[Guide] Dunning-Kruger effect in Planetside

>People are ignorant ********
Welcome to society!

The "average" player will change over time based on population and experience levels. If an influx of new players comes to PS2 because of the PS4 release and some marketing, then the "average" player will be less skilled and/or less experienced.

Whereas if PS2 stops advertising and the player population as a whole stops growing, those playing will still gain experience and skill at the game, so the "average" player will get better... just on pure experience gained.

Statistics based on win percentages, kills/deaths, score per minute all change based on the amount of experienced players playing.

A "scrub" is a scrub is a scrub. It's just a derogatory term for someone who is less skilled than average and not likely to get better. "average" can move... so therefore who is a scrub will also move in relation to it.

Thats about right, and the overall "average" in online gaming is definitely dropping to "scrub" level. TKing is done for fun and because you didnt "play how they wanted you to". Oblivious playing, drunk driving, cant fly, cant aim, no teamwork, poor tactics, horrific awareness, no suppporting each other......thats the average gamer......

Those who manage to break out of the scrubby "average" rise to become atleast.....tolerable to be around and play with.....but the current "average" player.......omfg...

Well, K/D and W/L aren't quite so flexible in terms of averages because they both approximate zero-sum stats (for every kill there's a death, for every win there's a loss). If the game lacks a "draw" condition, or draws are not counted in stats, then the W/L average is absolute: 1/1 (50% win rate). If there is a draw condition, then that depends on how you count draws in the average.

If the draw counts as a loss for both sides, or you calculate simple win rate by taking wins/matches, draws invariably drag the average down. If draws count as a "win" or "fractional win" for both sides, then they can drag the average up (because it produces a "win" without a corresponding "loss", breaking symmetry).

K/D is not quite as zero-sum as W/L due to TKs and suicides usually being more common than draw conditions, though how it is calculated means that the mathematical average has a hard cap of 1/1. This is because in the absence of TKs and suicides, every kill produces a death and you cannot do better than that in terms of global average. TKs and suicides can only drag the average down.

This also means that, for the population in aggregate, average K/D and average W/L is really just a measure of how common TKs/suicides and draws are respectively.

The W/L part is assuming either a two faction game, or that all matches are contests between two out of however many factions you have. If you have multiple factions (for example, three) and all of them compete simultaneously (for example, in alerts), then any time one faction wins all the others, by default, lose. In a three faction game, this means every win invariably produces 2 losses no matter who wins.

So in a multi-faction simultaneous competition, the global W/L average cannot exceed 1/(n-1), where "n" is the number of factions, by definition. Because every win automatically produces n-1 losses.

The only way to change that would be if you didn't have a zero-sum win condition (ie multiple factions could "win" at once).

TL;DR: Due to how math works, zero-sum statistics have little to no variance in their average. Statistics that are not zero-sum, such as score or accuracy, are capable of varying with the quality of the player base. However you should be aware of interfering factors, such as changes to the scoring system or weapon stats, or simply players using different weapons.

I also have to disagree. Other people have already said this, but the complaints often aren't the Dunning-Kruger effect.
If PS2 would have a suit that would cost 1 cert and 1 nanite to buy. The weapon is an instakill weapon and your armour reduces 99% of all damage. All teams have access to it, but would people like it? Same goes for real cheating programs. Everybody has access to them. Using them feels not fair.
The Dunning-Kruger effect is not present here. They don't necessarily lack skill. They can't judge the skill of the other, as that is being masked behind a low skill setting. Their 'inadequacy' is choosen on moral grounds. They feel that many things are cheating, making them avoid this setup.
While I agree that they do suffer from this effect for some things (example: shotgun usage), sometimes there is a real good reason not to do some tactic.
Redeployside is available to everyone. Crushing people with 3 times their numbers is a valid tactic. Is it lame? Yes. Is it worth avoiding? Yes. Is it a cognitive bias? No. Just a way of playing that you think is fair.

The pawn only chess game does not seem like a great example to me.

Suppose you are playing rock paper scissors and you use a strategy that you know works 2 out of 3 times against a person with another strategy. You pick randomly paper and they pick scissors. It's not illusory superiority to think that you were not outplayed and your strategy is still better if you then play 30 more times and win 20 of those times.

I'm sorry, this video seems to be comparing a game of deep strategy and forward planning played for 500 years with a shallow ephemeral video gaming genre title with absolutely no gameplay outside of TDM on a tiny map in a big map with gimped 1HK weapons?

And then passing it off with some psychobabble that that infantry player that just got instagibbed by the other player in their 1HK aircraft/vehicle/Max suit/100 round gun is better than them and it has nothing to do with broken gaming mechanics?

Nice try for attempting to raise the intellectual level of the argument of what is wrong with this title, epic fail for trying to explain away the higher RPM/3 times the number of rounds/one hit kill versus 15 hits to kill.

You want to get intellectual and win your argument - give everyone the same gun and get rid of aircraft/max/vehicles.

THEN you can argue that 2 EXACTLY matched players can't complain about who lost as it comes down to aiming.

And I'm not even going to get into latency exploiters.

Looking past the variable nature of "average skill" of a server and what it actually means (and what stats you can use to measure it)

While the zero-sum stats are pretty fixed, the interpretation of those stats can be skewed.

This is an issue when using historical stats to evaluate a current playerbase "skill" level and changes to that "skill" level over time. Changes to the game, changes to the population, changes to XP gain, adding of vehicles, playstyles, hell... even things like not recording a death if a medic revives you. (which is true in game, but not out of game stats, etc)... It makes for a nice little complex system to try and interpret.

I don't believe there is a single stat in the current game that hasn't had its foundations moved in some way or another making it extremely hard to use in comparisons like these. That's where all the stats arguments break down.... identifying and making subjective judgements on a players "skill" based on the objective stats given by the game.


For example... say for one week in the Month of November a youtube personality comes to PS2 and has 1000 followers play. They all come onboard, have low-Br guys and get wiped out... all getting 1 kill and 10 deaths each.

Thats an influx of 10000 deaths and 1000 kills... then those players stop playing. You can see how that would skew the active veteran stable populations K/D ratios for the month, the same K/D that was being used by some as a measure of "average skill" of the server.... How would you interpret the "average K/D" of the server in that instance. It would skew the average, mean, median, etc.

I'll Dunning-Kruger my foot up your ****.

I don't want SOE to balance weapons? When did I say this? Are you capable of discussion without making **** up and attributing it to me?

"You wouldn't understand the Dunning-Kruger effect, because your mind is so ripe with symptoms of the Dunning-Kruger effect that you Dunning-Kruger'ed your entire argument, making this a classic example of the Dunning-Kruger effect." - Liberty

E Z P Z

Hilarious how you're still salty about Polystigma destroying you to the point of you getting a shotty just to be on even footing with him. Anyways, wasn't this like 2 months ago?