I think it was probably very good for me, in my childhood-to-early-teenagerhood, that I was able to be one of the most skilled / competent people within my social sphere, for one of my social spheres.

Specifically: there was this browser game site, Funorb, that used to exist. And I played games on it a lot, and was pretty competitive about it. There weren't official all-time highscores, but there were official records of who'd done the best at achievement hunting across the site's games, and there were unofficial all-time highscores, and I did my best to climb towards the top of several of these. I never got #1 anywhere, but at my peak I was somewhere around #10 on the achievement-hunting scoreboard, and there were several games in which I was able to get to the #2 or #3 slot on the all-time highscores.

And this helped me a lot, psychologically, on multiple axes. It let me develop a sense of what it's like from the inside to be an expert in a field, and to chat with and learn from the other experts. It let me build an intuition that getting really good at a thing is possible, that it's an option available to me, even if it might be hard and take practice. It let me learn how to become good at things, on the practice-techniques level. In general, it... trained me in what it's like to be more successful at something than most people are? Which is a useful thing to be trained in, for almost any goals a person might want to pursue!

(Not the only useful thing, by any means. Most of my particularly-ambitious projects post-Funorb, thus far, have mostly fallen down for reasons such as lack of project-management skill, executive dysfunction, the projects taking long enough for me to lose interest and go off to do other things instead, and so forth. But a useful thing. One important one among many, good for building momentum and un-greying options.)

The question, then, is: how can we scale this? Funorb had ~300,000 accounts with at least one achievement unlocked in at least one game, of which at least a few thousand were reasonably active for a reasonably long time, and the impact of being among the top players would have been much less if the pool had been smaller. But then what about everyone who wasn't one of the top players, who just played casually at the mid-tier of a handful of games? How can we arrange things so they, too, get the chance to have that experience of Unusual Competence?

The answer, I think, is siloing. Don't just have a single metric of competence; have lots of them, such that a person can become an outlierly expert in a small handful while others do the same elsewhere, and thus everyone can get the experience of being an outlierly expert at something. Which Funorb, in fact, did! It had a few dozen games, many of which had multiple modes. I was one of the top players overall, sure, but the place where I had the most visceral sense of expertise wasn't for my overall rank but rather in the specific games and modes where I was one of the all-time top players. And there were relatively few such games! Only four coming to mind off the top of my head, and perhaps another few I'm forgetting right now, out of the ~40 on the site.

Actually implementing this at maximum society-wide scale—engineering the world to allow any kid who wants to to get to experience being a world-class expert at something, and have it be real and not a fake participation prize, because lots of other people genuinely have done that same thing in a non-expert capacity—seems likely to be hard. And it's not the sort of ambitious society-engineering project I myself expect to find my comparative advantage in, relative to various others that I've been scheming about over the past year or two. But it is, I think, a worthy project in the abstract, one that I hope someone figures out some way to implement.