Daily Archives: August 27, 2018

How Many People Play EVE Online?

If we still had Blog Banters going this is one I would throw out as a topic because it is an exercise in estimation given incomplete data.

This is a question that came up in an email exchange with Bhagpuss.  I do, on occasion, communicate with people via sources other than the main page of this blog.

I dropped him a note about something I heard on fleet coms a while back.  A player opined rather firmly that EverQuest currently has more players than EVE Online.  I have reason to believe that the player in question is/was in possession of information that indicated this, that it wasn’t just BS on coms but somebody with enough connections in the industry to know.

I shared this with Bhagpuss because we both enjoy these sorts of informational tidbits, but I wasn’t sure it was worth a blog post.  And it probably isn’t.  If it was I would be done writing by this point.  Instead Bhagpuss asked the pertinent question, which is the headline for this post.

Neither of us has any insight or information as to how many people play EverQuest these days.  The game is coming up on 20 years in age and in many ways feels its age.  But a lot of people played EverQuest back in the day.  It was the gateway drug into MMOs for a lot of people and it retains a lot of nostalgia value.

However, Daybreak is a private company so we don’t get any financial numbers, much less subscriber or player numbers. Even when it was part of Sony its numbers were buried so deep in the financials of its various parent organizations as to be invisible.  All we really know are some estimates based on press releases and various guesses and whispered information.

Subscription estimates – 150K to 1 million

There is that old chart.  Click on it to make it bigger/legible.

It shows EverQuest peaking at 550K subscribers and then dropping down, with the last number being 100K at some point during 2010.  After that the estimates stop.

EVE Online is also on that chart and it peaks at 500K worldwide some time at the end of 2012, which coincides with the peak concurrency event for the Serenity server in China.  The Serenity server’s brief moment of popularity fell away rather quickly if you go and look at the numbers at EVE Offline.

The numbers for Tranquility, which hosts the rest of the world, peaked around 350K at about a year before, after which there is no further data.  I suspect that CCP switch to world wide data during the Serenity surge because that sounded better.

There is also the theory that some put about that the player numbers for DUST 514 were being folded into the overall EVE Online numbers because we were able to drop rocks on them from space and possibly see them in our in game chat.  The only issue there is that I don’t think that DUST 514 numbers were enough to influence the total that much.

Anyway, by the time we get to 2014 we are out of even estimates as both companies had clammed up.  There are no press releases for dropping customers, only for hitting new peaks.  And since the end of the data points for both EverQuest and EVE Online both have gone free to play in their own way.

For EverQuest went free to play back in 2012 for its 13 birthday.  Some bar mitzvah present.  At that point the live servers were free but you needed to buy at least the latest expansion if you want to play the new stuff and if you wanted to play on one of the nostalgia servers you had to subscribe.  This was a somewhat traditional free to play, with the nostalgia server bonus for Daybreak.

For EVE Online the free plan meant the introduction of Alpha clones with the Ascension expansion back in November 2016.  Alpha clones were free, but could only use a limited skill set and the client blocked you from multi-boxing Alpha clones.  They couldn’t cloak, run a cyno, or fly anything beyond a cruiser of their racial choice initially.  That loosened up with the Arms Race expansion last December, but you still can’t log in more than one unless you’re tricky.

Industry wide there has been the claim that going free increases the player base of an MMORPG, but we have to stop talking about subscribers and just talk about players, since not everybody who is playing is necessarily paying.

For EverQuest that is all interesting, but doesn’t help us much.  For EVE Online though we have the data points referenced above at EVE Offline.

Lots of data

The question is, how does concurrency map, if at all, to the player base.

We know that the player base is greater than the concurrent users online because not everybody is logged in together, and all the more so for a server that hosts a world-wide player base.  So the simple answer seems to be to find the ratio of concurrent users to known player base numbers from the past to see what estimates that gives us now.

Looking at the late numbers for Tranquility in 2012, we have 350,000 users.  It is a little more, but I will take the round number for ease of use.  The average concurrent users number for 2012 was 43,000.  That gives us a ratio of 8.14 to 1.  One concurrent user equals about eight total users.

If we use that number and multiply it by the average concurrent users for the last 12 months, which is 33,000, we get an estimated player base of 268,604.

That number seems problematic to me.  I guess you could convince me that EVE Online has that many players, absent other data.  But if we are buying into the statement that EverQuest has more players, then that is where I stumble.  It might be just me, but I have trouble with the idea that the EverQuest population jumped so much with free to play that it is 2.5x its last subscriber entry six years later.  Is there that much nostalgia for Norrath?

Maybe 2012 isn’t the right year to go to for the ratio.  Let’s back off to 2010.  The subscriber number on the chart is 300,000 for that while the average concurrent users number is 47,000.  That gives us a ratio of  6.38 to 1 and a possible user base of 210,638 players.  That gets us closer to reasonable, but it still seems like a big number for EverQuest, beloved though it might be.

If we go back another year to 2009 the subscriber number seems to be about 225,000 and the concurrency number 44,000.  That gives us a 5.11 to 1 ratio and a possible player base of 168,750.

Now I have three numbers with a range of 100,000 players and my like or dislike of them is based solely on my gut feeling for how many players EverQuest might have these days.  The lower the number gets the more comfortable I am with it for EQ, but the smaller the ratio gets the less likely it seems to be accurate.  Again, that is my gut speaking, but 20% of the EVE Online player base being logged in at any given time seems like a lot.

And I can keep moving around the years and getting different numbers.  2008 gives a ratio of about 5.71 to 1, so more, while 2007 sums out to an even 5 to 1, making for less.  The subscriber base is between 150,000 and 225,000 for those two years, while the average concurrency is 30,000 and 35,000.  The average concurrency for the last twelve months sits between those two numbers, so perhaps the current number is also between 150,000 and 225,000.

But with free to play the ratio ought to be higher, there ought to be more casual players on as Alpha clones.  After all, we have been told time and again that a free option increases the player base.  Then, however, I get back to the “must be less than EverQuest” parameter which seems less likely to me the more the current player base estimate grows.

So I don’t really know.

The initial assumption could be wrong.  Those old subscriber number charts could be wrong.  And what constitutes a “player” in any case?  There is always the differentiation between subscribers, players, and active accounts.  Few games reward players for running multiple accounts the way EVE Online does, so the ratio of players between EQ and EVE might be different enough from the ratio of active accounts as to be significant.

In the end, all of this is just food for thought as I don’t think anybody is going to give us any real numbers, but I’d be interested if anybody could come up with another way to try an get an estimate on numbers.

Addendum: After all of that Hilmar spits out a number in an interview with Venture Beat:

GamesBeat: Is the game still well into the hundreds of thousands of players?

Petursson: Yes. The MAU fluctuates a bit, but it’s 200,000 to 300,000 people.

Well, a range of numbers that is still 100K wide, but the base is 200K.  MAU is still a BS metric the way some companies (like Blizzard) use it, but in this case it gives us something to stand on.  It doesn’t distinguish subscribed from free to play or count those playing the long game on training, but it is a number.