I saw this article, “Thousands of Amateur Gamblers Are Beating Wall Street Ph.D.s”
Economists have noticed that betting markets like Kalshi and Polymarket are pretty good at predicting not just political events but economic data, too.
As the author notes, this was an especially poor example, given that bettors and economists were wildly off the mark. The Bureau of Labor Statistics (BLS) report released on Wednesday, February 11, 2026, showed that the the U.S. economy actually added 130,000 jobs in January 2026, versus 50,000-60,000 estimated by either economists or bettors.
But the rise of ‘autodidactism expertise’ relates to my post “The ‘some guy’ phenomenon,” which is one of my favorite and most prescient posts of 2025. I wrote it on a whim in about half an hour, yet it so accurately captures where talent is often found–among unknown people. The best AI researchers are probably some guys you’ve never heard of uploading to github, not at OpenAI. The best trader and macro forecaster is your’s truly. The best writers are on substack, not a major publishing house. The best political commenters are guys on twitter or streamers, not pundits on Fox. We see this over and over.
Statistically speaking, this makes sense: with 8 billion people worldwide, it’s reasonable to expect that smartest and most talented are going to be obscure. The pool of talented human capital is too large to only be captured by large institutions. Thanks to prediction markets, there is a direct financial incentive for the top talent to rise to the top and profit from such expertise, on their own terms without gatekeepers.
In the case of myself, I invented and published without any Wall Street experience, the best hedging and investing strategy known. I was the only person who said to short Bitcoin to hedge tech stocks. No one else had said this. The assumption was this was too risky. There are so many ways this could have gone badly: I could have instead shorted gold, tech stocks, silver, or bought treasury bonds, or gone to cash. Instead, against all odds or expectations, shorting Bitcoin proved to be the optimal strategy.
Hedge funds hire teams of physics or math postdocs and pay them 6-7 figures to find hedging strategies, while running sophisticated computer models. Yet here was this random blogger who intuited the optimal hedging strategy, by having the correct model of the world. A superior model beats experience or manpower. Also, IQ vs ability does not scale linearly: a 160 IQ is not merely 23% better than a 130 IQ.
Intellectual endeavors are generally harder to quantify compared to physical feats. For ranking the best novelists, do you sort by sales or acclaim? Or societal relevance or context? A book can initially sell poorly and be poorly reviewed, but still be impactful later. Or who is the best coder, or even best dozen coders? Impossible to say. There are too many languages to choose from or other variables. I don’t think the ‘some guy’ framework applies to sports as well, e.g. soccer or golf, or chess (such as ELO scores, which are objective and rank-ordered), but it does apply to areas where talent is harder to quantify than points scored, rebounds, or how fast someone can run. Yet the talent is clearly there.
Rather, it’s more like there is a shortlist of writers or coders who can be considered ‘the best’, but further granularity is not possible. In trading and forecasting, skill is easier to quantify: how accurately does your model describe reality? And then,second, applying said framework or model to create an investment strategy. There is still some ambiguity, but track records matter. For example, those who predicted an economic crisis in 2025 due to Trump’s tariffs, or who argued that Bitcoin would serve as a reliable hedge against inflation or a strong investment at that time, were significantly wrong. So it is possible to assess skill to some extent, and then try to rank order this.