Nosy Gamer's rankings are based on data from Xfire and and thus reflect its fairly hard core demographic, so with this week's rankings out showing WoW indeed beating GW2, Tobold gets to crow about being right.
But since games are now played by people far outside the narrow demographic of hard core gamers measured by Xfire, I'd like to call attention to another source of data which can be used to measure a game's popularity—Google Trends.
Below I'll show that the Google Trends dataset reveals WoW's long-term decline despite the four expansions released over the years. It also shows that while a few western MMOs have spiked above WoW's popularity for a short time, only one asian MMO has been able to sustain it. There are a few western games that are more popular than WoW but they're not MMOs.
In the comments that follow, when I refer to a game's "popularity", please take that as shorthand for the game's popularity as a Google search term as shown by Google Trends.
When viewing the embedded charts, mouse over the chart to see individual data points.
WoW and Everquest
Google Trends has data going back to January, 2004; WoW launched in November, 2004, thus Google Trends conveniently captures WoW's entire existence. I believe that EverQuest was the leading western MMO at the time of launch, but WoW took the crown and hasn't looked back since:
EverQuest had a spike of its own due to the release of EverQuest II a few weeks prior to WoW's launch.
Note that prior to WoW's launch, EverQuest had a level of popularity that was higher than WoW's has been for most of 2012.
WoW and its expansions
This shows the popularity spikes associated with the release of each of WoW's expansions—The Burning Crusade, Wrath of the Lich King, Cataclysm, and the brand new Mists of Pandaria.
One can detect a gradual increase in popularity over the first three expansions; each one seems to slightly outdo its predecessor. This could be due to increasingly effective PR on Blizzard's part.
Despite the spikes, none of the expansions altered the fundamental downward trend for WoW.
WoW and the competition
This group of competitors include games that received a lot of attention at launch and each one was hyped as a WoW-killer for a period of time: Age of Conan, Warhammer Online, Rift, and Star Wars: The Old Republic.
Whether or not they contributed to WoW's decline, of the four Rift has had the most gradual reduction in popularity.
I omitted EVE Online and Darkfall (A.K.A. SynCaine's darlings) because they barely show up, and Trends can only show five searches at a time.
WoW and Guild Wars 2
Now we get to the subject of Tobold's bet: Guild Wars 2. First, a look at both Guild Wars and Guild Wars 2 using the same timescale as the earlier plots.
The first thing we see is that the popularity trend for the original Guild Wars looks quite different than the competing MMOs shown earlier. It has both a higher peak and a slower decline than anything shown thus far save for WoW itself.
The second thing we see is that with a peak in the upper 50s, Guild Wars 2 spiked higher than any MMO or expansion save for WoW itself (but see below). The last time WoW was over 60 was February, 2007, which is five and a half years ago!
The third thing we see is that Guild Wars 2's spike is over and its popularity is in sharp decline.
(I didn't embed a separate plot showing the GW1 campaigns, because unlike the WoW expansions the campaigns didn't get much attention compared to GW1 itself. Factions and Nightfall were stand-alone games that didn't require GW1, while the last release, Eye of the North, was a true expansion requiring one of the other titles to play.)
For Tobold's bet we need to zoom in so we can see a finer resolution than monthly. Showing just August, September, and October allows us to see data for individual days. Indeed, the weekend bumps are clearly visible:
And here we see that Tobold shouldn't crow too loudly, because as far as Google Trends is concerned, WoW beat GW2 for only a single day before dropping again to second place.
Now it's quite possible that in the next few weeks Guild Wars 2's popularity on Google Trends will subside further until it sinks below WoW.
[Edit: 2012-10-04] Shortly after publishing (and after dropping links in a few places, of course) I noticed that I'd forgotten to add together the values for WoW and Pandaria which I'd earlier realized would be needed to properly compare WoW's popularity with GW2's. Adding them together shows WoW maintaining its lead over GW2, although narrowly, as they both decline:
 |
| Combined Wow and Pandaria vs. GW2 (click to enlarge). |
[End edit]
In measuring GW2's face off against WoW through the fall it will be interesting to compare Xfire's data with the Google Trends data—it could well turn out that Xfire tracks hardcore MMO raiding players better, favoring WoW, while Google Trends reflects better the casual MMO player, favoring GW2.
WoW, RuneScape, and the asian MMOs
Now this is interesting:
I saved the chart of RuneScape and the asian MMOs for last since it's so different from the others. RuneScape beats WoW by a significant margin, and Aion's spike slightly exceeds GW2's.
Lineage II does surprisingly modestly for a game which (at least at one point) had so many subscribers.
I didn't embed a chart for Final Fantasy XI and XIV since neither scored particularly well.
Summary
I believe I've demonstrated the value in using Google Trends as an alternative or to augment Xfire statistics when measuring MMO popularity.
In a forthcoming piece I'll further show the utility of using Google Trends to track game popularity.
Posts
I have to say your choice of tool is definitely interesting and I'd like to thank you for the insight.
ReplyDeleteI have a question though, is Runescape Asian? It's been developed by a Western company and doesn't seem to have the language support or servers in the countries commonly associated with Asian MMOs (eastern and south-eastern Asia).
There might be more problems with tracking, the users may search for the games using a name written in their own alphabet (if they don't use Latin) or might prefer local search engines over Google (I believe this is the case in China if not other countries) so you won't track such users in this way.
This is an interesting way to measure popularity, but I'm not sure it actually measures what you want it to. Xfire, for example, measures number of Xfire users playing each game. So do you want to measure a game's popularity by number of players?
ReplyDeleteIf so, using Google Trends is not the best method. How many World of Warcraft players do you think are searching Google for "World of Warcraft?" That query is unlikely to return any results that will provide new information to regular players. Players going to Google for information will be looking for something specific associated with World of Warcraft, and they almost certainly will use "wow" or "warcraft" and not the full name.
I understand that you are looking for trends and comparisons, so an accurate measurement of the playerbase is not strictly necessary. But showing only the trends for the game/product names is more likely to show non-player interest in the game. This is an interesting comparison to explore, but I don't think it's what you were seeking.
Going forward, I would suggest you find something that is an extremely common search term for players of these games and analyze the trends on that. For ideas, you can start typing "wow" or "gw2" into Google and see what it suggests as common search completions. For "wow," the first suggestion is "wow Armory," which you probably know is a site to look at players' characters and is probably something that players, not potential players, are interested in. Similarly, the number one suggestion for "gw2" is "gw2 wiki," which is probably something that players are far more likely to search for than simply "gw2."
Imakulata, I've edited the post to correct the designation of RuneScape as an asian MMO.
ReplyDeleteFlip, I've addressed these issues in a forthcoming post. But the right thing would be for Google itself to produce the popularity ranking we seek. It could do this far more easily than we can since it has all the search data.