> The Medallion fund has been closed to external capital since 1993, and analysis of the flagship fund’s annual returns suggests that significant distributions

> The Medallion fund has been closed to external capital since 1993, and analysis of the flagship fund’s annual returns suggests that significant distributions are made each year to keep the fund about the same size. For example, despite the fact that Medallion reported annual net returns above 29 percent every year between 2010 and 2018, the fund’s assets under management stayed at about $10 billion throughout that period. This suggests to me that whatever opportunities they're exploiting, they would vanish quickly if they would increase the fund's size (e.g. by compounding returns). It's worth taking a look at the Numberphile interview with Simons (https://www.youtube.com/watch?v=QNznD9hMEh0 https://www.youtube.com/watch?v=QNznD9hMEh0). There he suggests that they're mostly using ML and other relatively simple techniques, with a bit of more advanced math when it comes to predicting how a trade will move the market. This chimes in well with the idea that they have to keep their trades relatively small in order to remain this profitable.