The distribution of wealth follows a well-known pattern sometimes called an 80:20 rule: 80 percent of the wealth is owned by 20 percent of the people.

This seems to occur in all societies at all scales.

It consists of N people, each with a certain level of talent (skill, intelligence, ability, and so on). This talent is distributed normally around some average level, with some standard deviation. [..] This is the same kind of distribution seen for various human skills, or even characteristics like height or weight.

The computer model charts each individual through a working life of 40 years. During this time, the individuals experience lucky events that they can exploit to increase their wealth if they are talented enough. However, they also experience unlucky events that reduce their wealth. These events occur at random.

The team studied three models, in which research funding is distributed equally to all scientists; distributed randomly to a subset of scientists; or given preferentially to those who have been most successful in the past. Which of these is the best strategy? The strategy that delivers the best returns, it turns out, is to divide the funding equally among all researchers. And the second and third-best strategies involve distributing it at random to 10 or 20 percent of scientists.

Fedi posts from first reading

The article doesn't say by what means "talent" does anything in the system. It says people get random good and bad events and then the most lucky individuals have the wealth. That's true but it's not informative (other than it apparently has the same 80/20 distribution they were looking for.)

I agree with their conclusion: it's best to distribute investments across multiple baskets. I don't agree with the method.

This conclusion is valid because of other reasons.

So I'm coming at this from reading Kiyosaki, Taleb, watching people play Cultist Simulator and hard life sims. One of the games from Kiyosaki is when he made a card game he called "opportunity." The game is based around you draw cards and they have events like you can buy a factory or not and other cards might say "if you have a factory, you can do this."

He observed people played the game and often complained that the opportunity to use something happened typically after they already turned down being able to do it, or they didn't have the money to buy it in the past so they missed out on other cards.

In other finance type advice you will see statements like "you only have to get rich once," or as Taleb less eloquently puts it "Fuck You money."

The oncoming of events is indeed random (what they simulate) but the talent for managing wealth is an understanding of the random nature and becoming invariant to it. to add yet another example you can watch x-com players.

X-com has similar risk management: you have to put soldiers at stake, which you may lose, and the losses have meta-campaign ramifications, so you have to carefully manage when you send the high quality soldiers or the middling disposable ones. And deal with things like "if i lose 3 90% rolls in a row, what do i do then?"

a common thread among ex. x-com, culty sim, and kind of hinted at by kiyosaki and others, are things like avoiding investments that would put you in to poverty if they failed. which is why the conclusion (distribute research funds evenly) is correct. it's not because of RNG, but because an intelligent risk manager understands they have to manage resources such that any one thing can fail 100% and it can't take the rest of the endeavor with it.

what makes being a smol so difficult is you may not be able to run multiple investments with reasonable returns, which exposes you to worse volatility, which means more likely to blow up and lose your savings with nothing to show for it.

some day i want to make an AI play Opportunity just to see what it learns to do with it. From what i've learned watching people in simulators, it's mostly about grinding hard to get out of the poverty line (which enables you to say "no" to toxic jobs) and then just not gambling all your wealth away from there on out.

Having a high IQ or being a great painter says nothing about your risk management skills, while risk management is basically the cornerstone of survival and solvency.

Whitepaper

• Talent vs Luck: the role of randomness in success and failure (Arxiv)