Wednesday, January 15, 2014

Why Rixx Is Right

Plateau? Maybe. Anything is possible. But this chart doesn't prove anything one way or the other.

If a blog banter is supposed to stimulate conversation across the Eve-o-sphere, then Kirith Kodachi's idea for Blog Banter 52 was brilliant.  As a device for looking at the future of EVE, Chribba's all-time average concurrency chart is perfect.  As the chart is relatively flat for the past 3-4 years, that plateau gives bloggers answering the call a relatively stable platform to jump from and make predictions of EVE's future.  But as Rixx stated in his own response to the topic, the chart by itself does not prove or disprove anything.

I think I'm at the point where I can back Rixx up and offer historical proof from Tranquility that just because the chart trends down that the subscription1 numbers do also.  On Monday I posted the results obtained by merging the data from with the data set containing EVE's subscription numbers maintained by  I then made a chart showing the changes in the average concurrent user and subscription numbers on Tranquility during five separate periods ranging from 6 to 28 months in length, shown below.

Falling activity does not mean falling subscription numbers

The first range, from July 2011 to March 2012, shows the one time I found when the activity on Tranquility increased by the number of subscriptions decreased.  Of course, that time period includes the Summer of Rage and the launch of Skyrim so I imagine that EVE was still suffering from all the rage quits and Elder Scrolls love.  However, those who remained were excited by Crucible and managed to make up the difference.

The next four entries are examples of when the ACU declined at least 8.5% but the number of subscriptions reported to MMOData increased.  The first period, January 2010 to September 2010, includes the launch of Tyrannis and the introduction of planetary interaction.  I'm going to guess that the reason for subs going up was an increase in PI alts combined with people getting aching hands from the PI click-fest that was the first iteration of the new feature.

The second period, from May 2009 to November 2009, is one of my favorites.  Anyone guessing that is when Unholy Rage occurred has probably visited the blog a few times before today.  The fact that EVE went 8 1/2 months without an expansion (Apocrypha launched in March) also contributed to the number of subscriptions going up and the ACU going down.

The third period, from January 2011 to July 2011, is very interesting.  That is the period right before Monoclegate and the "Greed is Good" disclosure brought CCP to its knees.  Players saw the declining activity and wondered what CCP was thinking.  CCP saw the 7.9% increase in subscriptions and figured everything was working great.  Sometimes numbers lie.

The last period I don't have a good answer as to why the numbers are so different.  From January 2010 to May 2012, the number of average concurrent users declined from 32,492 to 27,764 (-14.6%) but the number of subscriptions rose from 325,134 to 363,851 (11.9%).  While I don't have an explanation, I think this period is the perfect example for why Rixx titled his post, "The Graph Is A Lie".

So does this mean we shouldn't worry about the 9.7% decline in the ACU number from 33,911 in May 2013 down to 30,637 in December?  I have some thoughts on that, but I'll save them for another post.  What I do feel confident about stating is that if you see a website proclaiming falling subscription numbers for EVE Online, if that site is relying on the chart for Tranquility as the only source of information, you may want to take the claim with a grain of salt.  As far as I know, that information is securely housed within the offices of CCP.


1.  When I use the term subscription, I am referring to an EVE Online account, not a person.

2.  The numbers for average concurrent users and subscriptions from January 2009 to May 2012 are listed in the second chart in Monday's post.

1 comment:

  1. "While I don't have an explanation, I think this period is the perfect example for why Rixx titled his post, "The Graph Is A Lie""

    I think the explanation is very simple - players tend to acquire alts as they play EVE. L4 blitzing alts, Jita trade alts, hauling alts, spy alts, PI alts, research alts... and that's not counting professions like mining, where alts are pretty much a requirement.

    Do note that these alts will be all on separate accounts. This is because the ability to train two characters on the same account wasn't available until only recently, and it's still better to simply have another account now(because subscription is cheaper than PLEX and it's less hassle when running multiple clients).

    This is why the Chribba's graph spells doom for EVE - if players naturally "acquire" accounts with time, but the number of concurrent logged in accounts does not increase, then the only logical conclusion is that the number of EVE players is dropping at the rate at which the remaining players acquire new accounts.