Rational Expectations

The investment business seems to be a place where expectations are often rather disconnected from reality. Of course we all want huge returns and no drawdowns, but that mostly isn’t reality. I say mostly because of Renaissance Technologies (RenTec) and their Medallion fund which produces a compound annual growth rate of around 66% with a maximum drawdown of around 6% per Gregory Zuckerman’s book.

So, what is reasonable? Let’s start with the outer bounds omitting RenTec. One of the best track records belongs to George Soros. According to Sebastian Mallaby in his segment of the book George Soros: A Life In Full:

…[Soros] averaged annual returns of 36 percent between 1969 and 1989, the period of Soros’s direct control…

Per various books on Soros, drawdowns during the period could exceed 30%.

Also according to Mallaby, Soros’s fund produced an average 33% when combining the above returns with those of Stan Druckenmiller who ran the fund from 1989 to 2000. Per other public information, Druckenmiller also hit a drawdown of around 30% before leaving Soros in ~2000.

So, these guys were producing compound annual returns (CAGR) about equal to their maximum drawdowns (DD), give or take. The ratio of CAGR to max DD goes by “MAR”. These guys had a MAR ratio of about 1 or slightly over (per the data I have available).

Next up is Warren Buffett. Also according to Mallaby, WB’s CAGR from 1969 to 2000 was 32%. Using data from Yahoo Finance for Berkshire Hathaway’s stock (BRK-A) I show the max drawdown to be ~50% in early 2000 (data only goes back to 1980). Here is a chart of the drawdown:

So, 32% CAGR/50% DD = 0.64 MAR for Buffett through ~2000. His returns of the last 20 years aren’t as good; the BRK-A CAGR from 12/31/99 until today is around 9.5% with a max DD of ~50% for a MAR of 0.19.

With that in mind, Druckenmiller has been rather forthright in various interviews over the past several years that his returns since 2010 are nowhere near 33%. He also readily points out that there were a few dozen hedge funds when he started vs thousands now. In short, competition has increased a lot. As this has happened, the returns and MAR ratios of many legends have fallen.

Omitting funds that are closed, if you screen through databases of hedge funds and look for funds that have over $50mm or $100mm in AUM who have been around at least 10 years, and have max drawdowns around 30% with double digit CAGR (aka MAR of 0.3) you don’t find much. Most of what you do find belongs to CTA trend followers which tend not to get a lot of AUM since their narrative doesn’t make the masses feel warm and fuzzy like a prediction does. Another note on these CTAs, their MAR ratios were ~4 back in the 1970s.

What’s the point? There is very little evidence in the real world that MAR ratios above 0.3 can be sustained indefinitely on scalable strategies. Indeed there are windows of terrific MAR outcomes, but these tend to result from golden eras for strategies and/or minimal competition. The glaring exception in RenTec.

If you’d like to know more about what I think is happening behind the RenTec curtain, read this article: https://twoquants.com/decoding-rentec/ The TL;DR version is I believe it would be about impossible to replicate RenTec at this point and that something like their Medallion fund could only materialize via gradual evolution.

Bottom line, it is always good to dream, but it is probably best to set your investment expectations according to reality. When setting expectations, keep in mind that almost all real world strategies are subject to the 0.3 MAR ratio “gravity” over an extended period of time.