equally skilled at all professions, yet they sometimes swap what they play on-demand, or just for fun. Also, professions aren’t necessarily balanced. This adds even more noise.

Given these problems and the massive amount of day-to-day and match-to-match variance in player ability, the current system does an amazing job balancing the teams in SoloQ.

Good points. (Though some pretty dedicated, good players who are struggling down in the “percentages” might disagree about how good the matchmaking can be right now.)

Justin, to answer your question, I think the “class MMR” is very valid. It’s something I was thinking about a few days ago: that some classes are just more powerful for the current SoloQ team mixes, and there should be an MMR adjustment for what class you are playing.

Justin ODell.9517:

Kharr.5746:

There are a number of problems with assuming that there’s a magical solution to finding everyone’s perfect MMR in SoloQ:

1) Every player has variance in their ability depending on time of day/fatigue/health/how drunk they are.

2) Player skill and knowledge is constantly changing. The skill of inherently good, but new, players will go up each match as they learn new tricks from watching others/dying to them (e.g. terrain use/positioning on maps — you only have to see it once to know that you can provide pressure onto a point from certain areas better than others)

3) Players are not equally skilled at all professions, yet they sometimes swap what they play on-demand, or just for fun. Also, professions aren’t necessarily balanced. This adds even more noise.

Given these problems and the massive amount of day-to-day and match-to-match variance in player ability, the current system does an amazing job balancing the teams in SoloQ.

1) This is basically impossible to predict or deal with. If you’re constantly playing while in a state that your skill is significantly impaired, I’d say a lower MMR is proper.

2) The system is designed to allow MMR to raise or fall over time, just not as rapidly as when your deviation is high (i.e. new player or long periods of inactivity.)

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

If there were MMRs per profession, it might be easier to make balanced teams since this would eliminate some of the noise. I know I’m personally better with some professions than others simply because I have more time invested and muscle memory for my utility skills.

Justin ODell.9517:

Kharr.5746:

There are a number of problems with assuming that there’s a magical solution to finding everyone’s perfect MMR in SoloQ:

1) Every player has variance in their ability depending on time of day/fatigue/health/how drunk they are.

2) Player skill and knowledge is constantly changing. The skill of inherently good, but new, players will go up each match as they learn new tricks from watching others/dying to them (e.g. terrain use/positioning on maps — you only have to see it once to know that you can provide pressure onto a point from certain areas better than others)

3) Players are not equally skilled at all professions, yet they sometimes swap what they play on-demand, or just for fun. Also, professions aren’t necessarily balanced. This adds even more noise.

Given these problems and the massive amount of day-to-day and match-to-match variance in player ability, the current system does an amazing job balancing the teams in SoloQ.

1) This is basically impossible to predict or deal with. If you’re constantly playing while in a state that your skill is significantly impaired, I’d say a lower MMR is proper.

2) The system is designed to allow MMR to raise or fall over time, just not as rapidly as when your deviation is high (i.e. new player or long periods of inactivity.)

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

Id like profession specific MMR and leaderboards, even though it wouldnt affect me, I only play Mesmer anyway ;D.

Also, people can easily switch after they get a match, so for profession specific MmR to work, you’d have to lock people for character switching first :/. I dont, but I know that some people swap depending on composition.

I’d really love profession specific leaderboards, though :P

Justin ODell.9517:

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

ooooo very good!!!!

for super street fighter 4 arcade edition ranked matches, they have something similar.

PP player points = account rating points
BP battle points = character rating points

The system must have something to make it easier to increase your MMR over time…. I mean, as I can notice, the more people play, Lower their MMR is….

Lets suppose i am a top 10 player and my score is 20×0 , so i will stop playing not to lose and decrease my MMR, because until now it was easy to Win, but things are getting much more difficult and i never lost… But another guy that is much better than me has a score of 2430×3 and he will be placed second, because he failed 3 times…. A good system would take number of matches and victory into account and place the other guy much ahead of me….

Same in boxing, lets supose we take Sugar Ray Robinson 173 wins, 108 KO and 19 losses, the greatest ever lived and compare with an young boxer with 20 wins, 3KO and 0 losses… Your leaderboard will put the young boxer in a better position than a champion legend… That is not measuring skill at all !

Consider Fantasy Football team projections and player score projections…how wrong they can be.

Now think about the MMR rating trying to predict how well this team (that was formed for the first times just seconds ago) will perform against this other team that was just formed.

Justin ODell.9517:

Laserbolt.6731:

Though some pretty dedicated, good players who are struggling down in the “percentages” might disagree about how good the matchmaking can be right now.

In every instance I’ve investigated players are being matched very well as far as MMR spread is concerned. You don’t need to carry a team, you just need to provide enough of an advantage to win. The other team is in the same boat with the same chance at team composition.

What is especially significant, in my opinion, is the frustration of dealing with players that are less serious and less experienced, and I hope using ladders in matchmaking, among other things, will help.

Yes, but this assumes the MMR is accurate for each player. But their MMR came from the performance of their teammates…including the non-serious goofs, or a 4×5 match problem. There is a randomness there behind those exact-seeming MMR rating numbers. That’s the problem with a number. It looks authoritative. It should be more “fuzzy”.

fodem.2713:

The system must have something to make it easier to increase your MMR over time…. I mean, as I can notice, the more people play, Lower their MMR is….

That will always be the case, because a low sample size (games played) will always have a greater variance than a high sample size.

In other words, some players with few games played will have a higher win percentage than they “should” and some will have a lower win percentage than they “should.” Of course, we’ll only notice the ones with the high win percentages.

Think of it like flipping a coin. In 10 flips, it’s just possible that you could get 100% heads, but that would be unthinkably improbable over 500 flips.

Excellent point, NeverSayDie.

Justin ODell.9517:

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

This is something I can support, at the moment it is painful to take a class you haven’t had much hands on time with into team queue.

Justin ODell.9517:

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

That would be great, but some kind of class based queue should be needed prior to that.
Right now, if lets say 3 bunker guards are matched in the same team, someone will swap to a different character. With a class based MMR that swap should be forbidden, so the system itself should somehow avoid placing those players in the same team.

Justin ODell.9517:

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

I would love this…

Currently, I am terrible at playing Elementalist, so I won’t ever play that class in sPvP, my Ele having her own MMR could let me play that class in sPvP.

Vargamonth.2047:

Justin ODell.9517:

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

That would be great, but some kind of class based queue should be needed prior to that.
Right now, if lets say 3 bunker guards are matched in the same team, someone will swap to a different character. With a class based MMR that swap should be forbidden, so the system itself should somehow avoid placing those players in the same team.

make it so that once a character has entered the sPvP map, they cannot change professions.

Deimos Tel Arin.7391:

Vargamonth.2047:

Justin ODell.9517:

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

That would be great, but some kind of class based queue should be needed prior to that.
Right now, if lets say 3 bunker guards are matched in the same team, someone will swap to a different character. With a class based MMR that swap should be forbidden, so the system itself should somehow avoid placing those players in the same team.

make it so that once a character has entered the sPvP map, they cannot change professions.

That’s absolutely needed for having a class based MMR.
The problem arise when multiple instances of some classes are matched together, which can put a team in serious disadvantage.
Currently that’s easily solved by swapping characters. Once this option is gone, you want the matchmaking to avoid this team compositions.

this dual rating is a great soloq point. locking classes for it would be a must but maybee at prematch countdown there was a tag for teammates stating what type of build (burst, bunker ,condie) they were that a player could swap if wanted to an alt build of that character only. not sure if this would work but i like that the class mmr rating would be a instant help with class balance. not a perfect fix. not added for team q though.

Just want to throw in my support for the Character-bound MMR.

I’m pretty good on my Engineer… pretty bad at everything else.

Justin ODell.9517:

Kharr.5746:

There are a number of problems with assuming that there’s a magical solution to finding everyone’s perfect MMR in SoloQ:

1) Every player has variance in their ability depending on time of day/fatigue/health/how drunk they are.

2) Player skill and knowledge is constantly changing. The skill of inherently good, but new, players will go up each match as they learn new tricks from watching others/dying to them (e.g. terrain use/positioning on maps — you only have to see it once to know that you can provide pressure onto a point from certain areas better than others)

3) Players are not equally skilled at all professions, yet they sometimes swap what they play on-demand, or just for fun. Also, professions aren’t necessarily balanced. This adds even more noise.

Given these problems and the massive amount of day-to-day and match-to-match variance in player ability, the current system does an amazing job balancing the teams in SoloQ.

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

Yes, i have an opinion: I think having it based on account causes people to be pigeon holed into the class they are best with, not the one they most enjoy.

For instance, I am probably my best on warrior (shocker, right?) but I enjoy the Mesmer the most. Because my skill level on mesmer is probably 30% less effective than that of my warrior, I immediately risk damaging my MMR and leaderboard ranking by playing my Mesmer. So, my only option at that point is the play Hotjoin, which is a problem, since I don’t find that mode fun at all…as in, 0.

So, I think the MMR has to become character based. That will most certainly alleviate much of the issue of multiple classes of the same type being found in Solo Q. You are essentially forcing people to play flavor of the month classes and then penalizing them by destroying their MMR if they try to play with a different class.

Please take this back to the PvP crew and tell them this is a necessity.

Thank you

I think couple things affect outcome a lot

1) Some people are more eager to win match no matter what, so they go and change class after teams have been made. This can flip the balance a lot. Suddenly match which seemed fair at start turns out complete faceroll for other team.
2) A lot strong classes massing on same team. Sometimes unless like majority of team would swap classes there’s just no chance. Take for example 2 condinecros, 2 Warriors and mesmer or engi or something.. that’s something very difficult to overcome if other team is less than optimal. Or pet/minion/clone spamming team against single target team..
3) Too low ranked player in team will have 0 effect on outcome. Rank 6 guy is not going to do anything useful, like in 95% of cases. Like.. Team with one rank 50 3xrank30ish 1xrank6 is very likely to lose to team with like rank40 3xrank30ish 1xrank 20ish.. Since rank 20 guy is likely to be at least decent if not good. There’s of course some people who can but that’s like extremely rare. They’re either one-shot-wonders or just don’t know how to play for points etc. That hurts balance a lot. I’d rather wait extra 5 minutes to get in team with equally ranked people than just lose impossible matchup. Currently it’s more like playing dice who gets more newbs or happens to get more fotm setups.

I have a question regarding the old/new MMR. How long will it take the new MMR to stabilise?
In the old MMR,i was ranging in the 600+ rank on NA solo queue leaderboards.

After the MMR resets,I’ve gradually increased my rank from 900+ to 500+ to 200+ and now I’m in the 180+.

Also,what happened to the 400 players who had better ratings than me in the past?Did decay remove them or was the old MMR not consistent enough to give accurate ratings?

Thedenofsin.7340:

Justin ODell.9517:

Laserbolt.6731:

Though some pretty dedicated, good players who are struggling down in the “percentages” might disagree about how good the matchmaking can be right now.

In every instance I’ve investigated players are being matched very well as far as MMR spread is concerned. You don’t need to carry a team, you just need to provide enough of an advantage to win. The other team is in the same boat with the same chance at team composition.

So you’re basically saying the system can make either team suck because it doesn’t have a line of code that can differentiate between a player with 100 hours in pvp vs 1 with 1 hours.

Great….

Hours in PvP? I’d say that metric is only valuable if your hours in PvP is less than 10. Until you figure out what conquest is, I think it’s a poor indicator of MMR.

I do think that the players with a high MMR right now 1) Are decent, 2) have gotten good matches that raised their MMR, and 3) play a build that helps the team a lot, giving an advantage.

But I do notice that once people get in the top 25 they stop playing. And when they DO play again, they go down.

Justin ODell.9517:

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

Does profession-bound MMR even need to be discussed? It’s obvious that’s the optimal solution. Why account-bound MMR exists in the first place is beyond me.

Thedenofsin.7340:

Justin ODell.9517:

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

Does profession-bound MMR even need to be discussed? It’s obvious that’s the optimal solution. Why account-bound MMR exists in the first place is beyond me.

Easy there, sparky. I think there are definitely things to discuss.

@ Justin O’Dell, my opinions on profession based MMR

• 1) In GW2, it’s extremely easy to multi-class and have all of those classes at, or around your proper MMR. This isn’t a gear based MMO where your alts are all in poverty/welfare gear and you’re fighting an uphill battle. In conquest, mastering the specific profession’s mechanical skills are certainly important, however, skills such as cooldown management, proper rotations, and many of the ‘winning determinants’ are profession independent.

• 2) Deceptive MMR. 5 people get on professions that they’ve never played before and queue for team queue. They get matched up with the appropriately matched team, at their MMR level. Those 5 people then get on their mains and proceed to demolish opposing team with legitimately less MMR. For this to be a thing, you’d have to profession lock when queuing. And I don’t want that – I want to be able to make different picks if needed.

There are a few more minor issues, but these are my biggest two, in regards to profession based MMR. Especially my #2. But hey, if profession MMR was weighted very low, I would never say that diversifying the MMR determinants – effectively reducing the margin of matchmaking error could be bad.

Justin ODell.9517:

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

I would say just one MMR per character is enough. The skill at which one person plays a class does not transfer over to another character, this would have to be done when you are no longer allowed to swap classes after queuing due to queuing on a low ranked class then griefing with a different one.

However I have always been partial to a dual numerical system, glicko rating (impervious to point delay but retains the volitility) and a second one which would sort pf be your rating. if your “second” rating is much lower than your opponents glicko rating and you win, you get a lot more points, if you lose you don’t lose out on much. The same thing if you have a low glicko rating but somehow managed to gain a tonne of rating points you will be paired against people based on mmr so if you lose you plummet fast, and if you win you get next to nothing. Just my thoughts though.

Chaith.8256:

Thedenofsin.7340:

Justin ODell.9517:

3) I’ve pitched the idea internally of having two MMRs, one account bound and the other per profession. Does any one have an opinion on this?

Does profession-bound MMR even need to be discussed? It’s obvious that’s the optimal solution. Why account-bound MMR exists in the first place is beyond me.

Easy there, sparky. I think there are definitely things to discuss.

@ Justin O’Dell, my opinions on profession based MMR

• 1) In GW2, it’s extremely easy to multi-class and have all of those classes at, or around your proper MMR. This isn’t a gear based MMO where your alts are all in poverty/welfare gear and you’re fighting an uphill battle. In conquest, mastering the specific profession’s mechanical skills are certainly important, however, skills such as cooldown management, proper rotations, and many of the ‘winning determinants’ are profession independent.

• 2) Deceptive MMR. 5 people get on professions that they’ve never played before and queue for team queue. They get matched up with the appropriately matched team, at their MMR level. Those 5 people then get on their mains and proceed to demolish opposing team with legitimately less MMR. For this to be a thing, you’d have to profession lock when queuing. And I don’t want that – I want to be able to make different picks if needed.

There are a few more minor issues, but these are my biggest two, in regards to profession based MMR. Especially my #2. But hey, if profession MMR was weighted very low, I would never say that diversifying the MMR determinants – effectively reducing the margin of matchmaking error could be bad.

Your first point carries very little water. If you are equally skilled in multiple professions, then profession-based MMR will reflect that. If not, then it won’t.

To your second point: if team “A” swaps out against team “B” and wipes them, the only impact this would have is to marginally increase (if at all) the MMR of the professions that actually played the match, as opposed to the ones that queued for the match.

Thedenofsin.7340:

To your second point: if team “A” swaps out against team “B” and wipes them, the only impact this would have is to marginally increase (if at all) the MMR of the professions that actually played the match, as opposed to the ones that queued for the match.

Yes, but in the meantime a newbie team has had their face melt. Not funny at all. You risk to loose potential longterm pvp players. And if the team A keeps at it, they could get pretty high with zero to no effort, step after step, with every “marginal increase”.

For the topic about account and profession rating, let me add an example.
I’m a bit of an altoholic in pvp, as it’s so easy to set up and go, so I tried pretty much all professions. On some of those, I am pretty BAD, and I know it, but I like to play them once in a while for the sake of variety, for me and for the pvp world around me (I hate FOTM effects). A profession rating could help people like me to face reasonable matches when doing that, and prevent me from worrying about loosing my position with my “main”.
On the other hand, testing all the profs, tinkering with all the traits and things, gave me a kind of overall understanding on basic mechanics, in general, and class-related mechanics, in particular, that clearly contribute to my skill level when playing any profession. As an example, I had no problem switching from a staff necro to a grenade engi because I already had an ability to use large amount of ground targeting, or when I play against mesmers, having played as one helps me recognize from movements which one is the real deal. All of those kind of general based knowledge should be taken in account when putting me in a match, and an account based rating could help just doing that.

On how to deal with the rewards and ladder position when having those multiple numbers per player, however, is beyond me. Maybe a “weighted arithmetic mean” could be used: calculating general account rating per se, and then mixing it with profession rating in relation to games played on that prof.?

Thedenofsin.7340:

To your second point: if team “A” swaps out against team “B” and wipes them, the only impact this would have is to marginally increase (if at all) the MMR of the professions that actually played the match, as opposed to the ones that queued for the match.

So basically, you’re saying for the team that was the victim of an exploit causing them to face opponents much better than them, don’t worry – they didn’t get much of an MMR gain?

You have to consider the public interest and well being at large when deciding the right standards.

Chaith.8256:

Thedenofsin.7340:

To your second point: if team “A” swaps out against team “B” and wipes them, the only impact this would have is to marginally increase (if at all) the MMR of the professions that actually played the match, as opposed to the ones that queued for the match.

So basically, you’re saying for the team that was the victim of an exploit causing them to face opponents much better than them, don’t worry – they didn’t get much of an MMR gain?

You have to consider the public interest and well being at large when deciding the right standards.

I am considering the public interest. Eventually, if the disparity becomes so great, the team which swaps will not gain any rank for winning, and the other team will not lose any rank. Then the problem fixes itself.

Besides, you’re talking about team queuing, which will use a separate rank for the team. For this to be a problem in solo queue, you’d need a full team to sync, which apparently is no longer possible.

Sorry, I have to quote myself as it seems I went unnoticed

LorVan.1029:

Yes, but in the meantime a newbie team has had their face melt. Not funny at all. You risk to loose potential longterm pvp players.

I think that this was the issue about “public interest”. Newbies don’t care about ladder, or rank, or whatever, position, but about how funny are the first games they play in pvp mode.

Chaith.8256:

Thedenofsin.7340:

To your second point: if team “A” swaps out against team “B” and wipes them, the only impact this would have is to marginally increase (if at all) the MMR of the professions that actually played the match, as opposed to the ones that queued for the match.

So basically, you’re saying for the team that was the victim of an exploit causing them to face opponents much better than them, don’t worry – they didn’t get much of an MMR gain?

You have to consider the public interest and well being at large when deciding the right standards.

I think this is a good point. There would have to be some sort of “lock” system to prevent exploitation once the match starts so that people can’t swap classes mid-game as they currently can. I was under the impression that classes/skills were locked in soloQ but saw a ranger swap to his warrior mid-match last night in soloQ (was he exploiting? — if he was, mod can contact me for more details)

Suggestion: This is a technical barrier, but if Anet can show the roster with all the classes on your team before the match starts for 60 seconds for soloQ (similar to how TeamQ works right now — if you queue solo, you can see your team’s roster in the tPVP menu), players can change their class to match their team’s composition better.

If someone swaps to a class that’s way above/below their matched MMR, they get booted back into queue and the match doesn’t start. After it starts, professions get locked.

This sort of approach will allow players that are equally skilled at multiple classes to swap between professions pre-match to balance their team’s composition while:
1) preventing players from exploiting the matchmaking system by swapping to a profession they are better with
2) preventing players from dragging their team down by swapping to a profession they are worse with

LorVan.1029:

Sorry, I have to quote myself as it seems I went unnoticed

LorVan.1029:

Yes, but in the meantime a newbie team has had their face melt. Not funny at all. You risk to loose potential longterm pvp players.

I think that this was the issue about “public interest”. Newbies don’t care about ladder, or rank, or whatever, position, but about how funny are the first games they play in pvp mode.

With the advent of solo / team queue restrictions on players below a certain rank/level/# of matches, this problem would be obviated as well. New players would have had plenty of times to get their face melted in hotjoin and quit well before they queue solo (or team.

Kharr.5746:

Chaith.8256:

Thedenofsin.7340:

To your second point: if team “A” swaps out against team “B” and wipes them, the only impact this would have is to marginally increase (if at all) the MMR of the professions that actually played the match, as opposed to the ones that queued for the match.

So basically, you’re saying for the team that was the victim of an exploit causing them to face opponents much better than them, don’t worry – they didn’t get much of an MMR gain?

You have to consider the public interest and well being at large when deciding the right standards.

I think this is a good point. There would have to be some sort of “lock” system to prevent exploitation once the match starts so that people can’t swap classes mid-game as they currently can. I was under the impression that classes/skills were locked in soloQ but saw a ranger swap to his warrior mid-match last night in soloQ (was he exploiting? — if he was, mod can contact me for more details)

Suggestion: This is a technical barrier, but if Anet can show the roster with all the classes on your team before the match starts for 60 seconds for soloQ (similar to how TeamQ works right now — if you queue solo, you can see your team’s roster in the tPVP menu), players can change their class to match their team’s composition better.

If someone swaps to a class that’s way above/below their matched MMR, they get booted back into queue and the match doesn’t start. After it starts, professions get locked.

This sort of approach will allow players that are equally skilled at multiple classes to swap between professions pre-match to balance their team’s composition while:
1) preventing players from exploiting the matchmaking system by swapping to a profession they are better with
2) preventing players from dragging their team down by swapping to a profession they are worse with

You’re overthinking the problem. This really isn’t a significant issue, and if it is, ArenaNet can implement a band-aid that would effectively discourage that behavior.

If people are really concerned about this minor secondary effect, then all ArenaNet has to do is change the MMR reward thresholding for players that swap after entering the game. Instead of not rewarding points/rank/etc when the differential exceeds “X”, they change it to 0.5X for players that swap.

lol i would never wait a hour for a match – before i would delete the game^^

even waittimes now is hard borderline

Copy / paste GW1-Ladder and you’re done.
I’m not sure why you always need to reinvent the wheel.

What’s in gw1 STAY in gw1, we talkin GW2 here mate wish you saw the difference.
Bumping this topic so that people know what is the meaning of grinding ranks/achievement against competitive battle