Many investors, investment professionals, and pundits make comments regarding the relationship between stock correlations and opportunities for active stock pickers.

For example, here is a recent example from the Financial Times:

Correlation crash clears the way for stockpickers.

The basic (albeit flawed) intuition behind the statement is that when correlations are low, the variation in returns is high, and therefore, stock pickers have an easier time differentiating themselves.

These arguments appear intuitive. So they must be true, right?

Not exactly.

I will show that stocks can be perfectly correlated and have massive return variability, similarly, one can show that stocks can have zero correlation and maintain limited return variability.

How is this possible?

In short, correlation measures examine mean-adjusted returns, not absolute returns.

Back to Basics: How Correlation is Calculated

I have had this discussion many times in person, but never put pen to paper (or fingers to keyboard). However, I recently came across a “correlation & stock-picker’s market” discussion from an investment professional who I deeply respect. Moreover, this investor is well-known for pursuing an evidence-based investment philosophy. So I figured it was time to write an article explaining why these types of claims are flawed.

Let’s make use of my Ph.D. in mathematical finance and delve into a formula. I realize formulas are not everyone’s cup of tea, so I try to translate my argument into plain English. I even built a spreadsheet you can download. The spreadsheet shows how the numbers trickle through and contradict the notion of any relationship between correlation levels and opportunities for active stock pickers. Just in case mathematical formulas are too confusing or not compelling enough, I also provide a separate and intuitive way to reason around these misguided assertions.

Figure 1: Correlation Formula