How Much Do MMOs Drop Off After Launch?

The short answer is: “Usually a lot.”

The follow-up question is, “Yeah, but how much exactly?”

The semi-short answer to that is, “Unless developers start talking more about their active user numbers, we’ll never know. But we can form some reasonable guesses.”

Numbers don’t lie (much)

At the heart of much of my research on this blog is a reliance, as often as I can, on facts and cold numbers, whenever they are available. As mentioned, very rarely do we have any solid info on how many players an MMO has and, in the absence of real data, estimates are usually very heavily biased by the likes and dislikes of the reporter. “Everyone I know quit playing that game and I hate it so it’s dead.” Because, of course, you and your five buddies represent the bulk of the gaming community.

(As a consequence, if you have a dispute with anything I post here, I’m more than happy to debate you on it. But if I offer data, even if it comes from a shaky source, and all you can offer is, “I just don’t think that’s right,” save it. I recognize that my sources aren’t always going to be perfectly transparent, or even accurate, and I’m usually careful to phrase what I say in that way: “This source indicates that Game X has more players than Game Y” is different from the outright claim, “Game X has more players than Game Y.” Learn that difference.)

Now that has severely restricted how free people can access its data, and subscriptions start at an onerous $199/month — and go up from there — I’m forced to go back to Alexa for my website viewer numbers. This, combined with my increasing dislike for tracking and using Facebook and Twitter numbers to take snapshots of user activity has led to my pretty much abandoning the Buzz Index, but I’m messing around with various other ideas to bring it back.

Looking at the number of visitors to MMO websites can give you great insight into the histories of those games, and seems to offer a pretty good analog to the number of players in the games, at least as well as our “feelings” tell us. Yes, I know a lot of people never visit the sites of these games, but enough of them do that I think the sample sizes are relevant enough.

For instance, here’s Star Wars: The Old Republic‘s chart:

SWTORIt’s easy to pick out the launch, near the end of 2011. And that little spike near the end of 2012 is when the game went free-to-play. The rest of the chart displays the long, deep slide we all know SWTOR underwent throughout 2012, as subscriptions like grapes in the Tatooine heat. I think the mini-spike just after the midway point of 2012 was the result of the F2P announcement, which obviously would have gotten a lot of people visiting the site; telling, though, that it’s still less than the typical number of viewers to the site a full year before.

Slipping away

So how do I answer the question in the title of this article? How about like this: I’ll take what I saw as the four major MMO releases of 2012, along with SWTOR, and compare their peak — i.e., around launch — with their traffic three months and six months after.

For SWTOR, I’d say the peak hits at 0.16. FYI, that’s Alexa’s estimate of the percentage of global Internet viewers who visited on that day. Wikipedia claims about 2.25 billion people were using the Internet as of 2011, so that would mean 3.6 million visitors to SWTOR‘s site. Honestly, I don’t know how much faith to put in those numbers, but I could at least believe that a site Alexa tracks as “0.16” has roughly twice the visitors as one that’s “0.08.”

Three months later, I’d put SWTOR‘s site right about the 0.05 mark. After six months, I’d say it’s 0.02. Thus, the site had 31.25% of the traffic it did at launch three months later, dropping to 12.5% at six months.

How does that compare with 2012’s “big” releases? Here are their charts:

EnMasseTSWGW2PS2And the compilations of their “declines”:

Game Launch 3 Months 6 Months
SWTOR 0.16 31.25% 12.5%
TERA 0.025 16% 8%
TSW 0.035 17.14% 21.4% *
GW2 0.11 31.8% 22.7%
PS2 0.042 14.2% 11.9% *

The Secret World eliminated its sub fees before hitting its six-month mark. Ignoring the one dip in its graph, I’d estimate the sustained low point at about .003, or 8.6% of its launch traffic. PlanetSide 2 will turn six months old on May 20. The listed six-month figure is for its most recent traffic stat.

While it’s true that using this method means that games are “penalized” for having big launches — or for at least attracting a lot of viewers to its web site during its launch — I don’t know that we can say, in this day and age, that any AAA MMO isn’t hyped to the moon and back at launch. And the general similarity of all the charts — a major buildup prior to launch, followed by a huge spike at launch, and then a decline of various rapidity — I think indicates that one launch is pretty much like any other, relative to how the game performs later. Take away the numbers and dates and you’d be hard-pressed to identify one game’s chart from any other’s.

Based on these results, I’d feel pretty comfortable in saying that Guild Wars 2 has weathered the erosion of its player base through its first six months better than its big-time contemporaries. Meanwhile, it’s easy to see why TERA, SWTOR, and TSW abandoned their sub fees, with their numbers all hovering around 10% of their launch totals. PlanetSide 2 is on the same track, though I find its extremely narrow “launch spike” intriguing, and maybe not a true representation of how much attention it got at launch. Also, it’s fared better in the three- to six-month period than any other game. Here are the six-month numbers divided by the three-month numbers, using TSW’s pre-B2P stats:

SWTOR: 40%
TERA: 50%
TSW: 50%
GW2: 71%
PS2: 84%

Not all that close

Finally, in case it wasn’t obvious from the first column in the chart above, let’s look at one final Alexa chart, which shows all the above games in one chart:

All Here’s a chart using just the last three months’ worth of data:

All3moI don’t know if BioWare will ever recoup the expense of SWTOR, but for as badly as they bungled the initial game — and as much flak as they’ve received for their F2P practices, the game is doing remarkably well. It’s been on a solid upward trend following its F2P transition in November.

Go back to the parenthetical paragraph at the top of this post if you need to, but with this, I can feel pretty confident in saying that anyone who says SWTOR or GW2 is “dead” or that the other games are doing just as well as those “overhyped” games is talking out of their nether orifices.

This entry was posted in MMOs, Video Games and tagged , , , , . Bookmark the permalink.

3 Responses to How Much Do MMOs Drop Off After Launch?

  1. Thanks Jason, great article.

    For SWTOR, the business question I have is the royalty payments to Disney for the Star Wars IP. Not sure if there is a guarantee or minimum fee BW needs to meet to keep the MMO alive.

    For GW2, I uses to think it was how fast it would take GW2 to hit the 6.5 million box sales to overtake the original GW. I don’t know anymore, haven’t really heard sales numbers lately.

    Also for GW2, I was looking for the box sales numbers for their first campaign/expansion, but I’m not even sure they are going to do B2P box content.

    • jasonwinter says:

      The payments to Lucas, not to mention just the development cost will probably ensure BioWare doesn’t turn a profit from SWTOR for a good long while. GW2’s already sold 3 million copies (as of January), NCSoft makes $48 from each, so that’s $144m right there, which is probably around 2x what it cost to make.

      I would think they’ll do a typical expansion model like they did for GW1.

  2. Pingback: Words Are Wind | Wintry Mix

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s