Last year the numbers clearly showed that the number of users declined after the launch of Incarna. This year, the numbers took a drastic dive approximately one week before Inferno's deployment to Tranquility.
|Rolling 7-Day Average Before Two May Expansions|
Why did the 7-day rolling average of users on Tranquility drop 14.6% in the 9 days before Inferno launched? I included the numbers from before the Tyrannis expansion to see if such a drop is normal. I used Tyrannis because it launched 26 May 2010, which means it accounts for seasonal factors such as holidays, university exams, weather, etc. Such a big drop is not normal.
What also isn't normal is the release of one of the most anticipated video games of the past few years, Diablo 3. According to the Xfire numbers, the total number of hours spent playing all MMORPGs decreased by 16.8% in the first week after Diablo 3 launched. That Eve saw a massive decrease in activity is unsurprising.
I did find two other points of interest about the average concurrent user numbers for the 30 days before the launch of Inferno. The first is the effect of Hulkageddon V on the number of players logged in. Actually, I should state the lack of effect. Despite all the tears I saw on the forums, miners didn't log off en masse and stop playing the game. From the start of the event on 29 April thru 13 May, the 7-day rolling average of concurrent users remained relatively flat. The decline in numbers did not occur until Diablo 3 launched.
The second is that the average concurrent users in 2010 was slightly higher than that in 2012. Does that mean that the number of Eve Online subscriptions flat-lined for 2 years? Not at all. According to MMOData.net, Eve had approximately 320,000 subscribers in June 2010 and 361,000 subscribers in April 2012. So why the similar number of users logged into the game two years later? My answer is CCP's War on Bots™. Back in 2010 CCP still operated under the old mass banning at infrequent interval strategy when dealing with botters. Botters were not afraid to run their bots 23/7. This drove up the concurrent user numbers. With the new strategy CCP Sreegs introduced in March 2011 of banning botters continuously, bot devs started advising their customers to cut back on the number of hours they should run their bots. So even if the same number of people were running bots, the fewer active hours would result in a lower number of concurrent users averaged out over the course of a day. How much lower? I don't know, but on 22 June 2009, CCP banned 6200 accounts for botting as part of its Unholy Rage campaign. If CCP Sreegs and Team Security managed to keep just half that number from botting 24/7 then that could account for the difference between a 10% increase in accounts and no increase in concurrent users on Tranquility.