Seven deadly sins of investing: Confirmation bias & the death of the bear case

So far in Plato's 7 Deadly Sins series, we've written about overconfidence, the hot-hand fallacy, loss aversion, and the base-rate fallacy. 

While Part 4 was about the evidence we ignore at the start of a decision, the boring base rates that get drowned out by a vivid story, here in Part 5 we examine what can happen next. 

Once investors own a stock or have formed the thesis, another sin can take over - the tendency to stop seeing the evidence that would change your mind.

But before we get to any of that, watch the short video below. It is about ninety seconds long. Follow the instructions on screen carefully, then read on.

If you were counting passes for the first time, there is roughly a 50% chance you did not see the gorilla. A person in a full gorilla suit walks into the middle of the scene, stops, beats their chest, and walks off. 

About half of viewers, focused on the ball, miss it entirely. When told afterwards, many refuse to believe it until they watch the video a second time.

The experiment was run by psychologists Daniel Simons and Christopher Chabris at Harvard in 1999, and it has been replicated more times than almost any result in modern psychology. The lesson is uncomfortable. What we are looking for determines what we see. Everything else, no matter how obvious, becomes invisible.

This is the doorway into our fifth deadly sin.

Confirmation bias

Confirmation bias is the tendency to seek out evidence that supports what we already believe and to overlook evidence that contradicts it. It is the cognitive equivalent of counting basketball passes. 

Once the brain has a job to do, defending a thesis, justifying a position, protecting a decision, the gorillas walk past unseen. Investors are especially vulnerable, because the act of buying a stock is the act of forming a thesis.

Why this matters in markets

Inside an investment process, confirmation bias is rarely loud. It is quiet, gradual, and almost impossible to notice in real time.

A portfolio manager buys a stock. The thesis is sound. Over the following months, the broker notes that arrive on the desk get sorted, often unconsciously, into "interesting" and "noise."

The bullish ones get read carefully. The bearish ones get a glance and a dismissive "they don't get it." Management meetings reassure. The dissenting analyst on the team gets quietly moved off the name. None of this is dishonest. It is just the gorilla walking through the frame.

The result is that the bear case slowly dies inside the firm, even as the disconfirming evidence accumulates in plain sight. 

The ostrich effect

In a 2009 paper, behavioural economists Niklas Karlsson, George Loewenstein, and Duane Seppi studied how often retail investors logged into their brokerage accounts. 

The pattern was striking. When markets were rising, login frequency jumped. When markets were falling, it dropped sharply. They called it the ostrich effect.

It is confirmation bias stripped of its pretensions. The ostrich is not filtering the evidence, the ostrich is refusing to look at the evidence at all. And in a way, this is the more honest version of the sin, because at least the ostrich knows what they are doing.

The ostrich effect is also where confirmation bias compounds with loss aversion from Part 3 of this series. The investor sitting on a loss has two biases working in tandem. Loss aversion makes selling unbearable. Confirmation bias quietly screens out the evidence that would force the sale. Together, they are how a 10% loss becomes a 60% loss.

A familiar case study

A2 Milk (ASX: A2M) was once the great ANZ growth story. Premium infant formula into China, a booming daigou channel, and a middle class that seemed to grow forever. 

From 2015 to 2020, every new data point seemed to confirm the bull case.

 Export volumes were up. Brand surveys were glowing. Broker price targets kept marching higher. Then in 2020 and 2021, the daigou channel collapsed, inventory piled up in the trade, and the stock fell from above $20 to around $5. The disconfirming evidence had been visible for months in the ASX announcements. But most investors had their head in the sand.

Closing thoughts

It is worth pausing on how deep this bias runs. The scientific method, arguably the single greatest driver of human progress, is grounded in falsification. Karl Popper's insight was that a good hypothesis is not one you can prove, it is one you can disprove. Researchers are trained, against every natural instinct, to seek out the experiments that would break their theory rather than confirm it. We had to build an entire institutional framework to force ourselves to do the opposite of what our brains want to do. That is the scale of the problem.

The defence is not to try harder to be open-minded. Smart, well-trained, self-aware people exhibit the bias just as reliably as everyone else. Awareness helps, but it is not enough on its own.

What works is structure. Pre-mortems, where you imagine the position has failed and work backwards to the reasons why. Red-team reviews, where someone is formally tasked with arguing the bear case. And, ultimately, systematic processes that re-evaluate every position on the evidence each day, with no memory of what they concluded yesterday and no emotional attachment to being right.

The gorilla will always walk through the frame. The only question is whether your process is built to see it.

Learn about Plato Investment Management

This article is part of series of investor letters by the Plato Global Alpha Fund team. 

The Plato Global Alpha Fund has delivered +22.5% p.a. (after fees) since inception (1 September 2021). The Fund is available as an unlisted unit trust or via the Plato Global Alpha Complex ETF (ASX: PGA1). 

Click here to learn more about Plato Global Alpha. 

Managed Fund

Plato Global Alpha Fund

Global Shares

View