Guild Wars 2

OK so YB and JQ are now full TC and Maguuma are not. My question to anet is how can you justify this? JQ have lost 1 major guild (and gained another) but our numbers are not that great this week and last, not even in SEA, and Maguuma has been pushing BG in T1. It is especially ridiculous with JQ being without a linked server.

WvW Server open reset every Monday.

Population calculation is not a good idea for Server Open

Need a loop for server open

i dont think YB or JQ strong let server full o.O

Blackgate already two months still full

your english….i don’t understand…….

It takes some time for the status to be updated, however, do take note mag has better quality than most servers.

As per the devs when the wvw population changes were first rolled out:

“It’s not looking at a short period of time, and it’s looking at many aspects of the ways players are interacting with WvW to determine a server population.”

One week is a short period of time, I think the “recalculation every monday” is all in our heads.

morrolan.9608:

OK so YB and JQ are now full TC and Maguuma are not. My question to anet is how can you justify this?

Players are more likely to leave lower tiers to go to higher tiers.

With Tier 1 open and JQ full, there are likely more players leaving JQ to tier 1 servers.

This is a business after all.

Moreover Tier 1 servers have deeper pockets that can sponsor transfers and have sponsored and are more likely to sponsor transfers.

And this is good business as well.

Jerry CCH.9816:

WvW Server open reset every Monday.

Monday is a perfect time to open servers for players to start thinking about transfer.

By Monday the winning and losing servers are already transparent,

but the match does not end on Monday.

It goes on 5 more days because the extra days can increase the pressure for transfer for players who want to win but actually experiencing the stress of losing.

Lord of Rings.5371:

morrolan.9608:

OK so YB and JQ are now full TC and Maguuma are not. My question to anet is how can you justify this?

Players are more likely to leave lower tiers to go to higher tiers.

With Tier 1 open and JQ full, there are likely more players leaving JQ to tier 1 servers.

This is a business after all.

I don’t know anybody on JQ who’d want to go to the T1 servers. It’s quite literally a blob fest, with T1 terrified of running anything less than their entire map queue to take things. And in true enough fashion…you catch them without the entire map queue, they all fold really easy to a group of 30.

I’m sure it’s based off numerous factors and its most certainly is based off a longer evaluation time period than you describe.

The communities of the game move too sporadically given the game mechanics and cultures. It can’t be balanced by use of a formula over time in such a manner. I believe this is one of the issues Arena Net will face when trying to balance WvW with the current mechanics.

I would also like to point out I believe that your actual rank and/or your tier and not if you’ve won a match effects the formula that determines your status. (Can’t confirm) Just been watching consistently.

Based off that, I can use YB as an example as to why their locked.

YB is locked because even though YB is currently winning this match up by a land slide, They are rank 7 and this match-up isn’t decided so it’s not taken into account. Which means, their normalized spot is in T3. Yet they had a surge in population base and because of this, they are standing up to and crushing Tier 2.

DB has recently lost a large chunk of their player base and coverage(Over many weeks). They are dropping like a rock, yet they are STILL ranked 4 overall. Which means currently in Tier 2 there is a huge imbalance. JQ is rank 5 and beating rank 4, yet being creamed by rank 7. (ArenaNet should have reset the volatility)

If you look at JQ: last week was the first week of skirmishes, JQ gained +18 and the fact that match making is calculated at the end of the match up when Jade Quarry was rank 4 and Yaks Bend wasn’t in the tier dominating

Considering my above analysis, Yaks Bend is dominating and is imbalanced purely based off speculation that a servers actual rank is a factor. Thus, causing them to be temporarily locked. I say temporary, because their rank will even out. Predictions show that not only will YB take 6th but their closing in on 5th as well. Once this happens, and whatever reevaluation period comes next will indeed effect Yaks Bend based off their coverage over time.

*Take into account that the formula for rating was recently modified when the skirmish system got added. This is also having a effect on server placement and ranks. Thus, it’s even effecting the server status upon reevaluation that just happened. *

“Regarding how Glicko will be updated at the end of a skirmish-based match. The same inputs and outputs will be used as usual, except that total Victory Points from the week’s match will be fed in as the score for each team, instead of the week’s total War Score.
This can have different effects on how ratings will change. If team A consistently beats team B— but only a little— their ratings will be split apart faster than before. If team A steamrolls team B consistently, their ratings won’t split apart as quickly as before.” -ArenaNet Chris Barrett

Also, notice that last week MAG blew out JQ and remember “If team A steamrolls team B consistently, their ratings won’t split apart as quickly as before.” You still gained and JQ was recently open. TC was initially locked.

All the status (T1|T2) updates that just happened are predictable IMO more so than not. Just the same as the link reevaluations every 2 months.

Here is the NA server activity since the linking.

I’m not sure if Northern Shiverpeaks could actually compete in Tier 1 but its activity is comparable.

Dragonbrand’s activity is taking a dive.

TorquedSoul.8097:

Here is the NA server activity since the linking.

I’m not sure if Northern Shiverpeaks could actually compete in Tier 1 but its activity is comparable.

Dragonbrand’s activity is taking a dive.

What do the numbers on the Y-axis represent? Is this something that is more appropriately used to compare only a server’s activity against itself over time rather than an indicator of who might win a matchup and what tier they should be in?

TorquedSoul.8097:

Here is the NA server activity since the linking.

I’m not sure if Northern Shiverpeaks could actually compete in Tier 1 but its activity is comparable.

Dragonbrand’s activity is taking a dive.

And JQs is at T3 level so why is it closed? If population is based off the last month during that whole time the JQ population has been less than Maguuma and now YBs. YB is full but Maguuma isn’t, there is an inconsistency here. Moreover YB’s population is not at Maguuma’s size.

Lord of Rings.5371:

Players are more likely to leave lower tiers to go to higher tiers.

With Tier 1 open and JQ full, there are likely more players leaving JQ to tier 1 servers.

This is a business after all.

You’re saying anet has deliberately closed JQ to get players to move? If I thought anet were that manipulative I’d leave the game.

TorquedSoul.8097:

Here is the NA server activity since the linking.

I’m not sure if Northern Shiverpeaks could actually compete in Tier 1 but its activity is comparable.

Dragonbrand’s activity is taking a dive.

As being a resident on Dragonbrand yup our server is pretty much done. But not for the reasons many like to force them selves to believe. DB has declined so far so fast. Because of the meta, and the change of the mentality of current WvW guilds and players.

Most if not all of the guilds that was carrying DB pugs, and sub par PvX guilds. Was pure fight guilds. These are guild who pride them selves with fighting and not PPTing/PvDing. So atleast for me it came at no surprise. That these guild started to get demotivated to play. When the paradigm switched from who was the best fighters, to who was the best at PPTing, PvDing, and running away to avoid the PPK loss/Fights. Many of the those fight guilds tried to adapt to the paradigm shift. They started to blend into the crowd and siege hump and PPT for the servers. But at that point many just did not find that fun at all.

So now they are gone. Many of the fight oriented DB players or now playing other games or have found other hobbies. So of us still come on the forums or log in time to time. But for the most part fight oriented players like myself, barely push 8 hours of GW2/month anymore. A far cry from the minimal 6 plus hours a day, plus the many hundreds of dollars spent a month on the gem store.

ANET has always been incompetent when it comes to server balance.

Sylvyn.4750:

TorquedSoul.8097:

Here is the NA server activity since the linking.

I’m not sure if Northern Shiverpeaks could actually compete in Tier 1 but its activity is comparable.

Dragonbrand’s activity is taking a dive.

What do the numbers on the Y-axis represent? Is this something that is more appropriately used to compare only a server’s activity against itself over time rather than an indicator of who might win a matchup and what tier they should be in?

I assume those data are simply base on captures and kdr, retrieved from API. However, it lacks actual population so these activity level is questionable as the most important factor, active population, is not present.

Grim West.3194:

ANET has always been incompetent when it comes to server balance.

Its also not Anets fault that people keep moving, some people actually like the server they are on and arent 2 dorra hookers who can get bought… Deal with it or make a good suggestion which may or may not be responded to by Anet

SkyShroud.2865:

Sylvyn.4750:

TorquedSoul.8097:

Here is the NA server activity since the linking.

I’m not sure if Northern Shiverpeaks could actually compete in Tier 1 but its activity is comparable.

Dragonbrand’s activity is taking a dive.

What do the numbers on the Y-axis represent? Is this something that is more appropriately used to compare only a server’s activity against itself over time rather than an indicator of who might win a matchup and what tier they should be in?

I assume those data are simply base on captures and kdr, retrieved from API. However, it lacks actual population so these activity level is questionable as the most important factor, active population, is not present.

That is correct. The y axis value is a combination of captures and kills and deaths.

You are also correct that it lacks raw population data, but a person that is logged in and neither capturing an objective or engaging in fights is probably not a factor except for maybe scouts.

I still think is activity is the best way to tier a server. Equal activity doesn’t mean equal performance anymore than equal population would mean equal performance. Organization and coverage are still factors.

morrolan.9608:

TorquedSoul.8097:

Here is the NA server activity since the linking.

I’m not sure if Northern Shiverpeaks could actually compete in Tier 1 but its activity is comparable.

Dragonbrand’s activity is taking a dive.

And JQs is at T3 level so why is it closed? If population is based off the last month during that whole time the JQ population has been less than Maguuma and now YBs. YB is full but Maguuma isn’t, there is an inconsistency here. Moreover YB’s population is not at Maguuma’s size.

Depends on how Anet measures the activity. If they do unique logins on the maps throughout the week, that could be dramatically different from manhours on the maps for the week.

If they are trying to balance the competition, the should be tracking manhours.

JQ should be open or linked. If it is neither they are kinda getting screwed.

MangoCrush.7819:

Its also not Anets fault that people keep moving

It´s not Anets fault that people want to move, that is i think natural and has always been that way. But it is clearly Anets fault to let them move unhindered. They could have easily implemented some method to prevent that bandwaggoning but they either did not forsee it or were too greedy because they want to sell gems. Either reason is a joke and only shows how distracted they are from their own game when it comes to wvw.

TorquedSoul.8097:

morrolan.9608:

TorquedSoul.8097:

Here is the NA server activity since the linking.

I’m not sure if Northern Shiverpeaks could actually compete in Tier 1 but its activity is comparable.

Dragonbrand’s activity is taking a dive.

And JQs is at T3 level so why is it closed? If population is based off the last month during that whole time the JQ population has been less than Maguuma and now YBs. YB is full but Maguuma isn’t, there is an inconsistency here. Moreover YB’s population is not at Maguuma’s size.

Depends on how Anet measures the activity. If they do unique logins on the maps throughout the week, that could be dramatically different from manhours on the maps for the week.

If they are trying to balance the competition, the should be tracking manhours.

JQ should be open or linked. If it is neither they are kinda getting screwed.

Kinda why there’s a “Oh heck with this!” attitude. Dev’s are pulling way too many levers. That kind of manipulation is not interesting, it’s not refreshing, and it certainly isn’t fun. Stable and balanced populations and matches were better for the mode.

Population calculations are probably getting skewed by players coming in for dailies, for harvesting, and a biggie is having alt accounts. I wouldn’t even count a player that just runs reset as a viable WvW player. You should at least be playing for 5% of the week to be counted.

TorquedSoul.8097:

I still think is activity is the best way to tier a server. Equal activity doesn’t mean equal performance anymore than equal population would mean equal performance. Organization and coverage are still factors.

I think the factor coverage is quite important, organisation will immediately have an impact on population. Kill your zerg twice in a row and 1 out of 5 will leave wvw.
We discussed your “activity” rankings before and I think we both agree that if arenanet would do the tiers like you posted, then it wouldn’t work out well, but I am sure they have better predictors for “activity”. I would think they use the system currently in place for activity: the “participation” point system (the same thing that makes your reward track to go up when it isn’t maxed out) and maybe adjust it slightly. So they could add up all the points players would (theoreticly) get in participation over an hour to measure activity.
I agree that manhours probably is a good way to measure it, and a very easy one, and without any data it is hard to say what is superior but I guess both systems do not include possible coverage gaps that will have high impact on total scores.
So if you want to compare servers in their strength in the current system, the participation during off-hours is worth more than during primetime, because other servers will be unlikely to match it, so you should adjust for that (and thus increase the value of off-hour participation by 2-10 depending on when it is). And close servers depending on that total value that does also adjust for off-hours.

Instead of doing that for closing servers and matchmaking I would like it better if they adjusted scores to make primetime more valuable. And I think they are working on that.

I think alt accounts might have a lot more of an effect than ANET or any of us can measure. Given the number of players who brag about having accounts across servers and all sides of a match-up it is no surprise to see WvW populations vary wildly week to week. Heck I am not even hardcore but have access to two different alt accounts due to inactive family members if I feel like experiencing different tiers or world pairings.

Take a look at NSP for instance who had been the underdog for a long time and then had their WvW population take a big jump shortly after their first win of this pairing. People just want to have fun and worlds that have maintained good communities through this linking mess attract a lot of those alt account world wanders but only as long as they continue to put out. They can always jump back to their “main” account when the fun takes a downward turn and there is no way easy way for ANET to parse that.

As pointed out on another thread servers are probably gaming the system by having players transfer to their linked server instead of the main server hence the population of the main server doesn’t change or even decreases.

Maguuma has been pushing BG in T1. correction.TC.BP.MG and Kaineng have been pushing bg. its a 4vs 1.