Psychochild's Blog

A developer's musings on game development and writing.

28 July, 2006

Why Achievers win and Killers lose
Filed under: — Psychochild @ 5:57 PM

I’ve posted this up in other places, but let me explain why DIKU-style games (e.g., EQ and WoW) dominate and why PvP games are doomed to failure. It’s actually very simple once you think about it, and the theories come from the respected Dr. Richard Bartle.

Read on for more insight.

You need to reference Bartle’s landmark paper Players Who Suit MUDs. Let us set aside the arguments about whether this is the best measurement for players or not: it exists and it matches the experiences of experienced game developers to a high degree. I’m aware of Nick Yee’s fine work on motivations, but he hasn’t worked out as many theories as Bartle has. (Plus, I slightly mistrust this assertion that Explorers don’t exist, I’ll explain that in a minute.)

The part we’re interested for this post is a summary of the interactions between the four major types. Here is a graphic:

(Click on the graphic to go to the original paper.)

(Personal to Richard: Yes, I know you have the newer 8 archetype model that adds another axis. Unfortunately, you don’t have complete information about how those interact. Or, at least, I don’t remember it. ;) You should have considered human memory: it’s easier to keep 4 things in memory rather than 8. Sorry, your new model was doomed from the start….)

The arrows indicate how each group affects the other. A green indicates growth and red means a decrease. The shaft color indicates the cause, as expressed by a change from the originating box. The arrow head indicates the effect as change to the box being pointed to. The thickness of the shaft indicates the size of the change. So, for example, the top arrow between Achievers and Killers signifies that a drop in Achievers causes a moderate drop in Killers.

Now, take a look at that graph. Consider that most of the larger games focus largely on Acheivers. Increasing focus on Achiever-style gameplay results in: an increase in Killers and a potentially slight decrease in Socializers. This matches what we’ve seen in these types of games: socialization has taken a back-seat to achievement, and there are always griefers in the game to disrupt the sheep, er, Achievers. Of course, most developers believe griefers are something to be eliminated, so we try to eliminate the Killers, which leads to an increase in Achievers and a HUGE increase in Socializers. Socializers are self-reinforcing, so that increases the Socializer population Thus a game focusing on Achievers can easily attract them (and Socializers) by limiting the number of Killers. Therefore, a game which focuses on Achievement and restricts the Killers will have a large population of of Achievers and Socializers, leading to high subscription rates.

As a side note, I think you could argue that one of the flaws in Bartle’s theories has come about with the large, modern games: a limit to the absolute number of people interested in Killer type gameplay. I think that there is a vastly large number of Achievers interested in online games than their are Killers. So, even if you have the biggest Achiever game ever, there is a hard upper limit to the number of Killers you’ll find in the game even without policies to discourage them.

But, let’s look at what Bartle says about catering extensively to Achievers:

Tilting a MUD towards achievers would make it obsessed with gameplay. Players would spend their time looking for tactics to improve their position, and the presence of other players would become unnecessary. The result would be effectively a single-player adventure game (SUD?).

Rather interesting when you consider the trend toward making our games more solo-friendly. WoW lead the charge on this, showing that a single-player game with other people present (someone has to appreciate the status you’ve accumulated) is the Achiever heaven.

Now, let’s look at what happens when you make your game PvP-focused: the primary effect is that the Killer population increases significantly. What does that do to the other populations? Well, it decreases the Achiever population, and nearly decimates the Socializer population. It can also decrease the Explorer population. Again, note that Socializers have reinforcing feedback loops and that a large decrease in Socializers will tend to send that segment of the population into a tailspin. Frankly, this is what we have seen with just about ever PvP-focused game: the killers move in and most other people move out. Eventually a lack of Achievers as targets will drive away the Killers, leaving pretty much nobody playing.

What does Bartle say about catering extensively to Killers?

Tilting towards killers is more difficult, because this type of player is parasitic on the other three types. The emphasis on causing grief has to be sacrificed in favour of the thrill of the chase, and bolstered by the use of quick-thinking and skill to overcome adversity in clever (but violent) ways. In other words, this becomes an arcade (“shoot ‘em up”) type of game.

Again, we see Bartle being correct, as most PvP fans seem to want these online RPGs to become more like typical shooters: less investment, less “building”, more going out there and demolishing people. He also seems to support my assertion that a game catering to Killers will tend towards low populations since the Killers need the other types to prey upon.

Now, let me digress a bit and talk about Explorers. Another big flaw in Bartle’s theories it that explorers have another factor: time. As a game ages, there’s less unknown to explore and find. In modern online RPGs, the role of the explorer has been reduced to exploring mechanics. Most of the content is explored very rapidly by intrepid Achievers (with a bent on Exploring) and cataloged very efficiently by various information sites about the game. These sites fulfill the traditional role of the explorer in old text MUDs where they would share morsels of information and know secrets the Achievers wanted to know. I think it would be hard to have a game that caters to Explorers, because they would consume content at a rate that makes Achievers seem laid back.

Now, don’t think that I’m happy about this situation. I love PvP-focused games, but there is the undeniable trend that they tend to launch, do okay, then fall flat. These games have fallen out of favor, with older classics like M59 and UO either fading to obscurity or radically redefining the game (Trammel, anyone?) in order to survive. But, it is grim reality that PvP games tend not to be as popular for as long a time as Achievement-based games. When you’re pouring money into a business, this is a hard reality to ignore.

So, what do you think? Do you agree with two of the brightest minds in online game development? ;) Or, do you think Dr. Bartle and I are out of touch with reality?


« Previous Post:





16 Comments »

  1. I see only one small problem with the graphic above. There is a lack of reinforcement arrows on the Killer box.

    Very similar to the way that Socializers will logg off if there are no Socializers to talk to, PvPers will log off if there is no one to fight. Anyone to fight at all, including themselves.

    As much as it pains me to admit it, I think the solution to creating a good PvP game is consentual PvP. As long as there are server wide rewards for good PvP (not just the server wide punishments for not being good, like there are in M59) then the model can work.

    If you can make the achievements available through PvP, and allow a way for the socializers to not be exposed to PvP, then I think a game could be made to appeal to all of the types. (Something similar to what GW has done. From what I’ve heard they’ve actually added a little bit of PvP to the game instead of all that arena silly stuff.)

    Comment by Warr — 28 July, 2006 @ 7:02 PM

  2. I want to interject a topic of discussion I’d like to see worked on in this thread: Eve Online. Open PvP areas and full loot drop deaths make this game seem awefully PvP-focused. Yet, there seems to be alot for achievers to strive towards. Maybe the real-time limit to training skills plays a part? Is Eve doomed to fall prey to the Killers eating up the game? Or is Eve not so PvP focused to avoid this fate?

    Either way, I’d like to see EO discussed in particular to this thread.

    Comment by Black Molly — 28 July, 2006 @ 7:16 PM

  3. There is a lack of reinforcement arrows on the Killer box.

    Well, that’s handled indirectly: most Killers like to prey on others, so a decrease in their main prey, the Achievers, will lead to a decrease in the number of Killers. Sometimes Killers go after each other, but that tends to be risky. Unfortunately, I think the type of “ideal PvPer” that plays for the sheer challenge against other capable killers is very scarce.

    I think the solution to creating a good PvP game is consentual PvP.

    Even if you put in rewards for consensual PvP, this doesn’t mean they will be used. Over the past few years we’ve introduced many different mechanics to encourage PvP in M59, including the mutual guild wars and soldier shields. As you know, Warr, these haven’t exactly been resounding successes. The griefers don’t like the systems because usually only capable people show up to fight in these types of scenarios; no more preying on the unsuspecting weak. These mechanics also tend to be gamed quite significantly: complaints about people killing mules to gain soldier shield status are common. As with most PvP systems, anything that can be exploited will be in order to gain even the slightest advantage. Keep in mind that the original Hunter vs. Necromancer scenario was supposed to be a PvP scenario, but people didn’t hunt each other according to the “rules”, mostly because there was no reward for doing so. It was the people outside of the scenario that worked to disrupt it, mostly so they could get everything reset and get one of the nice things for themselves.

    But, in theory, there’s a good market for this type of game out there: Counterstrike has had sustained popularity based on competitive player vs. player gameplay. But, translating this to a persistent world online game has proven to be, uh, non-trivial.

    My further thoughts,

    Comment by Psychochild — 28 July, 2006 @ 7:27 PM

  4. You’ve managed to pleasantly surprise me, Psychochild. You admit that you have a preference for a particular type of gameplay, yet you still weighed the reasoning and evidence before you and accepted conclusions that are contrary to that self-interest. Kudos to you for intellectual integrity.

    I think those who don’t see Explorer types in MMOGs are looking in the wrong place. They exist in abundance in MMOG betas. Or rather, since few players are all of one type, those (like me) with a decent amount of Explorer in them have their day in the sun during beta then, once the MMOG is live and there’s no exploring left, they switch to their other type(s): Socializer and/or Achiever (rarely Killer). Nobody watching them during the live game would know that they’re actually Explorers to a significant degree.

    Have you ever played in a game where all types have equal influence (though not, of course, numbers)? I’ve had that pleasure. Full PvP — anywhere, anytime, anyone, yet griefing never got out of hand. Socializers, Achievers, and Explorers thrived (the latter got new areas and easter eggs fairly often, unannounced and without hints from the staff — thus Explorers could keep them secret). And it was stable for years, not just for brief shining moments or in limited circumstances as in the anecdotes from old-school UO and elsewhere. When you’ve seen such a thing happen, you really realize just how beautifully Bartle’s player types theory fits.

    My two cents, for what it’s worth.

    Comment by Dellaster — 28 July, 2006 @ 9:24 PM

  5. Whoa. It has been a long time since I replied to a game design thread that wasn’t on Damion’s or TN’s site.

    Anyways… some mindless thoughts. First off, I wonder if the inconsistancies in Bartle’s player type resolve from the fact that some types may evolve into others. For example, socializers may lead into other types since most MMO games are based off of the social aspects of the internet (an assumption).

    Second, I wonder if Killers types don’t do well in traditional MMO games because they are geared towards the Achiever over the Killer. FPS multiplayer games seem to do very well catering to the PvP crowd. This is probably do the fact that one of the reasons one plays these games is due to the PvP, and thus every player is a Killer type in some fashion.

    Also why hasn’t the social systems haven’t evolved from their historical ancestors. We are limited to the basic chat system, IM, and e-mails. Since the communication is a base system that every MMO uses, why hasn’t it gone through a revolution. I am surprised that game forums aren’t in-game and haven’t been turned into a mini-game.

    Anyways as I said… mindless thoughts.

    -N. Johnston

    Comment by Nathan Johnston — 29 July, 2006 @ 1:17 AM

  6. While I can see this for most of the big graphical games of today, I wonder how much it really applies to some of the big text MUDs that are thriving today. For example most of IRE’s games have a heavy focus on PvP, but also many social and achievement related aspects. While those “griefers” can and do cause other types of players to leave, I don’t think it is as bad as this article makes it sound.

    In most of the big graphical games, achievement is defined by getting the best items, money and levels. What if there were other, non-material ways to gain achievement, which couldn’t be taken away by PvP?

    In Ilyrias we are designing a player run, and created, organization system that I hope will prove to be this other type of achievement. Besides just creating clans, our players will be able to create their own guilds/sects/political parties/etc all the way up to their own cities. The bigger, and more influential the organization is, the harder it will be to make. So cities, the largest type, will be extremely hard to make and require large groups of players working towards the same goal. No matter how many times you die that cannot be taken from you. Of course with that comes various other political positions which would appeal to both the Achiever and the Socializer.

    In addition to that there are trade skills, which will greatly effect the overall economy of the game. Want to expand your city, you’re going to need to hire some player carpenters and other trade workers to complete the work. Weapons and armour made, and designed, by players. Clothing, jewelery, houses, etc are all designed and created by players. Both Socializers and Achievers would enjoy these systems, as not only do they get to create but they can run it as an in-game business and achieve more money and status.

    PvP is a big part of the game design, and the main source of revenue. But with player run organizations, which benefits to be in them, players can help police their own. I’ve seen it work many times before. Player Bob gets into PvP and goes crazy killing everything that moves. This upsets other players, they retaliate with their big fighters, Bob dies. Bob gets better, and goes back to his old ways, other players not only retaliate on Bob but any organizations he’s a part of. This can spark wars, which can be fun, but often it leads to Bobs organizations telling him to calm down or be kicked out. If Bob gets kicked out, things will start to get boring for him and he’ll find some other game to terrorize. If he calms down, then he becomes one of his organizations fighters and protectors of their interests, a good addition to their group.

    So how is this model effected by other forms of achievement, and players policing their own? Perhaps it wouldn’t work in the games with millions of players, perhaps it would.

    Comment by Joseph Monk — 29 July, 2006 @ 2:21 AM

  7. I think EVE has set much of the standard here for interaction between Killers and Achievers.

    2 reasons.

    1. Grinding cannot and will not help anyone achieve their goals, so catassing and the domination of the lifeless teenager are removed.

    2. PvP+ and PvP- areas, rather than switches. If you don’t want to get involved in any trouble, you don’t go into the PvP+ areas. Better by far than WoW’s Battlegrounds. I think they actually missed a trick with the Security Rating system for Solar Systems, though. Graded levels of danger could be exactly what’s required to lead players gradually from one archetype to another.

    The explorer-type is also pretty well void except for exploring the vast and arcane mechanics of the game. I’d have liked to see less emphasis on Autopilot for Everything and a bit more on actually searching for yourself.

    Comment by Cael — 29 July, 2006 @ 4:31 AM

  8. Psychochild,

    How do you explain Eve-Online?
    It’s been slow burn, but anything but a failure, and the China launch will probably double the subscribers. (I know CCP themselves are wildly enthusiastic about it, but realistically..)

    Comment by Andrew Crystall — 29 July, 2006 @ 5:51 AM

  9. I haven’t played Eve Online, but to me, from the outside, it appears to be a stable Killer/Achiever game. Like DAoC. As Dr. Bartle wrote, that doesn’t mean there aren’t any Explorers or Socializers. They just aren’t significant influences.

    I’d be very interested in reading how Dr. Bartle categorizes the various MMOGs.

    Comment by Dellaster — 29 July, 2006 @ 6:59 AM

  10. Well, that’s handled indirectly: most Killers like to prey on others, so a decrease in their main prey, the Achievers, will lead to a decrease in the number of Killers. Sometimes Killers go after each other, but that tends to be risky. Unfortunately, I think the type of “ideal PvPer” that plays for the sheer challenge against other capable killers is very scarce.

    Honestly, I think that Achievers as the primary prey for Killers only occurs because Achievers are the largest group in most games. Killers are just looking for targets.

    As for your Ideal PvPer, I think that MMOG designers need to discover that a large percentage of Killers are Killer sub Achievers. The achievement that they seek is, of course, “beating their opponent,” however they define that. As long as there is a system in place to reward players for seeking out strong opponents while punishing them for attacking unworthy opponents (and yes, I realise that this is the amazingly huge and painful hurdle to overcome) the amount of ‘ganking’ that occurs should remain at a minimum. Game designers should be able to appeal to the secondary aspect of why people hunt other people in order to encourage the behaviors they want and discourage the others.

    As near as I can tell, the other major group of Killers is the Killer sub Socializer group. Admittedly their preferred method of socializing seems to be /laugh, /taunt, and /e whizzes on your corpse, but these are still social interactions. Honestly I have no idea how to combat this one. The social aspects of MMOGs are too foreign to me to really be able to comment on them.

    Even if you put in rewards for consensual PvP, this doesn’t mean they will be used. Over the past few years we’ve introduced many different mechanics to encourage PvP in M59, including the mutual guild wars and soldier shields. As you know, Warr, these haven’t exactly been resounding successes. The griefers don’t like the systems because usually only capable people show up to fight in these types of scenarios; no more preying on the unsuspecting weak. These mechanics also tend to be gamed quite significantly: complaints about people killing mules to gain soldier shield status are common. As with most PvP systems, anything that can be exploited will be in order to gain even the slightest advantage.

    Yep, M59 is an incredibly hard game to attempt to balance rewards for. Personally, I think the risk/reward ratio for the Soldier game is fairly out of whack, since the payoff of having the stat increases is reducing your build time a few hours and the risk is the very real possibility of death, which will cost you several hours to repair. However, I personally can’t think of a better way to do it that wouldn’t be inherently unbalanced.

    Well, unless you made it out of the realm of balance. If some pointless prestige marks were added to Soldier Shields then they could be a lot more popular. (Little stamps on the colored portions of the shield to show other soldiers that you had killed… kinda like the stamps that pilots used to put under the canopy of thier fighters.) Or if the Faction Leaders started putting up monuments to their greatest warriors. Neither one of these affects game balance at all, but appeal to Achievers enough to encourage them to play.

    Mule abuses in M59 are a large enough issue that I won’t even try to get into them here.

    But, in theory, there’s a good market for this type of game out there: Counterstrike has had sustained popularity based on competitive player vs. player gameplay. But, translating this to a persistent world online game has proven to be, uh, non-trivial.

    You’re absolutely correct. Of course the market has proven that fully instanced games can experience success. From what I understand from the (non-PR) guys at NC, Guild Wars is now doing well enough that it will stay afloat if they can pump out an expansion every 8 months (and they’re doing them every 6).

    And remember, if your job were trivial, everyone would be doing it ;-) (not that they don’t think they want to anyway)

    Comment by Warr — 30 July, 2006 @ 2:23 AM

  11. Planetside is small because it’s also the fairest PvP experience you can find outside of CounterStrike servers. In other words, yes, hunting the hunters is fine and great but most MUD/MMO “killer” types would prefer a certain win, such as that imparted by a level-differential or loads of twinked gear.

    There is a reason why most games have “carebear” residents who characterize the PvP enthusiasts as smacktalking 1337-kiddies -
    usually, they are. And yes, they like ganking newbs. Don’t believe me? I think AC1′s Darktide cluster is still live – try it.

    Planetside caters to PvPers rather than griefers. PvPers don’t necessarily lack enough from their lives that their fun comes from making somebody else miserable. It’s unfortunate that DIKU and level-play have tended to mean that in most games, PvP is exactly the best way to make someone else miserable.

    That’s the main reason i dislike the Bartle System of Classification. You can be a PvPer without being a griefer. And i dislike being associated with those whose idea of social interaction is playground bullying in VWs.

    Comment by Cael — 31 July, 2006 @ 4:46 AM

  12. I’ve never been a hard believer of the “taxonomy” but I really think that we’re putting too much emphasis in labeling PvP as only a “killer’s place.”

    PvP gives the most obvious path a killer can take to aggravate others, but there are plenty other uses for PvP that work for all the other player motivations. The problem is that the code frequently caters to other expectations or the developer seems to decide that the PvP zone/server is the “killer’s sandbox” and doesn’t cater to the other players as much.

    Roleplay (also not just a socializer’s sandbox) drama often has interplayer drama, and could make considerable use out of a combat system. Roleplayers aren’t playing to optimize leveling, though, so they’ll often be ill-equipped for the interference of an achiever or killer armed for bear.

    A “self-made-faction” system would allow roleplayers to organize their own sides and their own rivalries, with leaders capable of removing participants that aren’t playing fair. This is similar to the “guild war” mechanic some games offer, but socializers often interact beyond their guild, and this often confines participants to the “entire guild or nothing.” The faction could be long-lasting or just set up for a little in-the-streets one-night riot.

    Such a system could be of obvious use to the achiever, who can set up his own rules and have competetive matches where unfair or unsportsmanlike conduct could be expelled. These could evolve into leagues, with championships and prizes.

    Both of these allow a degree of “self policing” that can make things tough on the PvP-centric killer (who could use PvE “interference” griefing to get some of his kicks). A bad reputation gets one bumped from a group- or never invited to one. Of course, the killers could build their own group, where everyone accepted killing tactics and hardcore attitudes, but what fun is it doing this to people who enjoy that kind of play?

    Comment by Chas — 31 July, 2006 @ 7:06 AM

  13. “Roleplayers aren’t playing to optimize leveling, though, so they’ll often be ill-equipped for the interference of an achiever or killer armed for bear.”

    That’s something else I like about Eve, actually. Several of the RP factions, and one in particular, are very highly organised PvPers with a substantial industrial prescence and can deploy bigger fleets than other alliances several times their size… ships JUST from their RP race (of 4 total), which they use vey effectively.

    Psychochild,

    Eve has been slow-burn yes, but it hasn’t had a publisher afer its first few months either and its development budget was tiny compared to most of the competition. In a 5-year to-profit model for a MMO, it can be argued that slow burn is preferable to fast burn, even. Fast-burn games population stablises and then decreases, sometimes quite quickly.

    Comment by Andrew Crystall — 31 July, 2006 @ 10:40 AM

  14. I think there is K in all players. Dark Ages of Camelot hit on that K in all their players by tying it to advancement. There is answer in there somewhere that I need time to work out, but I think Bartle’s graphic still holds true. All killer and no filler games will flop, but the careful blending can lead to a great game. And for some reason I think Mythic may be on to it with WAR.

    Comment by Heartless Gamer — 31 July, 2006 @ 5:17 PM

  15. HeartlessGamer-

    I’m SEA on the Bartle scale- literally no K. About 75% each Socializer and Explorer, 40% Achiever, and no Killer instinct at all. I think it might be connected to the fact that outside of RPG, I am mostly a puzzle gamer- very non-hostile and fairly non-competitive. Mandatory PvP is a quick way to get me to skip a game.

    Personally, I’m a plot fiend. Combine the Social urge with the Explorer urge, and I run through plot like it’s going out of style, then get very bored when I only have mechanics to screw around with. If I haven’t hooked into some player-driven world dynamic by the time I run through official plot, I’m done. So far, I’m disappointed with the few MMORPG’s I’ve tried. No underlying plot, just a huge mass of mini-quests. Bleh.

    Comment by Firefairy — 17 February, 2007 @ 9:21 AM

  16. What is missing in our games?

    [...] Tesh’s question does bring up most contemporary games hyperfocus on the Achiever motivation and the consequences of this focus. If we remove the Achiever motivation, that being the loot and level upgrade tracks, [...]

    Pingback by Psychochild’s Blog — 12 October, 2009 @ 1:56 AM

Leave a comment

I value your comment and think the discussions are the best part of this blog. However, there's this scourge called comment spam, so I choose to moderate comments rather than giving filthy spammers any advantage.

If this is your first comment, it will be held for moderation and therefore will not show up immediately. I will approve your comment when I can, usually within a day. Comments should eventually be approved if not spam. If your comment doesn't show up and it wasn't spam, send me an email as the spam catchers might have caught it by accident.

Line and paragraph breaks automatic, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Email Subscription

Get posts by email:


Recent Comments

Categories

Search the Blog

Calendar

April 2017
S M T W T F S
« Mar    
 1
2345678
9101112131415
16171819202122
23242526272829
30  

Meta

Archives

Standard Disclaimer

I speak only for myself, not for any company.

My Book





Information

Around the Internet

Game and Online Developers

Game News Sites

Game Ranters and Discussion

Help for Businesses

Other Fun Stuff

Quiet (aka Dead) Sites

Posts Copyright Brian Green, aka Psychochild. Comments belong to their authors.

Support me and my work on