Greedy Goblin

Tuesday, January 3, 2012

Why 20% owns 80%?

It is a well-known phenomena that the top 20% of the population owns 80% of the resources, while the other 80% of people has only 20% of resources. This is often mentioned as clear proof of the "evil oppression" of the capitalist system.

In my blog I want to research and teach capitalist thinking through games that we play. They have the benefits of being known to people and being fair: you can't blame my score on the wealth of my family, as socialists always do on real life examples. This time the very important question: "Why 20% owns 80%?" will be answered trough a game, World of Tanks.

In this game 15v15 random matches are played between teams of equal strength. At the end one team wins by killing all opponents or taking the base and a result screen appears:

As you can see there are many players with no kills at all, while others have 1, 2 or more. I processed the results of 60 matches that I played with various tanks. There were no selection of results, after I started screenshotting the result pages I saved the next 60. 1800 players(60x15x2) appeared in the results, of course one could appear more than once, obviously myself appearing in all. I made the simplification of ignoring this and considering all 1800 players different.

During these matches 1332 kills happened, making the average kill/player/match being 0.74. Common sense would say that most players had 1 kill, while a smaller bunch had 0, and a few had 2 or more. Well, "common sense" should be called "social sense" as it comes from our prehistoric subroutines. The truth is very far from that:
  • 58% had 0 kills
  • Only 23.5% had 1 kill
  • 10% had 2 kills
  • 8% had 3 or more kills.
Now let's calculate "kill ownership":
  • The top 10% made 45% of the kills
  • The second 10% group made 25% of the kills
  • The third 10% group made another 13.5% of the kills
  • The bottom 70% made only 16.5% of the kills.
Maybe it's just luck! - the social would say. Well, luck provides normal distribution. I calculated the mean (0.74) and the standard deviation (1.12) of the data and using these I plotted the kill distribution according to the normal distribution (0 point: F(0.5), X point: F(X+0.5)-F(X-0.5)), along with the measured distribution:
As you can see, the two curves have very different shape. It's not luck, nor normally distributed player skill. Compared to these, there are way too much "poor", less "middle class" and more "elite". The difference is largest in this region. According to the normal distribution, only 0.6% should have 4 or more kills, while actually 3.3% have such results.

So the short answer is: the bottom 80% has only 20% because they are filled with "0-score" losers, who are just that bad and deserve no better.

Remember, my results came from a game where every match is a new round and previous results have little effect on this match. You can farm points with losses, so no amount of sucking can stop you from having equally strong tank as the best players. Still the top 20% had 70% of the "wealth". Since in real life, past idiocy (skipping school, crime) has serious effect on today, 80/20 can be considered a very modest and nice result.


Andru said...

What's more concerning is people in the 20% who actively support people in the 80%.

Strictly speaking the "We are the 99%" of OWS is a gross mistranslation, and that's why I have a problem with it.

When someone tells me "We are the 99%" I reply: "No, you're fucking not. You're the 99%-80% interval. You are top 20% and arguing for the people who are too dumb to even protest. You are being used by the morons who bought flat screen TVs on credits, who bought toxic papers on credit, and you are blaming the banks for what the morons did."

That's like blaming the raid boss for wiping the raid, when half of your raid is dancing in the bloody fire.

That's not even social thinking. I...I have trouble even fitting that kind of thinking into a paradigm.

chewy said...

You've compared two systems the world as we know it and an online game of tanks. The conclusion you've come to is that the none specific ratio of wealth to population matches the capability of 1,800 players to kill tanks.

Excellent and from this you want me to believe that uncontrolled capitalism is good ?

Prasejednomalo said...

While I normally enjoy your analyses, you have two elementary problems in your assessment here. First, the WoT matchmaker assures that at least 33% of tanks in a given battle has no chance to destroy at least 33% of tanks (bot tier tanks vs top tier tanks).
Second, the bot 33% only has a chance to destroy tanks similar to them, and the chances of encountering similarly low tier tanks before they die is also 33%.
Third, in every battle, the top 33% can damage all the tanks in the battle.
Fourth, you did not count damage done. Without counting damage done, you don't know whether people are kill stealing, and the current version of match reporting encourages kill stealing (because damage done is shown nowhere).
As such, the WoT data you presented do not support your 20%/80% conclusion (though some other WoT data analysis might).

Anonymous said...


If ppl are morons for buying what they cannot pay what is the banking industry as a whole when it actively pursued it's own "downfall"?

I'd say the suit&tie morons who fancy themselves the elite are the primary marks and scapegoats.

Ahtchu said...

+ 1

I love people who villainize the top earners. Those gifted with understanding just so happen to understand that the villainizers are nothing more than M&S (thanks for this term Gev) voicing their greed which spawned from their nature of entitlement.

Bobbins said...

The problem you state are of 'the 20%' acting as morons. The bankers lent money to people who they knew couldn't pay it back, toxic loans were actually bought by the banks, government spent money they didn't have etc. These people had to be bailed out by the taxpayers.

The allocation of resources with our society is wrong. We reward idiots for failure and worse.

It is crazy that Andru blames the poor for the banking collapse while ignoring the people who were paid millions of pounds who lent the money.

Riptor said...

@ Bobbins:
There are people with 25k/year income who want to buy a house worth 500k.. You can tell them to their face that, due to the low income und thus, low return you would have to ask them for 10% interest in the loan. You also let them know, that if the Housing Market should shift you will adjust the Loan % accordingly (down to 5 or maybe up to 30..). Then you also let them know that should they fail to pay monthly, you will take your loan/the house back... If you find someone dumb enough to sign, why the hell wouldn't you do it? It is the Goblin way. If they somehow manage you will get a large return on your investment and should they fail, you get a House on top of the Payments that were already made.
Or you could sell a TV for 1200 today, or get monthly Payments of a total 1400 for the same TV.
Or you have someone getting an education in expressionist dance, ancient Aramaic Literature as well as Chinese brush art. This “Education” takes 4-6 years and costs 250k while it is clear that the well paid corporate jobs (the ones that give a salary high enough to pay back the 250k) are nonexistent for people whose sole skill is that they can figure-dance the Aramaic Alphabet.
You see where this is going? If someone offers you a share in an Investment Fund that generates 10-15% revenue (while all others generate 1-3%) and you are stupid/greedy enough to put all you money in that, you are plain and simple an M&S and have probably met Bernie Madoff somewhere along the way, congrats…
Andru and Ahtchu are right, it’s obnoxious, entitled, stupid morons that cry and moan when they are asked to pay for their stupidity.

sha said...


Since you were there every game, what were your kills/gm average and kills in each game?

If we are to assume you are in the top 30% skill level based on your argument(call the top 30% people that try and know what they are doing) we should see higher than average kills/gm.

I don't play world of tanks but if you are in the proper bracket, I would imagine you would go from 0 kills sometimes where you are the worst player and 5+ when you are the best player in the group.

That could also explain the 20% getting 80% of the kills because they are that much more powerful relatively to others.

Overall i do agree with your argument, but you do have to remember there is a crossover point with wealth. It is much easier to make $1000 with $10000 than it is with $100.

Goodmongo said...

Thing is I support TRUE capitalism. What happened in the US was NOT capitalism. If it was then AIG, Fannie/Freddie and many banks would have gone under.

True capitalism is a risk/reward proposition. Bailouts meant that there wasn't much risk. See the OWS folks blame the 99% and want government. In fact they are mad that the goverment spent the money on banks instead of giving it to them.

The Tea Party in the US want limited government, no bailouts and real capitalism.

Prasejednomalo may have a point about WOT, but too many other studies have proven the 80/20 split. And this was just one more example that some people just do better then other people. No maybe the top 20% in WOT are complete financial morons but the theory holds.

Anonymous said...

I guess that, in any level playing field, competent people will do significantly better than idiots.

However, some truly competent people choose to spend time ensuring that the systems that allow the accumulation of value (government) continue to function.

Some of those people (I am among them) believe that the accumulation of vast wealth by a small percentage of the population leads to government being tailored to permit the accumulation of vaster wealth in simple ways* by that elite. *F'r, instance, US Gov't lends money at 0% interest to investment bank...investment bank buys treasury bonds at 1%...not exactly hard. One side issue is that this results in inefficient allocation of resources.

The main point is that, long-term, the _easy_ accumulation of wealth by an entrenched elite results in the rise of an elite that is populated by idiots...and the eventual fall of the government, collapse of the economy, and revolution.

So, self-interested people interested in long-term stability have an interest in some of the goals of the 99%. The trick that people seem to forget is that social controls (police) are inherently very, very weak - so an avalanche of idiots led by a few moderately incompetent people would win a civil war. In most regards, the social controls in Egypt, eg, were comparable to those in the USA and a bunch of unemployed students toppled that government.

Goodmongo said...

In most regards, the social controls in Egypt, eg, were comparable to those in the USA and a bunch of unemployed students toppled that government.

Do you really believe that some unemplyed students did that? Or peseant farmers led the Russian revolution? All those students were just the useful idiots that the real powers behind the scenes used to get their way.

Synkage said...

This is a largely pointless post, and this is coming from someone that typically likes reading the articles on this blog.

Using a video game based off pure performance to model a real world economic system isn't logical.

The rationality of this test fails foremost where you've eliminated all possibilities of that excuse you hate so much - "socialists" blaming the wealth of your family. In no way is this modeled in this game.

You spawn derision towards isolated incidents in which people that cannot propel themselves to a higher class bracket in the same manner that you idealize people who may have actually worked to reach a level of success.

Now, if one person spawned with a tank that carried a deployable ICBM, a chassis and body constructed of some high quality alloy, and an rail gun versus a few tonka trucks, it would have merit.

It's simple to blame people for laziness and failing to progress of their own volition instead of blaming the variables that may hamper their progress.

You say "competent" people assuming that people become competent of their own volition. You do not simply open up a book and become competent about life decisions and business choices. You become competent because someone else is there to assist you in achieving competence. The constant bashing of the lower per capita of the population in this thread only hides the lack of depth or brevity in this project.

Anonymous said...

You've basically just pointed out that a particular trait (in this case tank killing, or anti-social plotting) is apparent in a relatively small portion of the general population, and in an activity that rewards that trait, those people will be rewarded and the others will not be.

Well that's a bit of a tautology. Of course that is true.

But the critics of capitalism would contend that there are lots of traits in the human experience and choosing a society centered around a system that rewards mostly one or two traits (anti social plotting primarily) disproportionately isn't a great way to run things.

Azuriel said...

So the short answer is: the bottom 80% has only 20% because they are filled with "0-score" losers, who are just that bad and deserve no better.

Oh, my. The naturalist, "they deserve to be poor" argument.

As mentioned by others, your analysis is asinine here, and actually supports the counter-argument. Do you believe that 58% of the WoT players are completely useless? I'm not even talking about value from being a distraction or whatever - I mean that they did zero damage to the opposing team. If you are not absolutely certain that ALL of them are, then we are automatically talking about a system in which the rewards (kills in this case) is slighted in the favor of some over others.

When you look at real wages in United States since 1964, it has actually gone down despite business being as profitable as it has ever been. The value of one's work has increased, but said value is extracted in higher proportions than ever before. It is the opposite of "being 0-kill losers," unless someone is a loser when they have their tank kill stolen.

You are drawing an incorrect parallel between capitalism and merit/effort. Show me someone living from paycheck to paycheck then becoming a CEO via hard work, and I'll show you 10,000 CEOs who do less than any one of the janitors they employ.

@Andru: The 80-95% make an average of $97k/year. Do you think those are the people out in the tents? The 60-80% make $56k. I find that the more likely group, as the 40-60% make $34k and aren't likely able to afford to take months off from their decent, dead-end jobs to go camping.

The chart I looked at was from 2001 though, so those figures are likely outdated. Not in the direction supported your argument, though.

But, yeah, those unwashed masses swindling those poor bankers who made bets against their own loans. It's not as though the bankers have a fiduciary obligation to check the ability of customers to pay back loans or anything.

Anonymous said...

I don't feel that you have succeeded in adequately answering the 'why?', just demonstrated the phenomenon once more.

I would really like to see a derivation of this from more basic principles, because this is really important.

Frostys said...


While I agree the top "own" most of the "work" done in game, the stats take are shallow in meaning. The killing blow on an enemy tank should not be worth 100% of it's value unless it was solo'd. Many tanks are being destroyed after taking shots from many opponents.

Playing as artilery, the kill count is extremely volatile even if played well because of the innacuracy of the guns and the long reload letting you deal awesome damage to a tank while having a medium get in for the kill. SOmetime you will one-shot stuff while otehr times, you will wreck 4 tanks in the same game from 100% TO 5% and finish with 0 kills.

Some player definately do heavy lifting and probably have really high damage/games but we don't have easy access to those stats unless you were to dig up many player profiles.