Jump to content
  • Sign Up

World Restructuring Status Update


Recommended Posts

I'm excited to hear you are addressing time of day considerations. I thought I'd read in the past that was not being considered or was a low priority. For my play experience WR has been great, for w/e one data point is worth ; p

  • Like 2
  • Thanks 1
  • Haha 1
  • Confused 7
Link to comment
Share on other sites

Quote

most teams are showing reasonably similar performance in terms of player hours, average war scores, and victory points.

Hmmmmmm....... are they though?

NA TI - 30 vp gap

NA T2 - 30 vp gap

NA T3 - 68 vp gap

NA T4 - 26 vp gap

NA T5 - 36 vp gap

------------------------------------

EU T1 - 15 vp gap

EU T2 - 40 vp gap

EU T3 - 12 vp gap

EU T4 - 103 vp gap

EU T5 - 59 vp gap

EU T6 - 67 vp gap

EU is crazy out of whack.

Also I'm more interested to know what the average playtime levels have been for the first 3 weeks of WR, week by week, and compared to before WR. And no, not the kdr. But I know that will never be revealed. Overall, meh, feels the same as before. 🤷‍♂️

  • Like 31
  • Thanks 1
  • Haha 4
  • Confused 5
Link to comment
Share on other sites

You guys promoted alliances, so people went and created alliances in preparation only for you to dump the project and then not pivot this project to compensate for what you knew players had done in terms of creating and sticking to said alliances (evident in the extra guild slot we were provided)? Fairly hefty oversight on a system specifically designed to "matchmake".

I find it hard to believe you cannot pull enough time zone data from players to make this system better as is. Instead you suggest anyone but NA/EU core splits off from their alliances they may have been with for over 2 years just to be allowed to play the game instead of being jammed onto packed servers? And for a seemingly not-short timeframe based on your wording. I don't know why you're trying to kill the game for a portion of your players. On NA servers, you quite literally don't even have enough guilds in either OCX or SEA to fill every server with one and yet you have servers with more than one guild in these time zones somehow.

I know it's the first match up being in this permanent framework, but this has been years in the making and time zone balance concerns have been brought up and discussed more than once with devs that this update just feels weak and empty on that front. And then you compare VP and KDR between servers as an additional note of imbalance.

  • Like 22
  • Haha 1
  • Confused 9
  • Sad 1
Link to comment
Share on other sites

Posted (edited)
23 minutes ago, amberrosalie.5381 said:

I find it hard to believe you cannot pull enough time zone data from players to make this system better as is. Instead you suggest anyone but NA/EU core splits off from their alliances they may have been with for over 2 years just to be allowed to play the game instead of being jammed onto packed servers? And for a seemingly not-short timeframe based on your wording. I don't know why you're trying to kill the game for a portion of your players. On NA servers, you quite literally don't even have enough guilds in either OCX or SEA to fill every server with one and yet you have servers with more than one guild in these time zones somehow.

It's due to these guilds being in the same alliance. The guilds (by this I mean a super guild that has a couple of guilds inside it) overstacked some servers in OCX/SEA, not Anet. This is why there is one server with a blob at OCX/SEA, and an opposing server that is a ghost-town. You can't blame Anet for something the players have done with their guild alliances. I pointed out this problem would occur months ago, and it's been treated as a non-issue. Well, here we are.

Edited by Hesione.9412
clarifying what I mean by guilds.
  • Like 11
  • Thanks 2
  • Haha 2
  • Confused 3
Link to comment
Share on other sites

16 minutes ago, amberrosalie.5381 said:

You guys promoted alliances, so people went and created alliances in preparation only for you to dump the project and then not pivot this project to compensate for what you knew players had done in terms of creating and sticking to said alliances (evident in the extra guild slot we were provided)? Fairly hefty oversight on a system specifically designed to "matchmake".

Can someone explain how having this sixth guild slot that is capped at 500 players is any different from having alliance that is also capped at 500 players? I feel there is a huge misunderstanding from this post.

 

  • Like 7
  • Thanks 1
  • Confused 2
Link to comment
Share on other sites

Just now, A Hamster.2580 said:

Can someone explain how having this sixth guild slot that is capped at 500 players is any different from having alliance that is also capped at 500 players? I feel there is a huge misunderstanding from this post.

 

The "super guild" probably is restricted to currently active players. Larger guilds may have players who haven't logged in for months, or who don't play WvW. Those players are excluded from the "super guild". In effect, the "super guild" is the same as an alliance.

  • Like 4
  • Thanks 1
  • Confused 2
Link to comment
Share on other sites

1 minute ago, Hesione.9412 said:

It's due to these guilds being in the same alliance. The guilds (by this I mean a super guild that has a couple of guilds inside it) overstacked some servers in OCX/SEA, not Anet. This is why there is one server with a blob at OCX/SEA, and an opposing server that is a ghost-town. You can't blame Anet for something the players have done with their guild alliances. I pointed out this problem would occur months ago, and it's been treated as a non-issue. Well, here we are.

I am very specifically talking about guilds that did not stack themselves but have been matched together and yes they exist. I'm aware of the alliances that have stacked themselves in certain time zones and I have my opinions on that as well (not kind ones), but there are a numbers of guilds who do not share alliances that are together that have been screwed by the system.

I am also among the people who have pointed out this issue, so this is not new to me either. What is frustrating is that Anet have done very little to combat it initially or now and have made comments in both directions of it being lowest and highest priority over the project's public lifetime.

  • Like 16
  • Confused 1
Link to comment
Share on other sites

6 minutes ago, A Hamster.2580 said:

Can someone explain how having this sixth guild slot that is capped at 500 players is any different from having alliance that is also capped at 500 players? I feel there is a huge misunderstanding from this post.

 

I don't think there really is a difference now - just an extra slot because people already complained about needing the extra slot throughout the beta. Alliances was supposed to have a completely different UI set up based on the initial information we got, I believe. Like guilds within a guild? I just remember there being something said about not having the UI sorted for some of the earlier betas, but not specifically what.

  • Like 4
Link to comment
Share on other sites

14 minutes ago, A Hamster.2580 said:

Can someone explain how having this sixth guild slot that is capped at 500 players is any different from having alliance that is also capped at 500 players? I feel there is a huge misunderstanding from this post.

 

We dont have Alliances, we have World Restructuring. We have actually never had an Alliance beta ever. Only WR betas that's guild based.

At some point they realised they were kind of trying to reinvent the wheel using spaghetti code (or something similar), and landed on just using guilds to get the WR up and going. Giving us a sixth guild slot to use in the same way the playerbase had already done to "fake" the not yet implemented Alliance system already from the first failed beta:; using umbrella guilds.

  • Like 8
  • Thanks 1
Link to comment
Share on other sites

Think of a database. Think of a table with a variable in there, for the sake of clarity, we'll just call "WvW alliance". Every WvW alliance sits in that variable in that table. Now you, as a player, have multiple tables/links to multiple tables in this database. There is your record in the table of GW2 accounts. There is a table with all your characters. There is a table with your material and bank storage, etc. You are linked to a table containing guilds, with your account name linked to each guild of which you are a member.

To implement alliances, Anet could link each guild in an alliance, to that "WvW alliance" variable. You are linked to the alliance through your link to your guild, and then your guild link to the alliance. Instead of that "WvW alliance" table/variable, Anet created a sixth guild slot. Instead of the link being your account, to your guild, to your alliance, the link is from you to your WvW "guild".

If people in guilds A, B, and C were in a new guild, D, and selected D as their wvw guild, guild D contains people from multiple guilds. The link is your account -> guild D. In the case of an alliance, the link would be from, say guild A, to alliance D, and the link is your account -> guild A -> alliance D.

The key point is the links between the database tables. Everything is links between database tables.

  • Like 1
  • Confused 7
Link to comment
Share on other sites

Posted (edited)

I will say that the alliance structure as I understand we were promised would absolutely be terrible.  Let's not pretend that the sweatiest, most-invested guilds in the game wouldn't collude to form into a super alliance of 500 people covering all times of the day and constantly sit at Tier 1 just utterly stomping everyone unlucky enough to get put against them.

Edited by Maekrix Waere.2087
  • Like 9
  • Confused 1
Link to comment
Share on other sites

Posted (edited)
5 hours ago, Cecil.4536 said:

The team-creation algorithm is performing as expected, and most teams are showing reasonably similar performance in terms of player hours, average war scores, and victory points. 

Right now im on T4 vs Mirror of Lyssa (red) and they are simply dominating the bracket. They are running around with 4 to 5 zergs every evening while blue and green have around barley one pub zerg. While the match ups before were rather close i never before experienced a more unbalanced match up, even before beta. I dont know how this a "better" experience than pre beta but please look into this match up and impede match ups like these in future encounters. Getting zoned to your starting point on ebg the whole week by different guild groups really is a bad joke.

Edited by Bale.3851
  • Like 17
Link to comment
Share on other sites

Currently t1 NA is a joke. 

Three of my accounts ended up in t1 on separate sides.

One server has poor na coverage and was very obviously backpacked to t1 by sea and ocx.

Another server with already active guilds was linked with what I'm guessing was dead guilds that are now very active leading to 247 coverage, and are gonna be t1 forever.

I have no idea how yall gonna fix it, but it sure is a mess.

  • Like 7
Link to comment
Share on other sites

43 minutes ago, C Cspace Cowboy.5903 said:

Currently t1 NA is a joke. 

Three of my accounts ended up in t1 on separate sides.

One server has poor na coverage and was very obviously backpacked to t1 by sea and ocx.

Another server with already active guilds was linked with what I'm guessing was dead guilds that are now very active leading to 247 coverage, and are gonna be t1 forever.

I have no idea how yall gonna fix it, but it sure is a mess.

T1 NA:
0.98 vs 0.81 vs 1.15 KDR (tied for the most balanced kdr with T3)
27k vs 25k vs 30k kills (the most balanced in terms of kills)
270 vs 240 vs 234 points (basicly looking like every matchup at this point in the week)

What a joke!

  • Like 4
  • Thanks 3
  • Confused 10
Link to comment
Share on other sites

4 hours ago, XenesisII.1540 said:

Hmmmmmm....... are they though?

NA TI - 30 vp gap

NA T2 - 30 vp gap

NA T3 - 68 vp gap

NA T4 - 26 vp gap

NA T5 - 36 vp gap

------------------------------------

EU T1 - 15 vp gap

EU T2 - 40 vp gap

EU T3 - 12 vp gap

EU T4 - 103 vp gap

EU T5 - 59 vp gap

EU T6 - 67 vp gap

EU is crazy out of whack.

Also I'm more interested to know what the average playtime levels have been for the first 3 weeks of WR, week by week, and compared to before WR. And no, not the kdr. But I know that will never be revealed. Overall, meh, feels the same as before. 🤷‍♂️

Well, honestly I don't find just counting VPs is enough to vizualize balance within MUs, since you don't really take into account the maximum gap.

From the reset until your comment, there has been 65 skirmirches. That means a 130 maximum gap that would occur if a team completly obliterates a MU and another team gets completly oblitareted. 

 

So

NA TI - 30 vp gap => 23%

NA T2 - 30 vp gap => 23%

NA T3 - 68 vp gap => 52%

NA T4 - 26 vp gap => 20%

NA T5 - 36 vp gap => 27.7%

------------------------------------

EU T1 - 15 vp gap => 11.5%

EU T2 - 40 vp gap => 30.8%

EU T3 - 12 vp gap => 9.2%

EU T4 - 103 vp gap => 79.2%

EU T5 - 59 vp gap => 45%

EU T6 - 67 vp gap => 51.5%

 

Percentage represents the numbers of skirmishes that saw the same server winning and the same server losing. To me, it looks like a big problem on EU (especially the lower tiers), not so much on NA. 

 

The fact that lower tiers are impacted shows some information

  • random repartition after relink results in several weeks of bad MU in the lower tiers
  • there was a team in EU that was unluckily placed in T6(=Mirror of Lyssa) after the relink, and has to climb all the way to the upper tiers to find teams of the same caliber. 
  • after looking at gw2 mists, it seems that this MU is not unbalanced activity-wise. This is an approximation, since it only uses data from players that gave an api key to mists
  • we might want to check MU after MU that "mid tier" result, to see if it is subject to higher gaps generally, or if it is just a random thing we see this MU being the most unbalanced on both NA and EU
  • it seems the "sorting" of the teams happens faster in the higher tiers than the lower

 

Honestly, we could go even further by considering that if you want "balanced teams", there are choices to be made, and it does not look that easy. From what we saw, even if activity-wise player count is the same, it might eventually becomes that reason there's an inbalance between teams.

 

Example, with that infamous EU T4

yesterday, GW2 mists registers activity peak at 20:00-22:00 - which is normal, for this is the time registered players on mists are more likely to be playing.

What we see in terms of "player activity" is that the greens were ~12% higher than the reds in activity % (around 1.3% difference) and the blues ~8% lower (around .7% difference). 

So, either a balanced situation or the greens have a slight advantage. 

And what did it translate as ?

  • highest k/d of the day for the reds
  • highest number of red kills of the MU
  • reds went below blue as the server with the lowest deaths on the MU 
  • greens didn't even see an inflexion point of the number of kills during whole 20-22 time period

Does that mean that Mirror of Lyssa is over-loaded in terms of players ? Not according to mists, at least. It could mean several things, however :

  • MoL is too strong, and shouldn't have been built that way. We will only know this when the server will join the higher tiers
  • the gap of the level between guilds is that wide and we can only fit "good guilds' in 9 teams before running out of them and having to do "teams that are doomed to lose". That is, assuming they measure accurately guild level with their metrics. They could - but do they ?
  • Like 5
  • Thanks 1
Link to comment
Share on other sites

13 minutes ago, Dawdler.8521 said:

T1 NA:
0.98 vs 0.81 vs 1.15 KDR (tied for the most balanced kdr with T3)
27k vs 25k vs 30k kills (the most balanced in terms of kills)
270 vs 240 vs 234 points (basicly looking like every matchup at this point in the week)

What a joke!

And yet the team with the highest kd is losing the matchup. In every other matchup except t1, the team with the highest kd is currently winning their respective matchup.

270

234

240

10619
7571
8750

  • Like 1
  • Confused 3
Link to comment
Share on other sites

22 minutes ago, C Cspace Cowboy.5903 said:

And yet the team with the highest kd is losing the matchup. In every other matchup except t1, the team with the highest kd is currently winning their respective matchup.

270

234

240

10619
7571
8750

... so you're saying that the other teams in the matchup is able to balance out their less effective killings with better coverage? 

Hm.

Should I assume you are on that loosing team? I mean it's clearly a joke if you're not winning. You should win. It's only fair.

  • Like 6
  • Haha 2
  • Confused 3
Link to comment
Share on other sites

Posted (edited)
31 minutes ago, C Cspace Cowboy.5903 said:

And yet the team with the highest kd is losing the matchup. In every other matchup except t1, the team with the highest kd is currently winning their respective matchup.

It's almost like a game mode whose victory condition is based on controlling territory and not kills doesn't actually care which team has the most kills if those kills aren't converting into territorial control.

Edited by Maekrix Waere.2087
  • Like 3
  • Thanks 1
  • Haha 1
  • Confused 1
Link to comment
Share on other sites

32 minutes ago, Dawdler.8521 said:

... so you're saying that the other teams in the matchup is able to balance out their less effective killings with better coverage? 

Hm.

Should I assume you are on that loosing team? I mean it's clearly a joke if you're not winning. You should win. It's only fair.


mate the numbers always beat 'skill' in wvw, that's literally the reason we have WR — some servers are way more emptier than other and so they get the short end of the stick. My server (Deso) *always* had the highest KDA in the matchup and yet we rarely won and rarely even got links because the server is one of the first and it's always 'full', so the 'algorithm' assumed we have enough people, and we could be casually put as solo server against 2+2 enemies.

It's not balance no matter how you put it. Small deviations like 48-52% winrate is expected in every game, but in WvW much higher KDA just means much higher population. No matter how 'skilled' you think you are, when every map is full for the enemy and you can barely scrape enough people for one squad on one map, you'll get destroyed. And since you think so high of yourself I assume you're from a winning team.

  • Like 3
  • Confused 3
Link to comment
Share on other sites

Posted (edited)
1 hour ago, latlat.4516 said:

Example, with that infamous EU T4 : <snip analysis>

So, either a balanced situation or the greens have a slight advantage. 

And what did it translate as ?

  • highest k/d of the day for the reds
  • highest number of red kills of the MU
  • reds went below blue as the server with the lowest deaths on the MU 
  • greens didn't even see an inflexion point of the number of kills during whole 20-22 time period

Does that mean that Mirror of Lyssa is over-loaded in terms of players ? Not according to mists, at least. It could mean several things, however :

  • MoL is too strong, and shouldn't have been built that way. We will only know this when the server will join the higher tiers

We're playing in that matchup at the moment. While I've not had a chance to get on during the evenings yet (so can't comment about that), the mid-morning has been rather hellish.

A slightly larger red group that jumps map and follows us around for the kills (we jumped to EB to take our keep back, they jumped and ran across the map from their keep to defend a T0 - you could see the long tail from the map load times). We put up a good fight where we could, but the only way to make progress was to completely change tactics. So a high k/d for red who earlier in the week were also dominating the PPT scoring as well. Red strategy now seems to be pushing greens and blues together to create a fight over who goes down.

I suspect your analysis is little off because you're not taking into account the history over the week. You can't get an undestanding of a matchup by analysing a single 2 hour skirmish: attendance in WvW is not the same as forming a squad, or even a boon-ball meta-squad led by a fight commander. Although you can highlight the apparent contradiction in that one session, as you did.

My hypothesis is that several GvG guilds have banded together to form a guild alliance, and that this is an edge case that the matching algorithm can't easily deal with. MoL seems to be the example. I'm left wondering when the realisation will come that these alliance guilds can really only fight each other in guild halls now, rather than at ye olde fight spots like Champions Demesne.

Edited by Charybdis.9042
  • Like 5
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...