> If they're "outliers", their success can be explained by chance. With many people tossing coins, some will do very well by the laws of probability.No, they cannot be. I meant outlier in the sense of a probability distribution with e.g. standard deviations, not to imply randomness. I’ll start by quoting what I wrote the last time this topic came up on HN:
“Virtu only lost money trading one day out of 1278 trading days between 2009 and 2014. In the most uncharitable analysis (1278/2; or the lost day happened in the middle), they had a 0.5^639 chance of doing that.
Maybe you disagree with 0.5 per day. Let's make it 0.9!
...But that's still 5.7 x 10^-30. How many firms do we need to exist for this to emerge by chance?”
Draw up a probability space for me and actually quantify what you’re asserting. What is an event? Is it a single trade? A trading day? A trading year? How many of these events are there, total, in that timespan? Of those, how many win? Can you map each of these events to a boolean function, such as a coin toss, or are the chances more nuanced? This is before we even get to the issue of quantifying risk as a metric.
How about we make the event a trading year; how many funds would need to exist to explain hedge funds like RenTech or Baupost? Moreover, can you explain three decades of extraordinary returns in a single fund as an outcome consistent with a uniform distribution across all eligible hedge funds?
This armchair probability analysis turns up every so often on HN, but to be blunt, the onus is on the analysis to actually make it empirical. You can’t just say “well, they’re doing this by chance, we can expect some of these to eventually emerge because statistics and coin tosses” without formally proving that there are actually sufficiently many hedge funds and criteria to satisfy that claim.
> Buffett beats the market - but he does it by understanding the businesses behind their stocks, their value and prospects, compared with their present price. He buys value for money, and generally holds it for the long term. He isn't an active trader, which to him is speculation.
This is false. Buffett is, by his own admission, an active investor. He picks stocks - what is your definition of an “active investor” if not someone who makes active decisions about how to invest? His letter, “The Superinvestors of Graham and Doddsville”, specifically outlines his investing and describes it as an active strategy.[1] In fact, the way you’ve described his work is an investment thesis consistent with modern trading strategies. At their core, statistical arbitrage and high frequency methods seek to identify price inefficiencies just as Buffett does, but on much smaller timescales. It’s a difference of degree, not category.
______________________
1. https://www8.gsb.columbia.edu/rtfiles/cbs/hermes/Buffett1984...