Does Your Bartle Type/Gender Influence Your Class Choice in WAR?
Posted by Sanya Weathers on Friday Oct 10, 2008 Under Market TrendsWhen a new game launches, the early adopters rush to make long-dreamed-of characters, and they choose the classes that sounded awesome on paper or in the demos. After that first mad rush, though, people tend to fall back on the archetype that they’ve always played.
For example, I am always a melee class. I don’t twiddle buttons, I don’t use magic, and I absolutely do not do anything that requires me to execute anything based on timing or reflexes. In fact, I don’t particularly like weapons. Weapons wear out, and get dropped by accident or by intoxication. Everquest had a monk class. I still miss that character. But in the absence of monkish goodness, gimme a sword and a board and a few specials based on a stun, and I’m good.
Now that WAR has been out for a few weeks, people are starting to settle into their real classes, not their hype-driven options. Classes that looked awesome in the demos may not work as advertised, and classes that seemed less than thrilling are hella fun to play. We here at GamerDNA were wondering – is there any pattern to what people choose to play? If there is, the people developing the MMO of tomorrow might glean something useful – and the rest of us can use this to talk trash on message boards. And after all, isn’t that the point of an MMO?
I’m kidding. Mostly. But seriously, let’s see what Bartle scores and a player’s gender might have to say about how you play. All of today’s data is derived from GamerDNA members who both took the Bartle and play WAR.
First, let’s look at the overall class distribution:
Pretty even, really, with pie segments ranging from 8% (Swordmaster) to 1% (the poor Squig Herder, which was absolutely the coolest character if you just looked at the concept art). Eleven classes have 5-6% of the population.
By the way, I know this is the chart all the board warriors really want to see:
It’s not totally cut and dried, but it does seem to show Order classes either have a lot of players or fewer, while Destruction is more evenly spread among the classes. In terms of the total, Destruction has 49% of the players, and Order has 51%. I would expect this number to tilt even further towards Order in the days to come. This is because of a little quirk of human nature – everyone goes to roll the freak characters, but eventually the majority will roll – and stick with – humans/”good guys.” But who knows, maybe WAR will break that mold.
Okay, that’s the general population. How does gender play into class choice?
Well. THAT’S interesting.
Before I ran the chart, I’d have said that women are stereotypically perceived as playing support roles – the healer to her boyfriend’s tank, as it were. I’d also have said that women make class choices just as men do. The only area in which women actually do have a preference is in the gender of their avatar. Women rarely choose male avatars, whereas men will often choose a female avatar if only to watch shapely female buttocks for the extended periods of time typical of MMOs.
The latter is proven out by the chart – Witch Elf, a female avatar only class, has more women playing than men, but men are still represented. Chosen, a male avatar only class, does not have a statistically significant number of women playing it.
WAR’s marketing of the classes (and to be fair, hewing closely to the lore established by Games Workshop) was more gender-oriented than usual for an MMO. That seems to be playing out in the class selections. Certainly no one thinks “delicate feminine beauty” when they hear “Black Orc.”
The Sorceress, in GW lore, is a female only class, and although Mythic ultimately made it available to both male and female avatars, it is considered female by hardcore Warhammer fans. Also, the sorceress is the most smoking hot character in the game. And women do like to be pretty when they’re online!
Along the “pretty” lines, I’m told that the Archmage is the most lovely and delicate fantasy figure without the crazed sexuality of the sorceress, making it the top choice of traditionalists.
And it’s a healer. Maybe I was wrong about support classes being a stereotype – or maybe it’s a stereotype because it’s based in truth?
All right, now, let’s see if Bartle can predict anything. Here’s the breakdown by “top” aspect:
In other words, if you are a K***, you are a killer for the purposes of this chart, and if you are an S***, you’re a socializer. In my opinion, the dominance of the explorer type simply reflects the fact that the product was recently released, and explorers prefer to play games before every inch has been spoilered to death. The killer preponderance was to be expected given the game’s theme.
But the 11% score representing achievers is a real opportunity for Mythic. By emphasizing the Tome mechanics, and leaderboard scores, there’s no reason achiever types shouldn’t see some marketing directed at their preferences.
How about class choice?
I can’t draw conclusions about the achievers and the socializers with the certainty that I can with killers and explorers, but I did find it interesting that even with a limited sample size, there were classes that people who identify themselves primarily as achievers simply don’t play.
The most fascinating thing about this chart is that there are four definite standout classes for the four types – Killers/Black Orcs, Explorers/Disciples, Socializers/Warrior Priests, and Achievers/Zealots.
I think… I think we here at GamerDNA need to run this data grab with other games!
Got any explanations for these trends besides what I’ve already given you? Tell me in the comments.
Did you like this article?
Register at gamerDNA.com and discover games played and enjoyed by people like you right now!gamerDNA connects the places you play (Xbox Live, Xfire, Steam) and the places you communicate (Twitter, Tumblr, and more soon) with your unique gamer identity, allowing you to showcase your discoveries, and explore the experiences of others who love games.
October 10th, 2008 at 10:23 pm
Interesting!
I am surprised that so many achievers seem to play a Zealot but NOT a DoK. Both are healing-type classes, which helps with survivability — something that is a must if you are trying to unlock a ton of things in the Tome (which I would presume equates to the need to achieve — titles and tactics and such).
October 10th, 2008 at 11:19 pm
Great article with lots of interesting statistics and analysis. I agree with the closing sentiment about running this data grab on other games.
October 11th, 2008 at 1:16 pm
Simply fascinating. I can’t wait to see if this data holds out over the long haul, and I agree we need to run this with lots of other games. It would be interesting to see if Bartle could be used to predict classes in a non-mmo, like say.. COD4. Or even if it could predict build-outs in games like Mass Effect, or Fallout 3.
October 11th, 2008 at 2:42 pm
So far I agree with the data presented here. I’ve seen a lot of the players and gamers attitudes to the way they play, and the way they approach the game itself. Course, the male/female bit I have to take your word for. We have 7 women in the guild, and no-one knows who they are (mostly due to the ‘idiot’ factor of some gamers), who like to keep that knowledge private for now.
The classes dominated by female players, they have a few things in common, game wise. Ranged Damage output, Healing abilities and Stealth. Three good traits for a solo player or quiet back-up player.
As gamers, we’re largely viewed as a sexist bunch… not exactly an undeserved title either. Considering the amount of harassment female players tend to get in other games, its no wonder they would learn their lessons and prefer characters that can be tailored for solo play. I’ll be watching this data with interest as time goes on.
October 11th, 2008 at 9:09 pm
@Solvoug, I actually think because Zealot is a more true achiever archtype. This is my personal opinion as a long time Achiever, but Achiever isn’t unlocking things solo necessarily. For me its working with a really solid, tight, uberguild which means we’ll need healers (maybe support) to get to the top. Almost every MMO I have played a healer after I learned how much help healers logging on in EQ1 swayed our ability to kill things.
I actually broke the mold a bit this time around since it is a PvP game and played a Sorceress, but my primary alt (the one most likely to also reach 40) is a Zealot. This is because I already see in my guild that we’ll need more Zealots than what we have.
October 11th, 2008 at 9:14 pm
Just to explain more why we need healers/support typically class distribution in an achiever guild is decided around beta/launch time. But as we all know people get into the game and decide they like/dislike things or simply don’t hang around leaving gaps. A well balanced team is needed usually to do high end content and usually healer/support end up coming up last - although there are certainly a lot of MMO vets that know this and just start as a healer/support class to begin with.
October 12th, 2008 at 11:10 am
Wow! Had I seen this article BEFORE I rolled my zealot, I would have known a lot sooner that it’d be my favorite class! I’ve never liked healer classes before, as I’ve always been a rogue and/or assassin in everything I’ve played prior to WAR. Then again, my Bartle says I’m an Acheiver, so the fact that my main ended up being a healer alt I created “just to shut people up in Nordenwatch” wasn’t without a grain of inevitability. I also find it interesting that not many females are playing Magus, as that’s my secondary character (because the witch elf, frankly, just got on my nerves after a week).
Solvoug — Maybe more of the acheiver types are female. I didn’t roll a DoK because there’s no female avatar for it, plus the idea of getting a knockback to finally put those IronBreakers in their place is pretty awesome.
October 12th, 2008 at 11:30 am
Very interesting - First off I have to say I didn’t take the test this time, however I tend to come off as Achiever. Also interestingly enough I play White Lion which seems quite rare for Achievers.
I’m not sure I can jump to any hasty conclusions, but it’s damn interesting anyway
October 12th, 2008 at 12:00 pm
This fascinates me, especially the disparity between the “paired” classes. The Disciple of Khaine and the Warrior Priest, for example, share the same basic class mechanic and mostly the same spells, only with different flavor text and visuals. Note that their gender representation is almost swapped - a larger percentage of males prefer the WP, while a larger percentage of females prefer the DoK.
I suppose that could be explained by art - the female DoK definitely appears more attractive than the WP - but the Bartle scores for these two classes are radically different too. Socializers and achievers are highly represented for the WP, whereas explorers are highly represented for the DoK.
Art direction and flavor text must be having a huge impact on these decisions, even more so than basic class mechanics.
October 12th, 2008 at 2:13 pm
[...] on their site. Not only did they crush that number, but they’ve put the statistics to good use. In an interesting research article by Sanya Weathers (who was just recently on the Massively speaking podcast), you’ll notice some [...]
October 13th, 2008 at 1:07 am
One important factor that comes into play with class selection is class mechanics. Certain mechanics just don’t appeal to certain players–even if the avatars, the role, and the backstory do. I, for one, would really enjoy playing a Witch Elf–if it weren’t for the whole “combo point” mechanic. (call it what you want) I tried a Rogue in WoW and they didn’t appeal to me because of that same sort of mechanic. I keep periodically trying the Witch Elves, but drop it quickly–because of that same mechanic. I don’t know why, but it just doesn’t appeal to me. The DoK or Sorcerer, however, feel a lot better to me–again, because of the mechanics.
WAR is a really tough game to analyze patterns and playstyles from. The developers were going for races that felt completely different from each other and classes that felt completely different from each other–so much so, that it creates a strange mix. The results can sit rather strangely with certain players because, unlike WoW, you can’t just choose a class and then look at the races that are playable with it. Your race determines your class options–exclusive options. Chances are, you would see distinctly different results if, for example, the shaman’s mechanic was instead only available for Dark Elves.
October 13th, 2008 at 2:12 pm
I like these articles from Guildcafe. I also wish that some articles could be dedicated to guild demographics, guilds as retention mechanisms, and guild friendly game features that MMO’s should consider based on those findings.
October 13th, 2008 at 3:51 pm
[...] This is just about the coolest analysis ever done of any MMO class. [...]
October 13th, 2008 at 9:25 pm
It’s… a bit disturbing to see how accurately the breakdown is reflected in my household, but it’s just so darn COOL to be able to learn about this sort of breakdown. Thanks a bunch for sharing!
October 14th, 2008 at 10:53 am
What is the sample size of the data you retrieved? While the charts and percentages seem all nice and your arguments for their support show that knowledge of the game is real but the editorials on the why of the data retrieved seems to take away from the statistics.
October 14th, 2008 at 12:59 pm
Really enjoying these statistics based commentaries on WAR. I, too, love the melee. I always play the rogue/assassin class if it is available, but I’m finding both the Witch Hunter and Warrior Priest classes enjoyable in WAR.
October 14th, 2008 at 2:51 pm
Definitely interesting to read.
If I can give some hopefully constructive criticism, a couple of things worry me about this research:
1) There’s no “Lie test” built into the gamer types that I can see. The questions seem to be “this or that’s” mixed up. How do you know this is reliable data?
2) Not enough diversity in the questions to get accurate results. I know for myself, I thought that the questions didn’t actually offer choices that reflected my preferences on a given subject.
I think my largest concern is that this is shaping gamer thought nowadays, as people love to have tools to categorize things. That can be a good or a bad thing. If the data gained is leading people to make assumptions about “The way things are” then it should either have a high degree of reliability, or a disclaimer stating it’s a self test and represents commonly held folk-psychological views on the subject.
October 15th, 2008 at 8:31 am
[...] hurting to do decent long-ranged dps, especially when melee closes the distance so fast. I saw a chart over at GamerDNA, where they did a survey and saw that the poor Squig Herders are the least-played class, period. [...]
October 20th, 2008 at 2:06 pm
[...] little while back the gamerDNA blog did a nice breakdown of how WAR classes correlate with how gamers do on the Bartle Test of Gamer Psychology, a widely used test that can break down [...]
October 20th, 2008 at 3:00 pm
[...] little while back the gamerDNA blog did a nice breakdown of how WAR classes correlate with how gamers do on the Bartle experiment of Gamer Psychology, a widely used experiment that can [...]
October 20th, 2008 at 3:02 pm
[...] Odds and ends, Blood Elves, ClassesA little while back the gamerDNA blog did a nice breakdown of how WAR classes correlate with how gamers do on the Bartle Test of Gamer Psychology, a widely used test that can break down [...]
October 20th, 2008 at 3:05 pm
[...] little while back the gamerDNA blog did a nice breakdown of how WAR classes correlate with how gamers do on the Bartle Test of Gamer Psychology, a widely used test that can break down [...]
October 20th, 2008 at 3:15 pm
[...] little while back the gamerDNA blog did a nice breakdown of how WAR classes correlate with how gamers do on the Bartle Test of Gamer Psychology, a widely used test that can break down [...]
October 20th, 2008 at 3:59 pm
[...] lowercase patch backwards the gamerDNA journal did a pleasant perturbation of how WAR classes correlate with how gamers do on the Bartle Test of Gamer Psychology, a widely utilised effort that crapper [...]
October 20th, 2008 at 4:02 pm
[...] Odds and ends, Blood Elves, ClassesA little while back the gamerDNA blog did a nice breakdown of how WAR classes correlate with how gamers do on the Bartle Test of Gamer Psychology, a widely used test that can break down [...]
October 21st, 2008 at 6:40 am
[...] little while back the gamerDNA blog did a nice breakdown of how WAR classes correlate with how gamers do on the Bartle Test of Gamer Psychology, a widely used test that can break down [...]
October 25th, 2008 at 2:01 pm
[...] little while back the gamerDNA blog did a nice breakdown of how WAR classes correlate with how gamers do on the Bartle Test of Gamer Psychology, a widely used test that can break down [...]
October 28th, 2008 at 7:06 am
[...] little while back the gamerDNA blog did a nice breakdown of how WAR classes correlate with how gamers do on the Bartle Test of Gamer Psychology, a widely used test that can break down [...]
November 5th, 2008 at 11:41 pm
[...] little while back the gamerDNA blog did a nice breakdown of how WAR classes correlate with how gamers do on the Bartle Test of Gamer Psychology, a widely used test that can break down [...]
November 26th, 2008 at 11:01 am
hi there!
xxoxo
I made on photoshop anime myspace banners.
have a look at them:
http://tinyurl.com/6mvtjg
Thank you for your site