I think there's actually enough in the book to get a reasonable idea of what's going on, as well as an explanation for why you can't just do this yourself. 1)

I think there's actually enough in the book to get a reasonable idea of what's going on, as well as an explanation for why you can't just do this yourself. 1) Trading PnL is very steadily profitable for their main fund. It must be trading a lot and winning due to the law of large numbers. 2) Their futures fund has normal returns. It looks like a lot of other CTAs, which also are not trading on subsecond time scales. Chances are they have some mix of liquidity provision and arb strategies in Medallion, which normally is not a hedge fund, but they started in an era where that was what people did, so they ended up with a hedge fund and gaining reputation as such. This is also why we know anything at all about the returns. Nowadays you would have a prop trading firm and nobody would know, but people who know, they know the returns are quite steady. In fact IIRC the book mentions specific people they'd come across in the early days trading certain kinds of strats that are known to be profitable enough to beat your average hedge fund. My guess is they found a few of these and refined them. For the CTA strat, they run a hedge fund because that's the appropriate vehicle for a thing with a sharpe/information ratio around the 1-2 range. The thing that really makes it work for them is probably incumbency. They've been around forever, people in the business want do business with them, and they know where all the landmines are, both in regards to trading strategies and legal/business framework. Even if you knew the strategies you would struggle to set up all the infra, contracts, relationships, and so on.