Why you DON’T want strategies with high win percentages — probably the most counterintuitive thing you’ll ever hear

Corona Del Mar, CA

Howdy Friend,

I always found it strange that the folks working at Jim Simon's Ren Tech kept talk about a 51% win rate. While they're closed to new investors, it did seem odd that they kept going on and on about that number. After accidentally adding a strange rule to a Meta ML multi-strategy, it all makes sense.

So here's what happened...

I was having a problem where my Meta ML was hanging on to a few trading strategies for several months at a time. They were doing so well that they were taking up a larger percentage of the portfolio than makes me comfortable. So I added a rule to the Meta that says "if you're in a strategy for more than three months, get out." That seemed to smooth results, but there was a period in 2020 that suddenly had a massive 31% drawdown.

How does a Meta go from a 6% drawdown to 31% when the rule just moves the money into cash? It suddenly dawned on me that there was WAYYYYYY more hedging going on behind the scenes than I thought. And the key to unlock this mystery was sitting there in my head...it was the reason the fellas at Ren Tech were so proud of a 51% win rate.

Here's why that's such a great thing when most amateurs and pros alike would say that sucks. If 51% of your trades make money, that means 49% are positioned incorrectly. That also means that if some rare event occurs, a Black Swan, you're hedged automatically because so many of your trades are unprofitable. After building thousands of strategies with The Boss, I can tell you that a typical win rate that maximizes profitability is around 65%.

That means 35% of the time, the trades don't make money.

Yet it helps make the gains buttery smooth because it absorbs sudden shocks like the suspension in a dune buggy.