Bartle Quotient
or "Pigeonholes for role-playing gamers."

"The primordial image, or archetype, is a figure--be it a daemon, a human being, or a process--that constantly recurs in the course of history and appears wherever creative fantasy is freely expressed. Essentially, therefore, it is a mythological figure. . . . In each of these images there is a little piece of human psychology and human fate, a remnant of the joys and sorrows that have been repeated countless times in our ancestral history"
    - Carl Jung (1875 - 1961)


The so-called Bartle Quotient is the result of research which was first discussed in a paper by Richard Bartle called Hearts, Clubs, Diamonds, Spades: Players Who Suit MUDs1, and is quite simply an attempt to partition MUD players into four distinct gaming archetypes: Killer, Explorer, Achiever, and Socializer. Every player contains at least part of each, but most lean heavily toward one or two of these four.

It is in the definition of the four archetypes and their various interactions that Bartle discovers and explains reasons why MUDs succeed or fail based on both game and interpersonal mechanics. The balance of certain player types is critical to the success of a MUD of any kind, and anyone who runs, writes for, codes on, or plans to do any of these should read the paper.


In January of 2000, Erwin Andreasen2 of Denmark launched a web-based test to measure the Bartle Quotient. Based on input from the MUD community, he patterned it--loosely--after the Kiersey Temperment Sorter (a watered-down Myers-Briggs personality test). It consists of 30 vaguely randomized questions with two answers each and you are asked to select one of the two answers for every question based on which action you would be more likely to take in a MUD or MMORPG.

The problems inherent in this system are many. For one, the test is heavily weighted toward the explorer archetype, which Bartle called the rarest and most valued type of player. In Andreasen's test, 29% of all players (leading the pack by 4%) are primarily explorer players, which is a ridiculous overstatement. Also, the Aristotelian "A or not A" questions often leave test takers with two options of which they would choose neither for the situation at hand. Obviously a larger question pool with more answers per question would make for a more accurate test, but for the moment Andreasen's test is the only one around.


One of the interesting things that Andreasen's test does is link player results to the games they play (using a MUD/MMORPG database derived from Mudconnector3), so that certain types of players can find muds which are populated by others of their kind. While it is a nice idea, there are two problems with it. The first being the inaccuracy of the test itself which renders the resulting statistics largely meaningless. The second--which is potentially quite important--is Bartle's findings that all MUDs require a certain balance of player types and that an abundance or deficit of certain kinds of people will likely destroy the game. Therefore if large quantities of players of a certain type flood a game it could create serious imbalances in the player population, which might eventually lead to its ruin.

Also interesting are the overall MUD results, which show an average of all players from a given game who have taken the test, and thereby gives the MUD itself a Bartle Quotient. There is also a Kiersey personality test with cross-references between player types and personality types. These are fun if you enjoy statistics, but also meaningless as well.

In Closing

The Bartle Quotient, as quantified by Andreasen, is a fine amusement for MUD/MMORPG players, but the flaws inherent in its present implementation make it useless for any real purpose. If a more robust and accurate version of this test were devised, the data could be used by game designers to enhance the experience of their players, but for the moment it is merely a nice diversion. The test and results can be found at

Update April 2008: Apparently Andreasen's site has been down for some time. The new location of the test is at

1 The Journal of Virtual Environments. Issue 1.