[mjburgess]

Some weak evidence of why "exceptional abilities" is not a bad idea, even if gimmiky, is that performance "at the extreme" is highly correlated. So that the people who tend, eg., to be concert pianists, tend also to be very accomplished artists, and the like.

So if you're hitting (a verifiable) top 0-0.5% in some field, there's a reasonable bias towards assuming a high general competence.

I did once hit 0.5 percentile in a multinational PHP exam in my teenage years however I did have a second window open with an interpreter running for the most fringe questions. -- who knows what that means.

[glitchc]

Wait, you expect a concert pianist to be a good software developer... on average? Most of the time? Not at all?

[Spivak]

Yes! Well specifically someone who is a concert pianist and then who also set out to be a software developer. The ability to get very good at a difficult thing is highly correlated to being able to get very good at another really difficult thing.

Case in point I have a friend who is a top 32 magic player in NA. She recently, not even a year ago recently, made it her goal to become a chess grandmaster and she's already 2000 ELO. You could argue that maybe some skills transfer but it's pretty shaky reasoning.

[michaelbarton]

Exactly. Olympic athletes in one sport have a much higher chance of being an Olympian in another. It’s not that they inherently “sporty” that makes the difference but they’re willing to put in the 100s of hours of training and get up at 5am regularly. You could say it’s discipline that would make them a good SWE?

See: https://www.nine.com.au/sport/olympics/olympians-who-changed...

[jimbokun]

Because a lot of the skills and personality trait of becoming elite at something are transferable to becoming elite at other things.

[mjburgess]

I expect a concert pianist who's applying for a software developer position to be worthy of an initial consideration.

I know a software developer who could well be a concert pianist, for example. Ie., that pool of people who overlap, in that overlap, are probably extraordinarily talented.

[butlike]

Yeah, I'm kind of with you. Just because they have the capacity to excel doesn't mean they're going to for YOUR company.

Also it smells like a false metric. People who are in the 0.05% of excellence are probably still heavily invested in the thing they're excelling at.

[evantbyrne]

It only make sense if you assume that people can generally brute force their way into the top fraction of a percent. Not a view that I agree with.

[dismalaf]

No, you'd expect them to have a capacity to learn at a high level. Which is a good trait for those who are also developers.

[DrillShopper]

They can't possibly be worse than half of my coworkers

[its_down_again]

disclaimer: not data or advice.

We once interviewed a former national [racket sport] champion from [country] for a React/JS SWE role. Our in-house [racket sport] expert, who happens to be the best player among us, was sure they’d ace the coding exercise. The next day, when we asked for his verdict, he gave us a few words: “He sucks. He got absolutely nowhere.”

Lesson learned: extreme talent in one domain not always a predictor of aptitude in another.

[ikr678]

It's a good filter for drive/ambition, not immediate talent.

The best grad/intern I have supervised had the least domain knowledge starting out, but was an olympic finalist level swimmer. They were easily the most competent by the end of the graduate rotation.

Unfortunately, they were also annoyingly perky first thing in the morning as they had already been awake and training since 4am.