A Simple Strategy for Shaking Confirmation Bias

A Simple Strategy for Shaking Confirmation Bias
Brett N. Steenbarger, Ph.D.



One of the most insidious cognitive biases affecting investors and traders is confirmation bias.  Once we hold a particular view, we tend to prefer processing information that fits with that view.  What’s worse is that, because of our bias blind spots, we commonly recognize biased thinking in others, but not in ourselves.  We know from research in psychology that fresh inputs and mental flexibility are essential to creativity of thought.  When we become locked in confirmation bias, we cease to innovate.

So how is one to overcome not only confirmation bias, but our own bias blind spots?  One way is to develop the discipline of reading articles and posts that have been selected for their quality by experienced, independent agents.  This is why I make it a habit to read selections curated by Abnormal Returns and The Big Picture.  At various times, the editors of those blogs, Tadas Viskanta and Barry Ritholtz, have highlighted posts from my TraderFeed blog.  It’s been my observation that, with rare exception, they have selected posts that I myself found to be among my more informative offerings.  By delegating information curation to multiple independent parties with demonstrated domain knowledge and critical ability, I ensure that what I read is not a mere reflection of what I believe.  Long-time readers will recall the curated links of Charles Kirk, who now offers those to traders he mentors through his site.  This is a dynamic I have noticed in medical education:  part of teaching is directing trainees to quality information sources—and helping them distinguish high quality material from the wealth of verbiage out in the world.

A second information source I use to ensure a fresh flow of quality material is SSRN, the Social Science Research Network.  This is quite possibly the greatest treasure trove of free, high quality information that investors don’t know about.  Interested in cognitive science, economics, entrepreneurship, finance, innovation, or leadership?  SSRN is a repository of journal abstracts and scholarly papers from leading researchers around the world.  Here the peer review system—and the quality of the authors’ published works and academic affiliations—help to ensure high quality among the readings.  One SSRN feature I particularly like occurs when you download a paper:  links appear for related papers that have been downloaded by those accessing the original manuscript.  It’s a great way to quickly survey relevant, high quality research on a topic.

So, for example, I recently scoured SSRN and came across a provocative paper on Factor Investing by Andrew Ang.  Drawing upon his experience working with Norway’s sovereign wealth fund, Ang documents how the lion’s share of investor returns come, not from timing and active management, but from asset allocation.  How did I find Ang’s work?  It popped up as a link when I checked out a work by Marcos Lopez de Prado on what to look for in properly and improperly constructed backtests of ideas in finance.  And, yes, alas, it turns out that a system I had developed to trade the S&P 500 Index qualifies as overfit by the criteria of that article.  There aren’t many better challenges to confirmation bias than that!

We are properly discerning about the food we put into our mouths, but not so much with the information we choose to process.  Fresh inputs, chosen for their quality, provide a healthy intellectual diet and catalyze creative thinking.  If we’re not regularly surprised by what we read and discuss, the odds are good that our brain’s diet has turned stale.  And that’s a danger zone for confirmation bias.


Brett N. Steenbarger, Ph.D. is Clinical Associate Professor of Psychiatry and Behavioral Sciences at SUNY Upstate Medical University in Syracuse, NY and works as a performance coach for hedge funds, proprietary trading firms, and investment banks.  He is the author of the TraderFeed blog and several books on trading psychology.

Print Friendly, PDF & Email

What's been said:

Discussions found on the web:

Posted Under