Ok. But since you’re walking a distribution if you keep applying that without insight into how much smarter, just smarter, you’ll understand the tail.

Identify A as smarter than B

A identifies A’ A’ identifies A’’ …

There’s no need to know the quantified smartness, just the ordinal smartness.

> and solves crosswords faster

I like this example, especially, as we're all aware, you can get better at both crosswords and IQ tests with practice.

> I dislike the notion that some individuals are off the scale on cleverness.

Why not? You accept that some individuals are off the scale of various physical abilities. That try as you might, you'll never be the fastest or tallest or strongest.

What makes intelligence different? Why should we believe that everyone is roughly equal in intelligence?

If you are mediocre in obvious ways, what's left? Everyone wants to be good at something. Everyone wants to be better at something than others. But it is fairly obvious with anything with a physical component. I am a mediocre guitar player. I am not tall. I am not fast. I am not exceptionally strong. These are things that can be sussed out fairly quickly with some basic tests.

But intelligence. Well. That's not only fluid, but it's also abstract. What does it mean to have an IQ of 129 versus an IQ of 130. Especially when one could score both depending on how much sleep one has had. So you can just say IQ is bullshit. Intelligence don't real. Everyone is smart in their own way.

No. That's just simply not true. Yes, one could test differently based on conditions. However, it will fall within a range. Which is why IQ is often expressed as a range. If you score a 129, you'll be within the 125-135 group or something.

And it's just like with physical traits. Like right now, I will be much slower than normal because my quadriceps are fucking cooked. I overexerted myself and I'm sore as shit. Even though I am not normally fast, timing me now wouldn't give an accurate representation of my average speed.

And you can tell when someone has spent any time among the truly gifted. Because running into people like this is something else. They will come into a situation blind, get what you've already gotten, and then make a realization you'd never would have. They do process on a different level. And it's not about knowledge. It's not how much they know. It's about how fast they can acquire that knowledge, understand it, and then build upon it. It's how they can relate knowledge from different areas in new and novel ways. People who have spent time with the truly gifted know there's a such thing as intelligence. People who have only spent time with people who are average to above average have never actually seen intelligence at work.

> You accept that some individuals are off the scale of various physical abilities.

No I don't!

I'm getting on, and not very fit; I can run about 30 yards, slowly, before I run out of steam. But even Olympic runners can't run 100x faster than me. I know how fast Olympic runners can run.

> Why should we believe that everyone is roughly equal in intelligence?

Why, indeed. Why do you think that I believe that?

> Which is why IQ is often expressed as a range.

I have avoided mention of IQ; I don't think it's equivalent to intelligence.

> and then make a realization you'd never would have.

Yes, of course. I've met a lot of people that are quicker on the uptake than I am. But I'm sorry; "truly gifted" suggests there's some kind of magic going on. I don't believe in magic, and I don't believe in gifts from the gods. And I don't believe that there are people that are orders of magnitude smarter than the average; my observation was simply that average people like me aren't in a position to estimate the smarts of brighter people. So there is a tendency to assume there's magic occurring; that they aren't just brighter, but there's some other ingredient.

There's no magic ingredient; we're all made roughly the same way, from the same ingredients.

I think you believe that because you say basically that. Then you say that you don't believe there are people that are orders of magnitude smarter than the average. There are.

There's no magic. Just like the fact Usain Bolt running the 100m dash in 9.58 seconds isn't magic. No one else has been capable of reproducing that feat. Including people who train to do exactly that.

And I don't suggest there is magic. You're forcing an argument on me that I am not making.

IQ is as good a measure of intelligence as there is. IQ is not a measure of human worth. IQ is not a measure of success. IQ is not a measure of a lot of things. IQ is the rough measure of the raw processing power of a human brain.

I am simply pointing out that intelligence is no different than any other attribute one can possess. And some people possess it at the extremes.

> Usain Bolt running the 100m dash in 9.58 seconds isn't magic

Sure it is. But it's not orders of magnitude.

> IQ is the rough measure of the raw processing power of a human brain.

[citation needed] That claim is both specific and vague at the same time. I'll try to work out how you did that :-)

Depends on your deviation.

And how do you mean "specific and vague".

IQ isn't precise because like all of our physical attributes, there's wiggle room. Like how you may think you're a certain height, but you actually fluctuate throughout the day. Same with weight. Or strength. Or speed. IQ is a rough measure. It can really only tell us what you get when you take the test. But if you take it several times, it should fall within a range.

And a good IQ test shouldn't test knowledge. It should test reasoning, spatial ability, pattern matching, etc. You know, the factors that make up our raw processing power.

Are you saying it's "specific and vague" because it's an imprecise measure of a specific ability? That wouldn't be "specific and vague" in the way you mean. You could say it's a vague claim about a specific thing, but that's not mysterious.

I meant "specific and vague" like this:

> Then you say that you don't believe there are people that are orders of magnitude smarter than the average. There are.

So you're stating that some people are orders of magnitude smarter than the average; but you've adduced no evidence. I don't agree; you're making the claim, so the burden is on you.

> IQ is the rough measure of the raw processing power of a human brain.

That's a very specific claim. I know what "raw processing power" means in the context of a CPU, and I know of various ways of measuring it. But a brain is not a CPU. It's not a processor; arguably what it does is not similar to information processing.

> And you can tell when someone has spent any time among the truly gifted. Because running into people like this is something else. They will come into a situation blind, get what you've already gotten, and then make a realization you'd never would have. They do process on a different level. And it's not about knowledge. It's not how much they know. It's about how fast they can acquire that knowledge, understand it, and then build upon it. It's how they can relate knowledge from different areas in new and novel ways. People who have spent time with the truly gifted know there's a such thing as intelligence. People who have only spent time with people who are average to above average have never actually seen intelligence at work.

I used to think this as well, having gone to a famous university. Some people just seemed to get it much faster than others.

With some hindsight I no longer believe it. I think it was all hard work, and very much influenced by environment.

One guy I knew seemed really quite bright, always seemed to know everything. Ended up getting a PhD as well. But I realized that he basically just studied a lot. I wanted to think he was smarter than me, but I also knew inside I wasn't putting in the same amount of practice he was.

I brought the same fellow into a business to work with me, and he was ordinary. Couldn't have known that he was one of the top grads of two world famous universities, just one among a sea of reasonably competent quants. I'm fairly sure a lot of firms make this mistake as well, they think they need the people who did best at uni, but actually it has a lot to do with what you do with people, who you stick them with, what training they get, and so on.

The thing about highly productive creatives is they seem to have special motivation. They ome into some field, they see connections, they are more motivated to read more, talk to more people, study more. That seems to be the driver. I never saw anyone good that was just smart but not engaged as well.

Distributions of human skill that we can measure tend to be on long-tailed power scale distributions. Chess is one of the most obvious examples of this. Chess world champions tend to be just dramatically better than the 2nd best players in the world. One early 20th century world champion (Emanuel Lasker) retained (and actively defended) his title for 27 years! And he was also a research mathematician (who made some significant discoveries) and more 'on the side'.

And while that exact stretch remains unbroken, similar levels of dominance in the game are typical. Kasparov was world champion for 15 years. And in current times Magnus Carlsen has been world champion for about a decade, and #1 in the world (by a wide margin) for years beyond that.

On things we can't measure, one can observe facts like the open problem that Einstein would go on to solve (in his twenties, while working at a patent office) was being dug into by countless brilliant minds, to no avail, since Einstein was 8 years old! And his work after that point makes it clear it wasn't just a freak coincidence.

---

I definitely do think there are plenty who are "that much" smarter than average, but I don't think they need to be paid to succeed. Nor do I think one having an official affirmation that he is a "genius" is conducive to clear-headedness. It naturally drives a sense of entitlement, and can be detrimental to actually putting in the work necessary to make things happen. All the potential in the world is irrelevant without a corresponding work ethic.

Chess is an interesting example. I learned to play chess as a kid. I never got even slightly good; people that were dimmer than me could beat me easily.

But how many chess grandmasters are also fabulously wealthy CEOs? They tend to stay chess-players.

I'm curious about what skill in chess consists of. It seems to correspond to general cognitive capacity; but that's certainly not all of it.

People aspire to different things.

It's difficult to explain if you've never really gotten seriously into chess, but the game is, by far, one of the most satisfying and rewarding activities in existence. You can put infinite mental energy into the game and continue to get ever more out of it. It is extremely addictive, especially once one reaches a certain level in the game. If I had the choice of retiring to a life endless beaches and scantily clad ladies delivering me all the drinks and pleasures of life one could imagine, or retiring to a [safe] ghetto full of nothing but old highly skilled chess players - I would pick #2 in a heart beat.

There are plenty of chess player anecdotes along the lines of Gata Kamsky. He became frustrated with his progress in chess at some point (as one of the best in the world, but not the best in the world) and just casually decided to become a lawyer. So he did that and breezed through law school and set himself to a traditional path of money and even power. But, as is always the end of these stories, the game called him back. Today he streams himself playing chess for hours a day (on Twitch) to an audience of tens and is clearly more content and satisfied than ever.

> long-tailed power scale

Yeah, but a power scale implies that there are outliers at any distance from the median. That can't be true, so the model is defective.

Except it is true! That is precisely what I was getting at with chess. Right now the #1 player in the world is rated 48 points higher than the #2 player, and this difference is typical and often even more extreme. And the #2 is rated almost 20 points higher than the #3 player!

I could try to explain what those numbers mean mathematically, but I think there's an even easier way to get the gist of it. There are hundreds of of thousands of FIDE rated players. That #1 player's rating is 2859. For him to be 50 points higher than the #2 player is insane in terms of any notion of broad-level approximate human equality, but perfectly normal in the not-even-remotely-equal reality of what we see in chess.

And this is mirrored everywhere, even in sports where the physical limitations on the body are presumably more limiting than the cognitive ones. Running is nice because it's a field where we are going to miss close to 0 outliers, and you can fairly compare generations regardless of time.

Imagine we take Usain Bolt at his peak, and have him run a 200m race against the 200th fastest human to have ever lived at his peak. People might expect this to be a pretty close race. In reality? He'd finish about 9.8 meters, more than 32 feet ahead of the other guy. It would look like 'fast' guy vs 'meh, more or less average' guy. When in reality that 'more or less average' guy is one of the fastest humans to have ever lived!

A true power law would say that there's a diminishing number of chess-players the further out you go along the long tail. But in practice, there are zero chess-players better than the world champion. It may look like a power law within some window, but power laws don't have a cutoff point.

That is what I'm saying about smart people; I don't accept that there are people arbitrarily smarter than the average. My guess is that the cutoff is somewhere around double the average (modulo your method for measuring smartness).

You can see various visualizations of chess ratings here: https://nycdatascience.com/blog/student-works/visualizing-fi...

It is a textbook power law. There are only 2 humans better than 2800, and one is pushing near 2900. Then there are only 37 humans from 2700-2799, 168 from 2600-2699 and so on with an increasingly exponential gain in brackets near the median. And each bracket tends to be exceptionally dominant of brackets just one below, let alone more.

A finite sample means any distribution will have a min and a max. If the odds of a Magnus are 1 in 10 billion, then we can expect to see one person of a roughly proportional skill every generation or two. But if the odds of the next tier of exponentially dominating outlier out are 1 in 500 billion then any given generation of humanity would be unlikely to see one, at least not until we start dramatically increasing our population sizes.

But none of those examples gives a reason why it had to be that particular person. Someone is gonna be the first person to solve an unsolved problem, or the new best chess player. It might well be that longevity comes from reaching the top?

The point with chess is to illustrate that relative ability follows a power law. If humanity was relatively normally distributed then of course there'd still be a top, but people in this top would all be roughly equal. And so, like you're implicitly stating, any sort of achievement would largely come down to throwing darts and seeing who, largely by chance, ends up being the first to hit the bullseye.

But in reality the top is not approximately equal. In chess there are hundreds of thousands of rated players. The #1 player is rated 50 points higher than the #2 player. The #1 player is rated 2859. Think about the implications of those three facts! And this is likely precisely the case with Einstein, and also the reason for his success.

Back to current times/domain I'd also argue this is why having exponentially more scientists than ever before has not really led to what somebody who assumed a normal distribution would expect to see.

The examples illustrate that relative outcomes seem to follow that kind of law. How do we know that the outcomes are strongly tied to some sort of potential, and not just self-perpetuating?

You get picked for the team because you're a little bit better than number 2, you feel motivated, you get more training... down the road you are a lot better.

Sports especially is filled with these kinds of stories.

In general it's pretty difficult to prove a negative. How do you know you don't have a friendly little goblin named Rakinichu, made of dark matter, floating just above your shoulder? You can't really prove that you don't, so the onus is on me to prove that you do. That said, in this specific case, I think there's even reasonably strong evidence against your hypothesis. For instance Einstein's "miracle year" would come while he was working as a low level inspector at a patent office. And he was working there because he applied to numerous universities, but none were interested in taking him on board!

Even in the chess world, "prodigies" are a dime a dozen and Magnus was a relatively late bloomer among his peers. Magnus received his International Master title (the second highest title in chess, prior to Grandmaster) at a later age than Sergey Karjakin, a fellow prodigy, received his Grandmaster title. He also achieved victory coming from a country that had relatively little chess culture. Grandmasters coming from countries with greater chess culture would have had access to greater resources, minds, coaching, and opportunities.

> In general it's pretty difficult to prove a negative.

But this isn't a negative, you're proposing that there's some sort of strong connection between observable potentials and eventual outcomes, am I right? So now we need the evidence.

Sports is a great place to look btw. The history of recruiting busts in the sports world simply incredible. People they guessed wrong on: Tom Brady, Ravel Morrison, Johnny Manziel, Carlos Bacca, just to give you a false positive and a false negative in two different sports.

It just seems to me that is really isn't all that obvious who might be really good, and so we should spread our bets rather than concentrate on a few promising geniuses.

> For instance Einstein...

So how does this say anything about skill? Couldn't he just be a low-probability ticket that won? There were other people looking at similar ideas at the time.

> Even in the chess world...

Thing is chess is international now, Magnus could be Norwegian but very cheaply access chess culture from other countries. Maybe in an earlier era this argument made sense. But just like in football, it's international now and people aren't surprised by regional styles.

No, that is not what I am arguing. I am arguing that people's skill distributions follow a power law, with outliers at nearly any level. To quote the earliest post in this chain, that indeed some people are "off the charts." And this inherent difference in these individuals plays a large role in their ultimate outcomes. And no, I don't think we should "bet" on anybody, whether focused or scattered. The point of Magnus, Einstein, and many others is that the fact that they ultimately were these "off the charts" individuals was in no way whatsoever clear.

And in any case, for people that are "off the charts" - they aren't going to have difficulty excelling in a capitalist economic system, nearly regardless of their humbleness of their beginnings. I would even be inclined to argue that acknowledging somebody as "genius" and plying them with money is more likely to lead to distorted rather than enhanced development. Just because somebody may be several sigmas from the mean cognitively, doesn't mean that they're immune to the trappings any human may succumb to: ego, vice, hedonism, entitlement, laziness, and so on.

---

As for your response, I find it uncompelling. You're trying to argue that there's some childhood butterfly effect that drives outcomes, while hand-waving away major factors that would imperil development (from an environmental POV). Similarly, the 'dart-throw theory' doesn't hold as it wasn't only Einstein's initial ideas that were exceptional - but rather a persisted trend throughout his entire career.

He published more than 450 papers over his 50 year career, with countless revolutionary discoveries. For some context, that's an academic output rate of publishing 0.75 papers per month, for his entire life until his death bed. It's not like he made one great discovery and then coasted on his laurels while reminiscing about the past. His "miracle year", for instance, while working at a patent office, involved the publication of 4 distinct (and ground-breaking) papers! All without any real access to academic resources, or even peers to really bounce ideas off of.