WoW Classic's Population Problems

I woke up to the shocking news this morning that come August, Blizzard is planning to shut down more than half of all BC Classic servers in the West. (There are currently 79, and the plan is to shut down 22 of the US ones and 21 in the EU.) My old home Hydraxian Waterlords is among those servers meant to be put out of their misery, after Blizzard already killed it off for all intents and purposes back in November by offering free transfers away from it.

Seeing Blizzard announce such a large number of actual server closures is quite shocking, as they've long had a reputation for never shutting down servers. Connect into clusters, sure, but outright shut down? Never!

Now, this was never entirely true... for example there were some EU servers many years ago that acted as unofficial Russian servers and were closed down when Blizzard decided to set up actual Russian servers, but by and large, Blizzard definitely made sure to avoid any bad press associated with that kind of thing. Players would point and laugh at other MMOs for closing and merging servers, but never WoW.

So, does this mean that Classic's dying? Not at all, but boy, does it have other problems.

I actually started writing this draft about server populations a couple of days ago, after Redbeard brought up the subject in a comment. You see, before Hydraxian Waterlords' sudden death I had been blissfully unaware of Classic having any population issues at all, but since then I've been looking around and the situation is actually pretty incredible, to the point that I'm continually surprised that there hasn't been more reporting on the subject.

Let me once again illustrate what I mean by using data from ironforge.pro. I've referred to this site before - it collects weekly data on how many characters are featured in raid combat logs or as having participated in arenas. As I've also said before, this obviously doesn't present a complete picture of a server's population, as there'll be many players who don't get captured by either of those measures, but it does give a pretty good indication of general trends. And they are crazy.

I'd like to illustrate this by comparing the historical data from the last week of OG Classic to data from last week, which was Burning Crusade Classic's 52nd week... so exactly a year later. I will be excluding arena data from this, as there's no comparable PvP data for OG Classic, so it seems more fair to only look at PvE (while keeping in mind that dedicated PvPers are also out there, somewhere).

First off, let's look at the overall numbers. The last week of OG Classic registered 207,365 active PvE endgame characters in the US and Europe. Last week in TBC Classic a stunning 275,243 characters were counted using the same metrics, which is nearly 33% more! I'm not sure we can interpret this growth in numbers as a growth in actual player population, considering that it's much easier in TBC to also do some raiding on alts, but we're definitely not dealing with a dying game.

Next, let's look at the overall faction balance. Blood elves have caused a pretty dramatic shift here - where Classic finished with the raiding population being 54.5% Alliance, it's now down to 47.8%, with the majority of players being Horde instead. Still, in the grand scheme of things that's not really a problem, and it's actually surprisingly close to even.

However, when we dig down to a server level, things don't look nearly as good. For the purposes of this post, I defined a "decent" faction balance as the larger faction making up less than 65% of the population. I know this is fairly generous, and I'm sure there are people who already consider that unpleasantly unbalanced, but I just had to pick a cut-off point for comparison purposes and that's what I went with.

The point is, even with that fairly generous interpretation of what makes for decent balance, less than half of all servers (38) qualified for it at the end of OG Classic, and now, a year later, that number is down to 13. If you actually enjoy encountering the enemy faction out in the world and standing a chance at a fair fight, your options have become increasingly limited.

The absolute extremes of imbalance show in the form of what can only be called "single-faction servers", where one faction has effectively died out to the point of the other making up 90% of the population or more. By the end of OG Classic there were already 11 servers that could be classified as such, but now we're up to 15.

However, all this wasn't the real shocker to me. Are you wondering why Blizzard is shutting down half of Classic BC's servers if there are actually more or at least a similar amount of characters being played each week? Because while by the end of OG Classic there were only four "dead" servers (defined by me as having less than one hundred PvE endgame characters logged), we are currently sitting on forty of these in Burning Crusade. Blizzard finally shutting these down is way overdue.

If the total population has grown or at least stayed roughly the same, and half the servers are empty, where did everybody go? The answer is that through a combination of free and paid character transfers, the Classic player base has increasingly congregated onto a bunch of mega servers. Before the launch of Burning Crusade, the largest server in the West was Gehennas (EU) with 6,706 endgame characters logged. A year later, the new "king" is Firemaw (EU) with no less than 29,163 known max-level PvE characters - nearly five times as many. Incidentally, Gehennas has also grown to a population of 18,560 - however, it has also gone from having good faction balance at the end of Classic to being 100% Horde now.

I can't even imagine what it must be like to play on one of those servers. Nethergarde Keep with its ~3k active endgame players is way down the list but still feels a bit too big to me even now. I miss the cosiness of Hydraxian Waterlords being less than half the size of that before it was killed off. However, apparently this is not what the majority of players want, and therefore it had to die.

And then it hit me: This is what has been wrong with Classic all along; I just never realised it. Even when #nochanges was the motto of the day and people like Asmongold were campaigning even for bugs to be reintroduced into the game, there were two areas were people were surprisingly quiet on matters of authenticity. One was the user interface, because the most outspoken campaigners were going to mod it all away anyway (when a client update suddenly introduced up/down arrows for gathering nodes on the mini map it took me ages to find even one person to mention that this was a Legion feature and not something that should be in Classic). And the other was server population.

I talked about this a bit during my private server days. During the heyday of Nostalrius, its fans were often going on about how amazing it was that the server could (supposedly) support 10k concurrent players. Meanwhile I was happily plodding away on the much smaller competitor Kronos and grateful for the peace and quiet.

When Classic became official and Blizzard sought input from its intended players, the Nostalrius narrative remained dominant. Never mind the fact that servers were limited to about 3k concurrent players back in the day, that was just a hardware limitation and there's no reason not to cram as many people as possible onto a single server nowadays! There are no downsides, honest!

I was never comfortable with this narrative, but during OG Classic, I was lucky without even realising it. I started on Pyrewood Village, which is now the biggest PvE server in Europe, and did bemoan even at the time that it was too busy for my liking, but I re-rolled on Hydraxian Waterlords after a few months and immediately loved that it was more quiet there

I remember commenter Kring complaining about Classic servers being too big in those early months, but thoughts like that quickly receded into the distance for me because I was now playing on a server that was actually fairly "vanilla-like" in its population: where known "server personalities" would shoot the shit in the LFG channel and you'd join a group for a dungeon you didn't need just because it was an odd hour and you felt like being kind to those strangers looking for just one more to be able to start their run already. People would complain about bots and boosters and GDKP runs ruining Classic on the subreddit and it all felt alien to me because none of those things were happening on my server. Because it was small.

It makes me wonder whether we aren't witnessing WoW being "ruined" all over again, just in a different way. WillE posted an excellent video on this only yesterday, but even he thinks that players congregating on mega servers "is the sensible thing to do" because apparently having the biggest possible player pool to choose from for your dungeon groups and having a full auction house are the only things that matter nowadays?

I feel that just like the automated LFD tool, this is one of those things that looks like it has no downsides when you look at it from a purely utilitarian point of view, but then you see some of the increasingly aberrant behaviours that arise in the environment you've created and go all surprised Pikachu.

The main difference when it comes to population issues is that they can't entirely be blamed on Blizzard. Players have campaigned for massive Classic servers from the beginning, and have paid good money for the privilege of jostling elbows with thousands of other players in Stormwind and Orgrimmar. I can't entirely blame Blizzard for letting those players do what they want, not to mention that the cost of all those server transfers must be raking in crazy amounts of money for them. However, I do have to say that I think it's part of an MMO developer's job to be able to say "no" sometimes and to recognise that there are things that are detrimental to a game's long-term health, even if there are people clamouring for them in the short term.

I wonder if mega servers will end up being one of those things for Classic or whether Blizzard will manage to come up with some sort of solution. Just like with automated group finding, I don't actually think that big servers must necessarily cause problems for an MMO, but they sure seem to be leading to some wacky behaviours in Classic right now and they are definitely not true to the original Vanilla or Burning Crusade experience.


  1. WillE caters to the hardcore progression content, so "moar peeps the better" for raiding content and everything else supporting said hardcore progression is his modus operandi.

    I felt that Myzrael-US, when it was about 3-3.5k raiding toons, was about perfect. But once the top two Horde raiding teams and #2 - 4 Ally raiding teams all left, that kind of left a huge gap in the pool of potential raiders. If you look at the Myz population right now, I do count as a "raider" since I run a weekly Karazhan raid. However, I'd bet my entire WoW gold supply on that the actual number of raiders to be less than half of the "raiding population". Me, raiding on one toon, is very much the exception than the rule on Myzrael-US.

    That alt-ness of the TBC Classic raiding scene masks a lot of problems. There's only a certain number of people on a server, and if they're raiding on multiple toons, it's still only one person who would utilize the automated LFD tool at a time. I'd imagine if you tried to multibox in a LFD tool, you'd find yourself bounced from the group before you'd get past two pulls.

    If I were Blizz, I'd turn off layering entirely on the huge pop servers and let the players deal with the consequences of their actions. You want a gigantic server, you got one. Have fun with that.

    1. I remember some people were extremely anti-layering at the beginning, suggesting that Blizzard might never turn it off. I didn't see why they wouldn't, but here we are...

  2. I'm on Hydraxian Warlords or rather I was when I was playing. I loved the peace and quiet there. Another reason not to return, as if I needed one.

  3. I'm also in the camp that I like smaller, quieter servers. There's something appealing to being one of a few adventurers out in the world. This is probably nostalgia, but it just felt nicer to not be competing with a crowd for resources. Of course, the downside was fewer folks to do group content with, but these days raid-size content isn't a concern for me.

    That said, I think given how the retail mindset has dominated (invaded, infested) the classic world, plus the death of #nochanges, moving more to a megaserver system is likely. Quantity of player bodies is more important than the quality of the server life. :sigh:

  4. My characters are still on HWL. Play time tapered off due to not enough people around after the free transfer debacle and other games. However, I did plan on returning once Wrath Classic was on the table.

    With the server migration I'm in a bind what do do now. My characters names are usually taken on other servers. From witnessing friends with similar RP background like me re-naming their character and then stopping to play, because they couldn't connect to the character again and/or couldn't get used to the cultural differences of a PvE realm, it's likely I will have the same experience.

    I don't know. Maybe I should just not play Wrath Classic.

  5. In "The Decline of MMOs" prof. Bartle speaks against mega servers:

    > Worlds should be made smaller-population and there should be more of them. Cloud-based servers allow this. If you have 100,000 players, then instead of 10 servers of 10,000 players each, try 400 servers of 250 players each.

    > https://mud.co.uk/richard/The%20Decline%20of%20MMOs.pdf

    There is also "Dunbar's number" which, in my opinion, makes a good argument for server sizes between 500-2500 player max.

    > https://en.wikipedia.org/wiki/Dunbar%27s_number

    Mega servers are just replicating the anonymity we already have in today's society - which isn't really an argument to play an online game.

    > The main difference when it comes to population issues is that they can't entirely be blamed on Blizzard.

    Without paid server transfers, and Classic didn't have server transfers (at least for the majority of its existence), we wouldn't have gotten to the point we're at. Shelling out 50 bucks for a transfer is much easier than spending the time to level and gear again - especially for older people who already played Classic.

    1. 250 players per server sounds a bit low even to me, considering that they won't all be online and interact with each other at the same time, but I guess it would depend on how big that world is geographically and what sorts of things players are expected to do in it.

      I pretty much agree with everything else, except that I still sympathise with Blizzard about the server transfers. I'd be happier without them, but it can't be denied that lots of players have been clamouring for them from the start of Classic. Like I said, Blizzard probably should've been more willing to say no, but I understand why "just give the players what they ask for" can be tempting.

  6. I'm sad to see Hydraxian will be closing. I've played WoW on and off since beta and Hydraxian was the only time I felt like I was playing with a bunch of friends. Sad times.

    Why does Classic have a population problem? The mega-servers are part of it. Blizzard's idiotic transfer policy that allowed transfers off smaller servers is another. But there's a third reason and that's TBC. When Classic launched we, the players, were able to play a version of WoW that no longer existed, an Azeroth that had been replaced on live by the Cataclysm version. But when TBC launched we were playing a version of WoW that still exists on live. Outland on live, barring the changes to the game itself, is broadly the same as the classic servers. So once the novelty of the launch was over people started to move on. Wrath will see something similar.

    However, if the rumours of fresh servers are true and if one of those is a replacement RP server then I might be tempted back.