[jjoonathan]

> But if you get lucky because you stumbled on something great, that's not science.

No, that's exactly how science works. The consistent delivery of predictable desired results is called engineering, not science.

The rest of your post revolves around this misunderstanding, so I won't insult you by repeating myself.

[maratd]

> No, that's exactly how science works.

No, it doesn't. There have been an ample number of scientists that have produced spectacular work on a consistent basis. No luck involved. And they have been highly valued.

What you don't want to admit is that scientists are like basketball players when it comes to how they're valued. You're either a superstar or you're worthless ... and being really good is equivalent to being worthless. You need to be spectacular.

I'm not trying to offend, that's just how the market works. I don't see anybody complaining about the market dynamics of athletes.

[jjoonathan]

First of all, these are two distinct claims:

1. The market undervalues science and scientists

2. Science is sub-optimally close to the "extremistan" end of the "extremistan-mediocristan" spectrum

I believe both (why do you think I would "not want to admit" #2?), but the claim I have been making is #1. The existence of well-compensated scientists is not evidence against #1 because #1 is an aggregate statement and #2 should guarantee the existence of well-compensated scientists regardless of #1.

Now let's talk about #2.

> There have been an ample number of scientists that have produced spectacular work on a consistent basis. No luck involved.

Examples?

Before you try to p-hack this one, let me specify the null model: scientist as prospector. Like gold in the ground, ideas cluster. Skill and resources determine your rate of sifting through earth, but at the end of the day initial positioning matters a hell of a lot. The founders of a new field are all but guaranteed decades of consistent productivity, but in order to maximize the amount of gold dug out of the ground it still makes sense to devote considerable resources to surveyors who haven't yet hit paydirt.

What would I see as evidence of the "skill" view? Einstein before 1906. People who are able to consistently make breakthroughs.

What do I see as counterevidence? Einstein after 1906. An overabundance of one-hit-wonders. Hundreds of theoretical papers in the arXiv that make the construction of GR look like a rite of passage rather than a singular achievement -- with the exception that Einstein happened to be right where they happened to be wrong or useless (it's not like either could tell ahead of time how things would turn out). Epistemology: correct useful ideas are very, very sparse in the set of plausible ideas.

> I don't see anybody complaining about the market dynamics of athletes.

Doubling the number of competent-but-not-superstar basketball players would add little marginal value (more games, maybe).

Doubling the number of competent-but-not-superstar scientists would about double the marginal discovery rate of science, assuming the scientist-as-prospector model.

[maratd]

> Doubling the number of competent-but-not-superstar scientists would about double the marginal discovery rate of science, assuming the scientist-as-prospector model.

Right. I reject that assumption. Because science has nothing to do with luck. You provided zero evidence to suggest that it does.

I would posit that there is substantial skill and talent involved in properly positioning yourself to be one of the "founders of a new field". It has nothing to do with luck.

Since it has nothing to do with luck, doubling or tripling the number of individuals who don't have the capacity to properly position themselves changes nothing. It just wastes resources on useless fields that produce nothing valuable.