The Dunning-Kruger Effect and Multiplayer Games

Let me describe a scenario that I think we’ve all been in. You pick up a game like Gears of War 3 or Starcraft II or the deck-building iOS game Ascension. You jam through the single-player campaign or do a little comp stomping in skirmish mode –maybe even on the second-to-hardest difficulty ‘cause you’re totally hardcore like that. And you’re better at the game than anyone on your friends list, judging by the local leaderboards and the way nobody will play with you anymore. You’ve got this game figured out, man, and you think you’re pretty good.

So you decide to venture online and try your hand at ranked ladder matches, a tournament, or maybe even just some pickup games via online matchmaking. You get creamed. Murdered. Owned. At the end of the match your competition has left you with a kill/death ratio in a realm of negative numbers so low that mathematicians hadn’t even bothered to really think about it yet because they figured nobody would ever use them. This baffles you, because by all previous accounts you’re totally awesome at this game.

Congratulations, you’ve encountered what psychologists call the Dunning-Kruger effect.

Named after the authors of a 1999 paper by Cornell University professor of psychology David Dunning and his then graduate student Justin Kruger1 the effect describes how those who really aren’t very good at something overestimate their skill while those who are experts tend to sell themselves short. The reason is that the more skilled you are in some complicated task2 the more you understand …that there’s stuff you don’t understand. Or that you haven’t mastered. Really good guitar players, for example, understand everything the instrument is capable of better than someone who has only now figured out how to bang out the beginning of that one Blink-182 song. Similarly, those of us who are really bad and inexperienced at a game often lack true understanding of what’s even possible. You can’t accurately reflect about your own opinion of yourself because you’re not good enough. And you’re not good enough because you can’t accurately reflect on your own opinion of yourself.

In their initial research, Kruger and Dunning gave students tests of logic, grammar, and humor (really, he had them evaluate the LOL potential of jokes from the likes of Woody Allen and Al Franken). When the researchers asked the subjects to guess at their performance on these tests, they consistently found that poorest performers overestimated themselves. Someone in the 12th percentile, for example, would guess that they were in the 62nd percentile. Further investigation showed that the poorer performing subjects overestimated their ability simply because they weren’t good enough to know how difficult the tasks were. And they didn’t know it.

I think I see this come up in video games a lot, especially ones with competitive multiplayer or even just those with challenges that let you compare your performance against others via leaderboards. It’s exacerbated by the fact that the single player versions of games often allow you to be incompetent in the pursuit of fun. You can soak up bullets in Gears 3 instead of using cover effectively or choosing the right weapon for the situation. You can brute force your way through a campaign scenario in Starcraft II using just Marines instead of appropriately countering the enemy’s army build. You can kite mobs around in World of Warcraft instead of using teamwork and assembling a set of equipment or list of perks with the optimal resistances. In each case, you’re frankly quite incompetent, but the limited feedback you’re getting doesn’t allow you to know it because you’re simply not that good at the game.

"LOL! Okay guys I played this level in the campaign! Just tell Liara to use Singularity until they're all dead."

Think of it in terms of known unknowns and unknown unknowns. I may not know the the melting point of Beryllium or the cooldown on an enemy mage’s frost bolt spell, but I know I don’t know that. But there’s also stuff I don’t know that I don’t even know exists or is a factor. Like what implications that enemy Sniper’s loadout has on my ability to sneak up and backstab him in TF2. The latter is the Dunning-Kruger effect in action. The great player sees every misstep and every missed opportunity for perfect play, and beats himself up over it. The novice bumbles along missing all that but getting the occasional headshot and thinks he’s doing all right for himself.

Some games are learning to address this fact by forcing novice players to learn the true scope of the game. Starcraft II, despite the fact that I’ve been using it to illustrate the Dunning-Kruger effect, actually tries to address it by inviting players to complete multiplayer-oriented challenges where they learn things like unit counters, defending against rushes, and other advanced tactics.This kind of thing helps, as well as curating of community guides and videos illustrating everything a game has to offer.

So next time you find your ladder rankings of your K/D ratio not living up to your expectations, take a second to reflect about all the things you don’t know and how your experiences so far may have been designed to make you feel more competent than you really are. Then go pick up some tips from those totally awesome hardcore players who know how totally awesome they really aren’t.


1. Kruger, J. and Dunning, D. (1999). Unskilled and Unaware of It: How Difficulties in Recognizing One’s Own Incompetence Leads to Inflated Self-Assessments. Journal of Personality and Social Psychology, 77 1121-1134.
2. Interestingly the Dunning-Kruger effect is more prominent for difficult, complex tasks and less so for simple, easy ones.

19 thoughts on “The Dunning-Kruger Effect and Multiplayer Games

    • I completely agree with you for dota. I am just getting into it and the depth of complexity of that game is just mindblowing.

  1. Pingback: | Superlevel

  2. If you only play “Halo” on Heroic, Line-Art Neil deGrasse Tyson just won’t respect you. Fortunately I finished the “Halo: Reach” campaign on Legendary (solo!) so that earned me some special pretend doll clothes to put on my little imaginary avatar. Sure, he may be slightly better at me than astrophysics, but I bet I’ve earned more imaginary doll clothes than him.

    Also I got the Achievement for making an Astronomical Katamari so maybe I am better than him at astrophysics too. That’s right, I’m calling you out to a Katamari challenge, Dr. Line-Art Tyson!

  3. I love the real world relation using psychology to explain playing video games. I’ve been a victim of the dunning-kruger effect. I do have to say though, getting owned in starcraft 2 may have a lot to do all of the map hackers n such on ladder.

  4. Pingback: Dunning–Kruger effect | Brodisse

  5. Pingback: Plato, The Apology | Grasping Ignorance

  6. Pingback: | Superlevel

  7. Sorry for the late response, but isn’t something like this also applicable to the difference between arcade games and those more modern ones with online multiplayer?

    Because I remember reading somewhere how quite a few people may have felt the same effects after going online in a game they’d previously played in a local arcade or group of friends. They were good at defeating the neighbourhood kids or a few mates in a local multiplayer experience, but failed to realise that didn’t translate in being good at the game in as far as doing well against people worldwide. Like how being a big fish in a small pond is fairly easy, but being a large fish in a large pond is a massive amount more difficult.

    Either way, it’s one of those things the internet made much clearer to gamers in general.

  8. Pingback: THose who oppose GMOs know the least about them , but believe they know more than experts | PesPro

  9. Pingback: Those who oppose GMO’s know the least about them — but believe they know more than experts -

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.