Unbalanceable Problems: Why The Server System Cannot Work (Part 2)

World vs. World has rarely, if ever, been balanced. At present, the only exception to the rules is Tier 1 of North America, where a consistent detente of sorts has reigned for years. This lack of balance is the result of the server system.

Servers are an approximation of a massed fighting force, with the ability for any number of timezones to be represented by people. And while this approximation was very useful early in the life of Guild Wars 2, its use has been outlasted by its weaknesses.

Population Shifts

The core weakness of servers is how prone they are to wild, inconsistent changes in population. This manifests in three, interrelated ways.


MMO players like to be rewarded for their effort. They really like to be extremely rewarded for their effort. For some, that means that when their server is winning, they show up in droves to secure the victory.

Conversely, if their server is losing, other things become more worth a player’s time. Put simply, if the match is going badly, it is near-assured to stay going badly as fairweathering takes hold on both winning and losing sides.

This is most clearly seen in the extreme rarity of last-second comeback victories. Oftentimes, the weekend alone decides the victor, because fairweathering cements the population and strategic advantages even further.

Designated Victor

DISCLAIMER: I did not participate in the third tournament. I was on hiatus from the game at the time.

For both the first and second WvW tournaments, silver and bronze leagues’ winners were known before the season began. When the lockdown period neared, players wanting to win evaluated the strongest-looking servers, and transferred to the best one.

With piles of additional bodies at all hours willing to overwhelm the lesser-populated opponents, victory went from “good chance” to “absolutely guaranteed.”

This was made even worse for the second tournament, when in an effort to “balance out” the various tiers, servers ranked 13-15 were made free transfer just before it. The 13th place server, Henge of Denravi, got dogpiled by so many people that those already on the server ended up complaining about how little WvW they got to do on account of queues. Spoiler alert: Henge of Denravi won silver league by a monumental landslide.

The Transfer Meta

Because of the relative ease of transferring, the most common answer to the question “my server’s WvW is dying, what can I do?” is “transfer to another server.”

And players and guilds do, with regularity. Server movements among the tiers are rarely because of vast improvements in strategy, or tactical skill. More typically, one server gets a lot of transfers at the expense of other servers, and moves up with stronger coverage.

The transfer meta has become more and more ingrained over time. At launch, free transfers caused ludicrous mass bandwagons, but now, players and guilds will build up a warchest of gold to transfer over to somewhere else whenever the fights become less appealing, or the server starts to look dead.

Combine all three of these, and servers aren’t working as semi-stable communities to gather together in the valiant cause of server glory. They are a gems-gated place where players and guilds happen to be at the time. Even the “ideal WvW” North American Tier 1 took 2 years to reach enough stability, most of it based around the idea of simply being how WvW is meant to be played*.

*I’m not commenting one way or the other about if they’re right, but that is their intent. Also, this stability recently changed by, you guessed it, guilds transferring off of Blackgate and onto Yaks’ Bend.

PPT vs. Fights

Another spanner thrown into the works is the chief division in strategic approach. On the one hand, some players who want to win in world score focus on Points Per Tick, the lion’s share of points gained. On the other, some players want to pit themselves against all manner of situations and see if they can succeed against all odds.

And they rarely, if ever, agree on the right way to WvW*. But one thing they have in common is this:

*Necessary caveat: some players PPT (as in, defend and attack objectives) purely to create fight situations, but both hardcore PPTers and hardcore fighters would say they’re “not playing the best way.”

The Nightcap Nightmare

In a 24/7, weeklong war, it’s a given that “nightcapping” is going to happen. People play at all hours, but not the same number of people play at every timezone. Thus, in non-peak times, whoever has more people gets a disproportionately larger slice of the pie.

While it can be argued* that the scoring system could do with some tweaking to better match player effort to their score impact, the strength of off-peak population drives a need to consider PPT at all times. Sometimes, this is to “make up for” a weak coverage area, and others it’s to beef up the lead even harder just in case coverage fails.

*And I and many others have in the past.

This grates on PPT players because in a very real sense, some players are more important than others purely because of what time they play, not because of any strategic or tactical skill they possess. It especially grates on people who want to win on score, as their efforts to siege, defend, and upgrade are more or less erased if they don’t also have a coverage advantage.

This also grates on players who want to pick a good, fun fight because over time, the coverage advantage becomes a stacked deck. Enemy structures are always fortified, and siege lines are always better developed on the opposing side. And oftentimes the joy of victory is erased by the fact that nothing comes of it.

“Just for the Fights”

Taking advantage of the massive combat that WvW regularly provides by seeking great fights seems like a good thing. But with the importance that scoring places on PPT, and thus on ranking, pursuing a “just for the fights” mentality alienates those who feel like they do the drudgework of defending, upgrading, and scouting.

Ultimately, this breaks the server, as people there “for the fights” eventually tire of others raining on their parade, and leave for elsewhere*.

*Consider both Sanctum of Rall before the Season 1 tournament, and Maguuma before Season 3’s.

In the present server-based system, both PPTers and fighters have to attempt to coexist, even when their approach to the game mode drastically differs. The tension of those aiming for a good brawl against those wanting victory in score never stops. And either people quit playing WvW, or transfer elsewhere.

Failing Pride

Despite all these problems, the myth of “server pride” persists. That the massive changes in population caused by transfers, fairweathers, disagreements in how the mode should be played, etc., are somehow staved off by a name. But discarding these, server pride has its own problems.

Servers are Meaningless

First and foremost, megaservers erased the backbone of PvX players who could potentially be recruited to “the war effort.” Add in the transfer meta, and servers as a fixed identity to take pride in doesn’t make sense.

When a player logs out of WvW, they go back to a hodgepodge of everyone, not only their server. Beyond WvW alone, the server is meaningless to the player. Often, servers are chosen not because of their history, but because of their position on the ladder.

Victory is Meaningless

Rewards for winning in WvW are terrible. They’ve never been good, even when unbridled karma training is encouraged*. And while it could be argued that mediocre rewards are to prevent people abusing the system for their own gain, that in no way justifies a terrible reward system that almost ignores who won and who lost.

*Here’s looking at you, Golem Event.

Knowing that the only reward is the temporary bragging rights of “we won” excludes a lot of people who would otherwise enjoy the game mode. Rewards matter, recognition matters, having something to show for the effort matters. And at present, none of those really exist for WvW.

Defending is Meaningless

Rewards for defending is even more nonexistent than capturing objectives. So at the very start, players have to weigh “pride” and intrinsic reward against the absolute lack of anything tangible. This is already a tough bar to meet for most players.

Because of the sheer strength of coverage, and the relative ease of doing Player vs. Door in off-peak times, the equation slips even further into the “don’t bother defending” side of things. Knowing that all of the hard work and effort will be removed as soon as people log off all but assures people won’t care at all.

Weigh all of these problems, and how they are closely tied with the server system*, and it’s a huge, unbalanceable mess. Servers are at once too fluid and too fixed, with the overwhelming option when things get rough to leave for greener pastures. Philosophies on how to fight beat out the actual act of fighting. And what little remains of the server definition is undone by terrible rewards and the fact they mean nothing anywhere outside of WvW.

*Honorable mention to the scoring system, but fixing an algorithm isn’t going to change how crummy massive population shifts are for the players, nor the core disagreement on how to fight.

The Myth of Self-Balancing

Many times when server imbalances are brought up, the “solution” offered has been that players need to move around to balance out the tiers and increase competition.

On paper, this sounds great. But players tend to be selfish, even in their altruism. Consider the example of Season 2 Henge of Denravi. Three servers were made free to transfer as an incentive, but only one got stacked, and ended up in the same place post-season as it had been pre-season. People came for their rewards, and left upon acquiring them.

Self-balancing is a chicken and egg problem of wanting competition, but also not wanting to be outmanned by transferring to a less powerful server. Short of a cold reset, there is no way to offer the right incentives to shift people around on servers until most matches are balanced. Again, NA Tier 1 took 2 years to become reasonably stable, and even that changed recently.


Looking back at asymmetric balance, it’s clear that servers fail at both major aspects of it. Resources for some servers are very limited, while for others available in excess. And rather than there being many different options for everyone, the server with more resources dictates what options the other servers get with fully upgraded structures to defend and paper ones to attack.

Servers are not the heart of WvW, and never have been. Focusing all efforts to building a structure around servers, with all their myriad problems, is not the right answer. What should be looked at instead is what has always been the backbone of the game mode: guilds.

The Alliances Series

  1. Putting Guilds Back into Guild Wars: WvW Alliances
  2. Unbalanceable Problems: Why the Server System Cannot Work
  3. Heart and Soul: The Guilds of World vs. World
  4. The Case for WvW Alliances
  5. The Alliance Structure: Sidestepping the Problems
  6. The Pros and Cons of WvW Alliances
  7. Rewarding the Competition
  8. Implementing Alliances

Putting Guilds Back into Guild Wars: WvW Alliances (Part 1)

World vs. World is a stale, buggy, unbalanced mess. Which is exactly how the game mode launched back in September 2012, minus the stale. Regardless of some minor changes to the format and a trio of tournaments, the only major change slated for the game mode is in the indeterminate Heart of Thorns future. And while that might fix the “stale” and “buggy” aspects, it’s very unlikely to do much about the balance.

WvW is also a ton of fun, with endless potential for unpredictable and challenging encounters each and every time players log in. This is why despite almost no attention, WvW hasn’t collapsed utterly. It offers something that structured PvP and PvE can only dream of at present: consistent spontaneity*.

*Not to mention mass combat, duels, open-field fights, extreme build customization, and straight-up siege warfare.

Yet the primary draw of WvW has been whittled away over time by problems in its very structure, a structure that by definition cannot be balanced. Slapping a new coat of paint on the mechanics via the new borderland map and a few other shifts will not solve the biggest weakness of the mode: the server system.

Over the next week (plus a day), I will lay out why the server system is broken, and what to do to create a better, longer-lasting system that WvW can thrive on far into the future. As a spoiler, it involves guilds and alliances.

But before I do that, I would like to write a quick aside about the concept of Asymmetric Balance, which is key to my entire argument about the server system being unbalanceable.

What is Asymmetric Balance?

Asymmetric balance is providing each team with a limited, but identical set of resources, and requiring the teams to decide how to use them across many different options. This is typically expressed with people, but also includes things like siege, supply, and upgrades.

Realm vs. Realm, World vs. World, whatever the term, thrives on asymmetric balance. It means that victory is determined by who deploys their resources best, either in defense or offense, distraction or brute force.

It does not mean that every fight that occurs on the battlefield will be “fair” in the sense of the same number of people on each side. Asymmetric balance is all about trying one’s best to make it unfair to the enemy team, pulling a fast one around their decisions and outplaying them on the strategic level.

But That’s Impossible!

Yes, it is. However, it is an ideal to aim for. Asymmetric balance can be reduced to two phrases:

  1. Limited resources
  2. Tons of options for using those resources

And these phrases will help illustrate exactly why the server system cannot create an approximation of asymmetric balance, but alliances can.

A look at all the problems that the server system has, and why they are impossible to fix is here.

The Alliances Series

  1. Putting Guilds Back into Guild Wars: WvW Alliances
  2. Unbalanceable Problems: Why the Server System Cannot Work
  3. Heart and Soul: The Guilds of World vs. World
  4. The Case for WvW Alliances
  5. The Alliance Structure: Sidestepping the Problems
  6. The Pros and Cons of WvW Alliances
  7. Rewarding the Competition
  8. Implementing Alliances

One Step From Death: This Should Not Be the End

As the recent specializations update and subsequent balancing pass have shown, the necromancer still needs a lot of help. Over the course of May, I released six posts that comprehensively looked at the necromancer.

The Necromancer’s Curse

Part 1 of the series took an overview of what makes the necromancer the necromancer, and the various ways that they’re quite frankly underperforming compared to every other profession.

And at its core, the necromancer is currently identified by three things:

  1. Attrition
  2. A Tank Tendency
  3. Afflicted, but Not Crippled

Shrouded in Black and White

Part 2 turned its attention to the unique profession mechanic of the necromancer: death shroud. More specifically, the binary, extremely limiting choice that death shroud forces necromancers into.

Worse still, this is a choice that by construction the balance team has to force on necromancers, because death shroud lacks counterplay. The entire mechanic needs a redesign to give enemy players counterplay, and necromancers options.

The Dwarf and the Demon

Part 3 surprised me. I sought to draw parallels to the upcoming revenant, but ended up in a different direction. Revenant has focus, self-synergy, and a concept I’ve termed burst attrition. They hit hard, but only ever so often.

All things that the necromancer lacks, but should have. And it’s a core identity issue. Necromancer as it is implemented cannot have these positive characteristics.  It needs a redesign from the ground up.

The Dance of (Self-Countering) Death

Part 4 began my look at each of the three game modes that necromancers see play, starting with structured PvP. Because of the various woes of the profession’s structure, necromancers lack both skillful play and counterplay.

Ironically, this is a result of design decisions meant to reinforce the necromancer’s core identity. Design decisions that in practice mean that necromancers must completely outclass their enemies to win.

To offer a way around these decisions, I suggested recrafting death shroud around the break bar mechanic to reinforce the necromancer identity as a tanky opponent who must be either locked down, or burst down, but actually has the tools to still be effective in the face of competent players.

Designed Inefficiency

Part 5 is a look at the clear loser for necromancers: PvE. Everything unique a necromancer brings to the fight isn’t needed, and what they have in common they do worse than every other profession.

But unlike the core balance problems in PvP, the solution does not lie in recrafting their mechanics (though that would doubtless help). Necromancers (and all other professions) need unique mechanical encounters that require more than simply damage.

At Home in War

Part 6 should be a triumphant win for the necromancer, because they are gods in WvW. Necromancers are the staple of zergs and single roaming everywhere.

But that staple status is very tenuous because either the weaknesses of the profession need to be built around by the team (almost to the exception of all else), or the dynamic way fights develop across the borderlands can null their advantage immediately.

Even in their best mode, necromancers are on the edge of irrelevance.

This Should Not Be The End

In the process of writing these posts, I came to several conclusions, stated above. But to conclude further on them (Concluception), here’s my condensed list of what the necromancer needs:

  1. Recrafting design decisions to retain the core identity of the profession, but built better
  2. Death Shroud needs more options, and more counterplay
  3. A from-scratch approach to skills and traits
  4. Creating skillful play and counterplay for the profession as a whole (e.g., adding a break bar to Death Shroud)
  5. Crafting more varied encounters that can play to a necromancer’s unique strengths (and also other professions’)
  6. Not using the profession’s position in the WvW meta as a reason that other changes aren’t necessary

Having one profession condemned to irrelevance purely because of a broken foundation is not okay. If it can’t be fixed with the current setup of the profession, then Question Everything should go into full effect and a foundation that can be fixed should be put in place.

This should not be the end of the necromancer, where it is stuck being crud forever. It’s time to look critically at all of the problems, and generate a solution that in time can bring it up to the same standard as the other 7 (soon to be 8).

Applying Emergence: A Case Study

To conclude this short series on emergence, and the potential it has to keep theme park MMOs from an unsustainable treadmill of content creation, consumption, and complaints, I’m going to look deeper at how emergence can apply in Guild Wars 2, particularly with its brand new shiny dynamic AI system.

Previous posts in this series:

  1. A Theme Park MMO’s Bane: The Trifold Curse
  2. Dumping Sand on the Theme Park: Emergence

Emergence can apply everywhere, even if a dynamic AI system isn’t readily extensible to events, the environment, and the like. To show this, I’m going to overview the phrases of the emergence definition, and how they can apply to enemies, events, and the environment.

As a quick refresher, here is my definition of emergence:

Emergence is the use of simple, but layered mechanics to create a complex, nuanced, and ultimately unpredictable encounter, while retaining the ability to be comprehended.

Continue reading Applying Emergence: A Case Study

Dumping Sand on the Theme Park: Emergence

A few days ago, I wrote about the Trifold Curse, and how that ultimately leads to burnout, both of countless MMOs and the developers who make them. It’s not a good situation, but how can it be avoided when the vast majority of the industry at present is based on building feature-rich theme parks*?

*A sizable portion of players, judging by the number of MMOs made to get their attention, prefer the simpler theme park approach to content over the situations that evolve organically in sandboxes like EVE Online.

Enter what I term emergence:

Emergence is the use of simple, but layered mechanics to create a complex, nuanced, and ultimately unpredictable encounter, while retaining the ability to be comprehended.

Sounds really complex, doesn’t it? Fortunately, elements of this concept already exist in some of the most recent content released for Guild Wars 2, which I will use as a case study for this post.

Continue reading Dumping Sand on the Theme Park: Emergence

The Dangers of Feedback Firestorms

Last week should have been an awesome week with prepurchase starting (and presumably, release getting closer), a metric ton of information about guild halls, and full details of exactly how big the patch that went live today is.

Instead, it turned into a nightmare of botched PR, hate and shaming on ArenaNet’s developers, and a viral level of outcry spreading from reddit to the same group that hosted the initial reveal of guild halls and prepurchase during E3 in less than a week. A literal firestorm of negative feedback.

And based on Monday’s post, the outcry worked. People are calmed down now, and most folks’ “fix list” has been accommodated.

But regardless of its effectiveness, it sets a dangerous, and ultimately unproductive precedent. I am going to show how first firestorms aren’t new to the Guild Wars 2 community, how these firestorms hurt everyone involved, and ways to keep a raging inferno from engulfing the studio.

Continue reading The Dangers of Feedback Firestorms

A Theme Park MMO’s Bane: The Trifold Curse

NOTE: This was originally released as a single post, but has since been split into three parts and expanded to better communicate my analysis and suggestions.

Theme park MMOs are all about the attractions. Dungeons, raids, open world, even PvP arenas are considered small pieces of the overall park. And just like real life theme parks, the attractions steadily lose their luster as time goes by, the newness fading into a sense of staleness.

Why this is can be traced to three particular elements, which when combined inevitably create staleness for the vast majority of the player base. Below, I trace these elements as they apply to the game Guild Wars 2, but the principles hold for any theme park MMO, even the behemoth World of WarCraft.

Continue reading A Theme Park MMO’s Bane: The Trifold Curse

Sometimes what you love just needs some TLC