Guild Wars 2

I’ve been trying to think through how ranks 21 (DR) and 22 (HoD) got matched up with a 14 (BP). If you look at the raw tiers, this is a +/- 3 tier spread. BP is tier 5, DR is tier 7 and HoD is tier 8. The description of the new system states that BP should never be matched up with a tier 8 because that’s three tiers away.

However, if you look at the raw server rankings and break THOSE up by tiers, you’ll see this:

14 Borlis Pass
15 Anvil Rock
16 Darkhaven
—-
17 Isle of Janthir
18 Gate of Madness
19 Sorrow’s Furnace
—-
20 Northern Shiverpeaks
21 Devona’s Rest
22 Henge of Denravi

When you just look at the tier distribution based on static triplets, HoD is top t8 and BP is middle t5, and they’re four tiers apart (making it +/- 3). If you look at it this way, on a strict numerical distribution, HoD and BP are polar opposites in a three tier spread, making it viable for a +/- 2 tier distribution match up like what happened.

There is no mention that I can see in the official blog post about whether or not this spread is based on raw rank or on the existing tiers.

If this is as designed, +/- 2 is too high of a spread in my opinion. I’m sure HoD/DR isn’t the only pairing that is getting stomped so far this week.

What are these tiers you are talking about, and why do you think they are arguments/inputs for the stochastic function that determines a server’s matchmaking rating?

Oh, and you can find the blog post that talks about the nuts and bolts of the matchmaking system here:

https://www.guildwars2.com/en/news/big-changes-coming-to-wvw-matchups/

You can see exactly how they are determined.

Poor design system Mr. Devon. 2 weeks in a row the matchups have been 600-50-40ppts for our servers. If you call this balanced, I’d really hate to see imbalance.

Sorry Devon I have no idea what you are trying to say. All i know is that we are ticking 400+ points all day because we have been lumped into an uneven match up.

In a one up one down scenerio, Tarnished Coast could slowly build the forces they need by recruitment or other means to compete in a higher tier. Leaving things to a random generator will/has lead to exactly this, poorly balanced and boring match ups.

This way of matching servers is 100% worse than the old system, it is nonsensical in every way. What is so wrong with one up one down? It would work better than both the old and current systems of matching servers.

Really, you have start thinking about the people that play WvW. Not only have you made arrows carts game breaking and unmanageable, you’ve now managed to alienate players further by creating unreasonable match-ups that allow for even more stomping of outmanned players, and that’s without brinGing in the initial design flaw of how the points system works. Neither of the other servers in my servers match-up alone stand a chance, we will run them over all week, driving more people away from WvW and ultimately the game.

Are people listening to him? He’s saying that server ratings decided these matchups. If these matchups are bad, it’s because the server ratings were bad. If the server ratings are bad, then you need to shakeup the matchups in order to get more accurate ratings, because the current ratings were gained from stagnant matchups.

One up one Down wouldn’t need ratings, it would just mean that people would fight harder to not be in third place. And quality of victory is laughable right now, and really will be for the foreseeable future. I don’t feel good about ticking 500+ all day every day, that is no quality of victory for anyone involved.

Invalidating ratings is the point. The rating system is what led to stagnant match-ups in the first place. I actually was fine with the old system because the matches were well-balanced, but if you must cater to whiners who need to see different tags over people’s heads, this new system is not a good way to go about it.

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in. The system we are using now creates varied matchups, which will allow us to get more accurate ratings. As those become more accurate we can dial down on the amount of variance we are introducing to have matchups that are varied, but less prone to large rating gaps between the servers. But, again, this is entirely based on server rating, which is the thing we have actually been calculating this entire time. Until that number is a more accurate reflection of server quality, we are going to keep mixing the matchups up.

This is pretty much the exact response that has been given since the game launched. So far there has been very little accuracy to the ratings across the board. The system is broken and allowing these types of matchups to “correct the ratings” just leads to further skewing of WvW populations on individual servers and therefore further invalidating the ratings.

Wrex.6798:

This way of matching servers is 100% worse than the old system, it is nonsensical in every way. What is so wrong with one up one down? It would work better than both the old and current systems of matching servers.

Really, you have start thinking about the people that play WvW. Not only have you made arrows carts game breaking and unmanageable, you’ve now managed to alienate players further by creating unreasonable match-ups that allow for even more stomping of outmanned players, and that’s without brinGing in the initial design flaw of how the points system works. Neither of the other servers in my servers match-up alone stand a chance, we will run them over all week, driving more people away from WvW and ultimately the game.

Maybe if you didnt super stack a server and turn it into the NA version of vizunah, which you guys left EU to get away from, you wouldnt be complaining. I told you guys before you left EU to come to NA that you needed to recruit other EU guilds to go to the other servers. You really should of listened.

Oh Wish, explain to me if you were super stacked when you got all your gold medals for winning tier 1? Pot Kettle Black at all? Especially as you guys worked hard to recruit us as well when in second spot!

Wish, I don’t care about other EU guilds, we actually have decent competition in our prime time. And don’t moan about stacking, we’re talking about the actual match-ups not how full servers are. It would make little difference if we were still on the EU tiers, they are as borked as the NA tiers.

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in.

Here’s the thing Devon – Ratings are rubbish and will never be fully accurate. Besides what is more important, fun for players or you having an accurate system? Right now you are shafting the former to try and create the latter.

Offski.4897:

Oh Wish, explain to me if you were super stacked when you got all your gold medals for winning tier 1? Pot Kettle Black at all? Especially as you guys worked hard to recruit us as well when in second spot!

When JQ was winning NA T1 JQ had lots of Sea and NA players, and thats it, thats the way JQ has been for a long time, and at the time no one really had any EU players, or super late night NA players. When Iron was thinking of moving JQ had already lost a bunch of it’s NA people, and BG had already gotten a bunch of new people who played during EU time, so of course JQ wanted to recruit you. Having played on Desolation with you guys for a bit when you were in T2 I personally thought it would be fun to play with you again, but because of that, I recognized exactly what would happen if you moved to any of the NA servers. You’d roll over everything and take it all. Which is exactly what I told you. I also told you that you’d need to recruit other EU guilds for the other servers you did not go to, or you wouldnt really have anything to do aside from PVD. Complaining about it now just makes you look sad IMO, since I know for a fact I told your rep on JQ all this ahead of time.

Does any anet employees actually PLAY the game themselves? Because if they did they’d realize that what they did to arrowcarts and now the matchups is game breakingly bad. this goes way beyond fubar. this is just dumb.

SoR has won tier1 NA for a few weeks in a row now. we’re the strongest server atm. then wham bam! we’re forced down to fight servers who don’t stand a chance. we had 3 waypointed garrisons when I logged on this morning and we’re currently ticking 500+. seriously anet. you messed up. BIG TIME. revert this nonsense and start actually listening to what the PLAYERS who PLAY wvw are actually saying needs done!

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in. The system we are using now creates varied matchups, which will allow us to get more accurate ratings. As those become more accurate we can dial down on the amount of variance we are introducing to have matchups that are varied, but less prone to large rating gaps between the servers. But, again, this is entirely based on server rating, which is the thing we have actually been calculating this entire time. Until that number is a more accurate reflection of server quality, we are going to keep mixing the matchups up.

All of that is completely true, but it beats the crap out of me why you folks (ANet) can’t grasp the fact that the fundamental fault here is the WvW population imbalance between servers … period. You can shake things up all you want … and yes, it will give more accurate ratings … but you’re still going to end up with significantly imbalanced matches each and every week, and the “fresher” they are the more imbalanced they will be. Everybody (well, almost everybody) on these forums knows that accuracy does not equate to equality. We’re still going to have lots of disparity, with match results for the week that are obvious after the first 24 hours of play.

There are mechanisms available to you to fix this, but you obviously aren’t willing to change the system sufficiently to implement them.

Maybe Iron should consider transferring to a balanced matchup every week from now on.

Wolbryne.4037:

SoR has won tier1 NA for a few weeks in a row now. we’re the strongest server atm. then wham bam! we’re forced down to fight servers who don’t stand a chance. we had 3 waypointed garrisons when I logged on this morning and we’re currently ticking 500+. seriously anet. you messed up. BIG TIME. revert this nonsense and start actually listening to what the PLAYERS who PLAY wvw are actually saying needs done!

I agree, something should be done. We should have the #1 EU server fight the #1 NA server every week, that would give them good matchups.

wouldn’t fix anything LOOK At every tier there is almost a blowout in all of them. The matches before were stagnate yes but were more balanced atleast

wish I swear at times u live in a bubble

Whats a bubble?

Maybe a quick, short term fix would be to reset the score, but not the matchups every 24 hours. You can keep track of the totals internally throughout the week, but at least when people press “B” they won’t see a 50k point difference and say “kitten it, can’t do much to turn this match around”. A lot of the issue with this point system is psychological. I don’t really mind being out manned or losing the match, but any time you are getting roflstomped you lose numbers. All that does is make some of the WvW community apathetic which leads to less people in WvW, which leads to getting beat on worse, which leads to less people……. you get the picture. This kind of “let the ratings work themselves out” attitude almost killed HoD’s WvW population when we started our fall from T1. It took so long for the ratings to correct themselves that anyone who cared about points bailed on us. We built a great WvW community from the remnants (although much smaller) but that took a lot of time and effort that I’m not sure people will be willing to exert again if this continues for very long. Reset the score and maybe even the Borderlands every day or 2, leave EB so that the winning server isn’t totally robbed of their earned conquests. Demoralizing a server’s WvW population will only make the rating system fail more completely.

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in. The system we are using now creates varied matchups, which will allow us to get more accurate ratings. As those become more accurate we can dial down on the amount of variance we are introducing to have matchups that are varied, but less prone to large rating gaps between the servers. But, again, this is entirely based on server rating, which is the thing we have actually been calculating this entire time. Until that number is a more accurate reflection of server quality, we are going to keep mixing the matchups up.

Yes or no question. You look at the current match ups and think “this is how it should be?”

I tried to point out this very problem before and it was solved with a very simple solution: instead of making the random rolls between -1 and 1, make them between -0.5 and 0.5. I showed that it is a fact that it makes for less stale matches, but also prevents the matches with rank spreads of 6+. However, I was told by devs that this kind of thing was actually rare, and that my results won’t apply and were useless. But, so far, they seem to be fairly correct. I was told that things would balance out “eventually”, and that they would lower the range “if they had to”. So, they came into the system knowing it would produce results like this, and still did it with the statement “we’ll fix it later if we have to”. The way to fix this system is so simple, and it would drastically improve the results. They kept telling people ’it’ll stabilize’, but they never discussed how long it would take, or it seems put into consideration the morale of the servers that got bad matches.

So, it’s really simple to fix (cut the random range in half, look at my thread to see why!), but it probably won’t be, sadly. For those in rough matches, just try to stick it out and try to have some fun

The problem stems from the design of the ratings system itself, which was meant to rate individual players or teams in competitive sports, and does a good job at that role.

The assumption that this system will translate to an MMO is where the disconnect happens.

The skill level of an individual player, or a sports team which has a set roster, tend to only increase over time (except in cases of injury/retirement/etc). This means when a player/team gains rating in relation to another, it can be extrapolated that their skill has increased proportionately.

The “skill level” of an MMO server is dependent upon many more factors.
Professional sports are not affected by:
-massive guild transfers
-seasonal differences in participation due to hemisphere
-differences in participation throughout each 24 hour period due to concentration of players in specific time zones
-“fairweather/casual” players avoiding participation in losing matches
or any of the other factors involved in a 24/7 online game with thousands of players per “team”.

I’m not in favor of “one up, one down” either. I’d actually prefer if ANet had someone closely monitor the matches and manually create the new ones each week based on observed trends, rather than relying on overly complex math. Attempting to remove the human-error factor from matchmaking hasn’t resulted in error-free match creation so far…..

Math performs many functions well, and often it is the best language to use when attempting to define specific things, but it tends to fail at defining human behavior. Perhaps a psychology major would be better than a math major for the job of WvW matchmaker.

Offski.4897:

In a one up one down scenerio, Tarnished Coast could slowly build the forces they need by recruitment or other means to compete in a higher tier. Leaving things to a random generator will/has lead to exactly this, poorly balanced and boring match ups.

Stop suggesting an even worse system. OUOD system is even more ill thought out and inspired. It does nothing but consistently give you T1vs.T2vs.T3 and T2vs.T3vs.T4, and so on down the line, every single week, and strands the middle server in permanent limbo. The system would be instantly frozen because the T1 server that loses in T1 will crush T2 and hence bounce up back to T1, then the T2 server in T3 will crush T3, and immediately bounce up to T2, and because of the nature of the system there will be a T1 server in T2 and a T2 server in T3, etc, every single week. You would strangle the match-ups instantly.

Cactus.2710:

All of that is completely true, but it beats the crap out of me why you folks (ANet) can’t grasp the fact that the fundamental fault here is the WvW population imbalance between servers … period.

Doesn’t matter what they do, we went from imbalanced match ups with a server consistently winning by wide, wide margins to… the same thing with more shake-up in the fact that you don’t have to lose to the same people (or beat the same people) over and over with the possibility that (so long as you aren’t T1 matched downwards) you might lose or win for a change.

This horrible logic from ANet is determining that SoS is becoming a better server by getting totally dominated by higher pop servers. We hold what we have in EB and we get maybe a tower and a couple depots in our BL. That is making us better? Our rating is going to INCREASE 60pts because we cannot keep our own BL and sustain a push in EB? We are LOSING and we are going to climb the ladder.

This whole idea is horrible and a waste of time.

This system would work perfectly if the T1 servers weren’t so stacked with bandwagon players, because there’s really no other servers they can face except maybe the winner of T2. Wherever you go, unless it’s JQ / BG / SoR, you will ruin the fun for everyone around you including yourselves and your server.

I’m slightly amused by this.

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in. The system we are using now creates varied matchups, which will allow us to get more accurate ratings. As those become more accurate we can dial down on the amount of variance we are introducing to have matchups that are varied, but less prone to large rating gaps between the servers. But, again, this is entirely based on server rating, which is the thing we have actually been calculating this entire time. Until that number is a more accurate reflection of server quality, we are going to keep mixing the matchups up.

Do you know that most servers wont go all out so that “your new system” test each server akittens real strength?
Most players who are in a server that is facing a bigger server will just not go to WvWvW till the next week, same for the players in that bigger server.

Lets say you achieved what you want, what will we get after that system? the previous broken system?

Thrashbarg.9820:

The problem stems from the design of the ratings system itself, which was meant to rate individual players or teams in competitive sports, and does a good job at that role.

The assumption that this system will translate to an MMO is where the disconnect happens.

The skill level of an individual player, or a sports team which has a set roster, tend to only increase over time (except in cases of injury/retirement/etc). This means when a player/team gains rating in relation to another, it can be extrapolated that their skill has increased proportionately.

The “skill level” of an MMO server is dependent upon many more factors.
Professional sports are not affected by:
-massive guild transfers
-seasonal differences in participation due to hemisphere
-differences in participation throughout each 24 hour period due to concentration of players in specific time zones
-“fairweather/casual” players avoiding participation in losing matches
or any of the other factors involved in a 24/7 online game with thousands of players per “team”.

I’m not in favor of “one up, one down” either. I’d actually prefer if ANet had someone closely monitor the matches and manually create the new ones each week based on observed trends, rather than relying on overly complex math. Attempting to remove the human-error factor from matchmaking hasn’t resulted in error-free match creation so far…..

Math performs many functions well, and often it is the best language to use when attempting to define specific things, but it tends to fail at defining human behavior. Perhaps a psychology major would be better than a math major for the job of WvW matchmaker.

ya this is true. WvW isnt based on individual skill, where all individuals are equal. wvw is based on group coordination, reenforced by group numbers and coverage since its 24/7. Don’t get me wrong when I say that it doesn’t have to do with skill becasue in order to win everyone needs to have there act together and have the right builds and team work.

Sure you may have skill as a player, but individual skill only goes so far, as an individual cant carry a server.

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in. The system we are using now creates varied matchups, which will allow us to get more accurate ratings. As those become more accurate we can dial down on the amount of variance we are introducing to have matchups that are varied, but less prone to large rating gaps between the servers. But, again, this is entirely based on server rating, which is the thing we have actually been calculating this entire time. Until that number is a more accurate reflection of server quality, we are going to keep mixing the matchups up.

One up, One Down atleast creates more balanced matchups. While the servers can move up and down much faster this system would adjusts to changes in how well a server does rapidly. Instead of keeping servers clumped together, or creating worst matchups due to a random roll of the dice.
f.e. a rising Super Server could quickly climb up to a level it’ll be comfortable at.

And yes, this will create situations where a server will go up after a narrow win and get into a tier where itll get beaten. As a server that narrowly lost goes down a tier where it’ll win convincingly.
But this pairing would remain relative close to eachother. Unlike the current system which could pair servers several tiers appart. So how is that any different, or better?

And dialing down the variance just means moving closer back to the system we left behind. And then you’d still have the random element that could create seriously odd matchups.
Atleast OUOD would be a predictable system.

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in.

Devon, ratings are already and have been invalidated. I could care less if my server wins or loses at the end of the week.

What’s the point? “Congratulations! You’ve been promoted to Tier 1. Here you will experience skill lag and massive zergs! Good luck doing anything but smashing #1”

I don’t know about you but, normally when I win I receive something. If you want your precious ratings to be valid, reward the players with something other than “+5% Crafting exp!!!” …really?

Wrex.6798:

Wish, I don’t care about other EU guilds

This we already know at Desolation. No pun intended.

But back to the topic:
It is obvious that the old match up system nor the new one aren’t working. Just a quick glance at current matchups and I can see so huge point differences in just 1+ day of the match up that it is pointless to go on.
http://mos.millenium.org/eu

Just to mention few:

Fort Ranik has more than twice as many points compared to Whiteside Ridge and Dzagonur combined.

Arborstone has over 4.5 times more points compared to Fissure of Woe and Vabbi combined.

I would say that already 1:2 point difference between the winning and losing server in same tier is imbalanced match up, but the above examples are much worse and I feel real pity for such servers who are stuck in such matchups. Our matchup (Desolation) vs Baruch Bay and Elona Reach is borked. Desolation has been already for many weeks being constantly ganged by 2 other servers. 2 vs 1. And guess who was once again outmanned on our own borderlands in primetime?

While this might be okay for a week or two, but being on a server which has been losing for months upon months in a row. Arenanet you are just constantly “balancing” this game for the winning side.

Fix the useless outmanned “buff” NOW and don’t make it visible to enemy. It cannot be a difficult programming task to change it so that it is not visible to enemy as it just attracts both of them to double team the outmanned side.

End point is:

We are being farmed at our spawn.

The match ups blow for lower tier servers. This has been proven two weeks in a row. I think its obvious that this needs to change. How? Lotsa suggestions on the forums.

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in. The system we are using now creates varied matchups, which will allow us to get more accurate ratings. As those become more accurate we can dial down on the amount of variance we are introducing to have matchups that are varied, but less prone to large rating gaps between the servers. But, again, this is entirely based on server rating, which is the thing we have actually been calculating this entire time. Until that number is a more accurate reflection of server quality, we are going to keep mixing the matchups up.

I really appreciate that you replied to this thread Devon.

Here are my few points:
The ratings are meaningless when a server has been stuck in an imbalanced match up for months. E.g. look at Vabbi or Fissure of Woe. Do you seriously think more accurate ratings are gonna balance the game for them?
http://mos.millenium.org/eu/matchups/map/744

In most cases the winner and loser of the tier is already obvious after just 1 day of the weekly long match up. Thus the remaining 6 days of the tier don’t matter that much. This is just bad game design. In an optimal game the winner is determined at the very end. Now the point difference can be so big that it is already mathematically impossible to get enough point for a win even if you would control 100% of all maps and enemy doesn’t control anything (and such sudden domination from a losing server is simply not realistic).

When a server is being dominated by two stronger ones (or one very strong one), it creates a vicious downward cycle. People have less and less motivation. If you want to experiment with more volatile system, at least make the match ups just 24 hour long for that period of time. To keep the match up 1 week long even though it is plain dead obvious many match ups (and in fact half of the EU match ups) are very far from being balanced, is just making many people feel extremely demotivated about WvWvW.

The key issue is coverage. A server with the best coverage wins. Simple as that. And the recent totally overpowered changes to ACs just alleviate this problem, giving even more power to the winning side (with better coverage), like I have explained in many threads.

If you want to balance the matchups, start looking at the coverage numbers e.g. how many players are actually playing WvWvW (not AFK) at each possible time tick, 24h7 total tally. Ideally a server should be matched with a one which has a similar coverage.

And you can really contact me or some other WvWvW players, as we got ideas how the balancing could work. I don’t claim I have solution for every problem e.g. at EU Visunah Square at least used to have superior coverage with no other server being close to theirs, thus they have no ideal coverage match.

LOL this new system is actually pretty great, Us mid lows get some pretty good random matchups vs the month long battles with same 2 servers.

To be straight up honest I hope the “super servers” get paired up with the lower severs for months on end. My chuckles would have no end then. LOL people being mad they have nothing to do for the rest of the week after they painted wvw their color an hour after reset. Isn’t that the reason you mostly likely transferred to that server?

My 2 cents as well, the ratings are absolutely meaningless if one server is getting completely steamrolled over.

You can’t expect the server to get excited they nabbed a high ranking if they can’t even make it into their own garrison. On the flip side, you can’t expect the dominant server to be remotely excited when there is no challenge present. This is even heightened by the fact if people give up and WvW is a baron waste land, and there is nothing to do for those who remain on the map.

Anet; while I completely appreciate the effort in trying to switch things up at the players request, this is clearly a case of “be careful what you wish for”. We wished it, we got it, and this system is much worse.

It’s not a case of rankings being the problem, coverage is the problem as others have pointed out. Now there is a couple of ways to address it, 1 believe would be temporary, the other one would need Anet to address an even bigger elephant in the room first.. skill lag.

The temporary solution: and I emphasize the word temporary; turn off the PPT during off hours. Have it active during normal hours, but turned off when realistically people in those time zones should be sleeping. Take a specific time zone in NA, and use central as your guide, institute a “blackout” so to speak between the hours of 11pm – 9am. During those hours the PPT does not tick, but every thing else in WvW remains status quo. Everything else can be done in WvW, the only difference is the PPT is turned off. Do the same in Europe. Offer free transfers for those who operate during SEA time if they don’t like it, so they can go to a server when PPT is active. Again, this is temporary until they find a permanent fix.

A permanent fix I think would work (this is assuming best case scenario in that they fix skill lag), would be to abolish the “servers joining WvW” and have players join 3 separate factions instead. There are already 3 factions in the game… make use of them, have players join one of those. There is also 3 borderlands.. how convenient. This would obviously make maps a lot larger and have far more people on them. Assuming skill lag is taken care of, I think this would both solve the coverage problem and make WvW more fun overall. If this can be achieved, then turn the PPT to run 24/7.

Thoughts?

Ok they are going to decrease the size of variation sometime in next few weeks once ratings adjust to more realistic values representing relative strengths of servers.

I would favour them dialling it back at least a little right now. Looking at current ratings I would try and limit to 200 rating points either way (so SoS could play anyone between CD and TC, effectively one tier up or down if you were looking it that way)

I think scoring as well needs to be considered if they are having these types of matchups – did I read somewhere that they were introducing a system where you gained more points by TAKING an objective rather than HOLDING an objective, which would mean that, for example, SoR wouldn’t accrue massive points just by sitting on it’s upgraded garrisons (perhaps receive 50 points for taking a garrison and 10 ticks for holding).

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in. The system we are using now creates varied matchups, which will allow us to get more accurate ratings. As those become more accurate we can dial down on the amount of variance we are introducing to have matchups that are varied, but less prone to large rating gaps between the servers. But, again, this is entirely based on server rating, which is the thing we have actually been calculating this entire time. Until that number is a more accurate reflection of server quality, we are going to keep mixing the matchups up.

Some variation is necessary to ensure that the overall server ratings remain accurate, but right now it is too varied. What purpose is there in sending the top ranked server into tier 2, for example? We already know that server #1 beats server #2 and #3 in tier 1. It goes without saying that they should be even more capable against tier 2 servers.

The fundamental flaw here is the assumption that the server ratings are not accurate or at least semi-accurate. Obviously, they are accurate within each tier. The only time they are not accurate is possibly at the fringe of each tier, e.g. the top server in tier 2 may actually be better than the lowest server in tier 1 but because of stagnant ratings/matchups it takes forever for movement between tiers.

Having a system where winner moves up and loser moves down where activation chance is based on rating seems like it would generate much more meaningful data and matchups.

Vena.8436:

Offski.4897:

In a one up one down scenerio, Tarnished Coast could slowly build the forces they need by recruitment or other means to compete in a higher tier. Leaving things to a random generator will/has lead to exactly this, poorly balanced and boring match ups.

Stop suggesting an even worse system. OUOD system is even more ill thought out and inspired. It does nothing but consistently give you T1vs.T2vs.T3 and T2vs.T3vs.T4, and so on down the line, every single week, and strands the middle server in permanent limbo. The system would be instantly frozen because the T1 server that loses in T1 will crush T2 and hence bounce up back to T1, then the T2 server in T3 will crush T3, and immediately bounce up to T2, and because of the nature of the system there will be a T1 server in T2 and a T2 server in T3, etc, every single week. You would strangle the match-ups instantly.

Cactus.2710:

All of that is completely true, but it beats the crap out of me why you folks (ANet) can’t grasp the fact that the fundamental fault here is the WvW population imbalance between servers … period.

Doesn’t matter what they do, we went from imbalanced match ups with a server consistently winning by wide, wide margins to… the same thing with more shake-up in the fact that you don’t have to lose to the same people (or beat the same people) over and over with the possibility that (so long as you aren’t T1 matched downwards) you might lose or win for a change.

I dont get your logic as there is none

You’re calculations for server ranking is not taking WvW populations into account clearly its should.

Do a better job, you seem like a bunch of shoot from the hip ‘concept’ clowns that cant do any heavy lifting.

p.s. Yes I code, but in an industry that does not allow so many…mistakes.

The problem is that the system is INHERENTLY bugged this way. It’s a BROKEN system.

People have done nothing but complain about your excessive “random roll” junk in this game since day 1. So…when faced with a REAL problem, what did you do? You tried to solve it with more RNG.

People did not want this. READ: People. Did. Not. Want. This.

Let me say that again:
People
Did
Not
Want
This

NOW, lets talk about what people wanted:

People wanted something very simple.

Extremely simple, actually, and something that you could have accomplished very simply by tweaking the way Glicko works.

It would have made a lot more sense and worked a lot better than what you have now.

People wanted you to make it easier for tiers that were unfairly stacked to be evened out. That. Is. All.

All you had to do to accomplish this was implement a system where traversing to the next tier, or previous tier was “compressed.” Before, you had to absolutely destroy the previous tier to move up…COMBINED WITH…a server on the previous tier completely tanking. That was a very hard condition to meet. All. you. had. to. do. was. tweak. glicko. so. it. was. easier.

Bandwagoning is really what has killed the entire system. Tiers were never meant to be based on skill. They were supposed to reflect server population. Less populated servers fight less populated and large servers fight larger servers.

Its this nonsense that assumes the tiers are some sort of skill ranking is what really is the problem. Being tier 1, 2 or 3 does not mean you are awesome at WvW. It means you have a ton of people who play WvW at all hours.

Aeros.2046:

Bandwagoning is really what has killed the entire system. Tiers were never meant to be based on skill. They were supposed to reflect server population. Less populated servers fight less populated and large servers fight larger servers.

Its this nonsense that assumes the tiers are some sort of skill ranking is what really is the problem. Being tier 1, 2 or 3 does not mean you are awesome at WvW. It means you have a ton of people who play WvW at all hours.

There is a clear and distinct difference between a T3 skill level and a T2 skill level. It’s like night and day, when you put a new T3 server against experienced T2 servers.

Yes, population has a ton to do with winning, obviously, but 30 people can wipe 60 people if they are “skilled” about how they do it.

Hickeroar.9734:

Aeros.2046:

Bandwagoning is really what has killed the entire system. Tiers were never meant to be based on skill. They were supposed to reflect server population. Less populated servers fight less populated and large servers fight larger servers.

Its this nonsense that assumes the tiers are some sort of skill ranking is what really is the problem. Being tier 1, 2 or 3 does not mean you are awesome at WvW. It means you have a ton of people who play WvW at all hours.

There is a clear and distinct difference between a T3 skill level and a T2 skill level. It’s like night and day, when you put a new T3 server against experienced T2 servers.

Yes, population has a ton to do with winning, obviously, but 30 people can wipe 60 people if they are “skilled” about how they do it.

oh yeah i saw that difference. I guested from DH ( T6) to JQ (T1) to visit some gw1 buddies. A member called out taco up so i decided to dropin. 30 people died on his first breath, leaving me standing there saying " lol what?"……

Oh wait my bad thats pve, Taco was pulling some 360 no scope “dont touch my last tater tot” haxasaurus moves on you. Im sure if he was romaing around dodging your attacksit would have been prettier instead of being all static and telegraphing his attacks 5 secs earlier then when they hit. We should get him nerfed……

\o/ SKILL !!!!!!!

random roll isn´t bugged, it´s the bug itself.

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in. The system we are using now creates varied matchups, which will allow us to get more accurate ratings. As those become more accurate we can dial down on the amount of variance we are introducing to have matchups that are varied, but less prone to large rating gaps between the servers. But, again, this is entirely based on server rating, which is the thing we have actually been calculating this entire time. Until that number is a more accurate reflection of server quality, we are going to keep mixing the matchups up.

If you guys want any kind of longevity to WvW match ups and player migration need to be improved. I’d go so far as creating new servers that are WvW specific and allowing players to transfer there for free (based on number of hours spent in WvW) placement would be based on a mix of skill level and server population. The number of blow blowouts week in and week out is a little ridiculous, and you have to see that in many cases it is the same servers getting blown out.

DevonCarver.5370:

One up and one down solves some problems but creates a bunch more. Including that it invalidates ratings entirely. It no longer takes into account quality of victory but simply uses the fact of victory to move a server out of the matchup they were previously in. The system we are using now creates varied matchups, which will allow us to get more accurate ratings. As those become more accurate we can dial down on the amount of variance we are introducing to have matchups that are varied, but less prone to large rating gaps between the servers. But, again, this is entirely based on server rating, which is the thing we have actually been calculating this entire time. Until that number is a more accurate reflection of server quality, we are going to keep mixing the matchups up.

I think the bigger problem your now facing with this grande failure of yours is the fact : How to keep people to stay in wvw, the losing servers are completely giving up, way to go, you ledt lower tier servers to a complete and utter slaughter.
Or was this all a great plan to get people to go into the failure called sPvP?

I just wanted top add my tuppence worth.

Firstly, I really do appreciate the intention to spice up the match ups. I know that at FA, we’d been facing TC/Dragonband/Kaineng for months on end, and variety is good for all involved. I also have a vague awareness of other servers routinely romping home, but not being able to go up, so again, kudos for making a change.

I do feel however, that this change has simply come too late. The previous match up system was so locked in place, that many guilds (and players) gravitated towards the top 2 tiers. This may have been so they could join up with other equally committed players. In my case, living in Asia, I had to join a US server with an Oceanic/SEA presence in WVW, meaning a top 2 tier server.

In allowing players to transfer for free for such a long time, and allowing matches to get so stale, a lot of the WVW playerbase really stacked up the top tiered servers (and the same on a smaller scale the lower you go down the tiers). As a result, many matches with two servers outside of their original tier is grossly one-sided; those servers had been preparing themselves for months for the coverage wars that categorized the old tiered system.

What keeps me playing wvw is logging in and seeing that your server needs help somewhere, and that there’s an awesome battle to join and contribute too. For far too many people, this simply isn’t the case at the moment. When I log in at SEA time zone, I see 400+ PPT, a complete lack of urgency, and my server mates desperately scratching around for a good battle. I imagine it’s even worse for our opponents.

As you’re posting here Devon, I’m sure you’ve seen the many suggestions made across the forum, particularly the one up/one down suggestion. While, as you say, it will bring it’s own problems, but I’d like to ask (the community too), to what extent is the quality of a victory now a contributing factor to forming enjoyable match ups for players? Unfortunately, as many posters here have said, the system as it stands isn’t providing the experience that will keep people playing, and I can’t help but think the answer isn’t to let the playerbase ‘go on vacation’ while the ratings sort themselves out.

The solution I’ll leave to people who may understand the system and maths behind it more, but I’d just like to urge the development team, who’ve made such an enjoyable game, to please act as a matter of urgency on re-evaluating these match ups so that the whole playerbase can log in and enjoy the awesome battles you only get in GW2.

(Disclaimer; I realize a lot of my statements about top 2 tiers are generalizations, and don’t mean to say the players from those servers are necessarily better; some players in our Emhry Bay match ups have brought some awesome fights this week, and I’m sure there are other examples throughout the servers. As a rule though, these lower servers can’t compete with the timezone coverage of tiers higher up.)