Alpha On Steroids, AKA ‘Microwave Alpha’
In the Paradox of Skill , author Brad Steiman accurately proclaims that ” confirming skill takes an investment lifetime, and you can never be fully confident that the alpha is not random. ” Alpha is the intercept in a regression of fund performance versus a benchmark. It measures success across time, and that is why it takes so long — you need a lot of observations (time periods) to gain significance. As shown in the following picture, it takes more than 140 years to identify the skill of a low-skill manager, and even an extraordinary manager will take 20 years to manifest statistical significance, and by that time the management might not be the same. Click to enlarge Nonetheless performance evaluators continue to use alpha as their skill barometer without ever questioning its meaningfulness. No one wants to wait decades, so we ignore the underlying theory. “Alpha” sounds like science, being a Greek letter and all, but there’s little science in its actual usage. But don’t despair. There is a new and better approach that can deliver statistical significance in a much shorter period of time. Call it alpha on steroids, or microwave alpha — shortening decades to years. The breakthrough determines statistically significant success in the cross-section rather than across time. I’ve written about this approach in Real Long-Only Due Diligence and Real Hedge Fund Due Diligence . A portfolio simulator creates all the portfolios the manager might have held, selecting stocks from a custom benchmark — thousands of portfolios. A ranking in the top 10% of this scientific peer group is significant at the 90% level, even if it’s for a short period of time, like a quarter. To state an extreme example, a return of, say, 1000% is significant, and you don’t have to wait 50 years to declare it significant. This process creates what I call “Success Scores. ” A statistician would call them “p values.” A ranking in this scientific peer group is the statistical significance of performance above the benchmark. Of course it’s still important to get the benchmark right, which means custom is highly advised. So you have a choice. You can continue to use alpha, but you really should wait the requisite time before you invest, or you can use Success Scores. An additional benefit of Success Scores is that they replace peer groups with their myriad biases, including “Loser Bias” caused by the fact that most members of peer groups underperform their benchmarks, creating a race against losers. Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.