In Episode 93, we welcome entrepreneur, author, and quant investor, John Reese.
We start with John’s background. When John was a child, his father was a subscriber to Value Line, and John related to the charts and numbers. Later, this love of numbers took him to MIT, where he researched how to take the wisdom from books and turn it into computer programs. Years later, when he sold his company to GE Capital, John needed to learn how to invest the proceeds. Yet, he wasn’t sure which investment guru to follow in doing this. He decided to study a handful of gurus, and was disappointed to find that there was no repeatability and sustainability of outperformance over multiple time periods.
However, John then came across Peter Lynch’s One Up On Wall Street. In the book, Lynch had provided enough detail about his strategy that John was able to translate it into a computer program designed to pick the stocks that Lynch might have chosen. The results were solid. John then moved on to Ben Graham, eventually codifying 12 different guru strategies. He then put his research up on a website, which eventually morphed into Validea.
Meb asks about the challenges of this – namely, many managers have a qualitative component to their stock selection as well quantitative. How did John account for this?
John tells us this was very challenging. He had to re-read the various books multiple times, determining whether the printed word actually matched what the guru did in the market, versus his actions revealing more information or biases. Meb asks about filtering the incredibly long list of potential gurus to follow, and John tells us the list actually wasn’t too long. Most gurus didn’t have a sufficiently-long track record of performance, or they didn’t describe their strategies in sufficient details as to be able to be codified.
Meb then asks how John determines when a period of underperformance reveals a manager has lost his touch, versus the manager’s style is simply out of favor.
John tells us that he first looks at the length of time in which the strategy worked. If it was long enough, he tends to believe that, at some point, the strategy will come back into favor. He goes on to tell us that in all of his research, he found that there was not one strategy that outperformed the market every single year. They were these periods of going-out-of-favor that paved the way for the outperformance that occurred when the style came back into favor.
The guys then jump into an actual example of how John’s guru quant strategies work, using Buffett. Be sure to listen to this part for all the details.
Moving on from Buffett, Meb asks if there are any common attributes to the models that tend to do the best – any broad takeaways.
John tells us that, over time, the more successful strategies tend to have a value orientation, some kind of debt criteria, and they’re all profitable.
Meb asks – “Okay, gun to your head, which strategy has outperformed?” I’m going to make you listen to find out John’s answer, but odds are you’ll be surprised.
Next, the guys turn to factors, with Meb asking if there are any combination of factors that John tends to prefer. John says he likes momentum and mean reversion. This leads into a conversation on timing factors.
As usual, there’s far more in this episode: practical guidelines for listeners looking to follow along… portfolio construction in today’s challenging environment… what John would have done differently if he could start over again on Day 1… a roboadvisor for income investors… and of course, John’s most memorable trade.
This one happened the day after Black Monday. What are the details? Find out in Episode 93.