11 Apr 2017

Not So Smart, Not So Dumb: Understanding the Difference Between Skill and Luck

Adam Geha

11 Apr 2017

positioner

I’m not sure if you too have noticed … but there are a few more geniuses walking in our midst lately.

From 2013 to 2016, a great deal of money was made in Australian commercial real estate, and we at EG enjoyed our fair share. Nothing wrong with that.

The problem arises when investors attribute market performance to their own genius when – truth be told – the market has made geniuses of us all.

It matters because over-confidence is so often followed by spectacular failure.

Why does this matter? It matters because over-confidence is so often followed by spectacular failure. Which is no small tragedy if you’re investing your own money – but a tragedy of Shakespearean proportions if you’re entrusted with the mighty responsibility of investing other people’s money.

Your Brain is the Problem

Owning and operating a brain is difficult. Owning and operating a brain, as an investor, is treacherously difficult.

We are all issued with a brain at birth. What we are not told is that the model arrives riddled with glitches and cognitive biases. It is also surprisingly prone to delusions. Oh, and one other thing, it also comes with a powerful over-confidence module that works around the clock to protect your self-image as an “unerring genius”.

You are, in effect, sleeping with the enemy (your brain).

So says David McRaney, author of the book You Are Not So Smart, which I highly recommend you read (even though, strangely, he doesn’t cover the bias I focus on in this article).

Self-Attribution Bias

One common glitch in our brains is something researchers have called “self-attribution bias”.

Self-attribution bias refers to the tendency in humans to attribute successes to innate factors such as talent or skill or foresight, while blaming losses on external factors such as the market or bad luck.

“Heads, I got it right. Tails, it’s bad luck.” This, in a nutshell, is self-attribution bias.

This type of flawed reasoning is far more common than you may suppose. It is well known, for instance, that sports fans will regularly attribute their team’s success to skill, but their opponent’s victory to luck. Harmless enough in this context, but positively toxic when applied to investment analysis.

Self-attribution bias means we tend to remember when we get it right, but readily forget when we get it wrong

Irrationally attributing investment gains and losses can impair investors in two ways. First, investors who attribute investment losses to “bad luck” are unable to learn from those mistakes. Second, investors who disproportionately credit themselves for a windfall gain can become over confident in their own market savviness – and as we all know, “pride precedeth the fall”.

What’s interesting is that recent studies show that:

  • Self-attribution bias increases with education. University postgraduates are more prone to this cognitive bias than university graduates, who in turn are more prone to the bias than high school graduates. In other words, far from helping, education inclines you to greater self-attribution bias. This is because the smarter you think you are, the harder your brain will fight to protect your self-image.
  • Self-attribution bias is highest among moderately experienced investors (those who have experienced 1-2 investment cycles only); second in novice investors (those who have not yet experienced a single investment cycle) and lowest in highly experienced investors (those who have experienced 3 or more investment cycles). In other words, the bias is most prominent in investors aged between say 27-40.

So beware of entrusting your hard earned money to a 35-year old fund manager with a PhD … unless of course the fund manager plays a lot of backgammon.

“Say what?” I hear you cry. Bear with me.

Backgammon as Business Metaphor

Dating back to circa 3000 BC, backgammon is perhaps the oldest board game still in existence today. Doubtless it owes its impressive longevity in no small measure to the delicate balance it strikes between skill and luck. Backgammon is a game that is roughly 50% skill and 50% luck.

As such, backgammon is a tantalising proxy for the business and investment environment, where the dice represent unpredictable market forces.

And so it’s the perfect game to play if you want to study and overcome your natural tendency to self-attribution bias.

First, you will notice that beginners will regularly attribute victories in backgammon to their skill and will blame “bad luck” when they lose. This observation alone will in time mollify the very same tendency in you.

But for the small minority of players who are truly intent on becoming champions, there is no substitute for objective post-match analysis. Today, various backgammon software packages allow you to review your games by providing a comprehensive analysis of all instances of mistakes and luck. In this way, aspiring players can systematically identify areas for improvement and (let’s face it) find out if the opponent really was as lucky as they thought they were.

And if all this effort is worth doing for the cause of improving your backgammon performance, how much more so for the cause of improving your investment performance.

Post-Investment Analysis

I’m staggered by how few investment managers study their track record to understand what part of their returns are attributable to skill, mistakes and luck.

Here are a few disciplines that I’ve found to be very useful:

  • At the point of acquisition, we deconstruct our base case financial model and separate the equity returns into four buckets:
    1. Yield (net income),
    2. Passive Capital Growth (market forces),
    3. Active Capital Growth (strategy and skill), and
    4. Debt Funding (extent to which leverage drives returns).
  • We then get the whole investment team to vote on every deal and record every person’s vote and ask each team member to provide their own forecast of what the equity returns are likely to be (these may differ substantially from the base case financial model because for example a particular team member may have a more sanguine outlook on the market). Then each member provides a comment to explain their forecast return estimate, and these are all stored in our PRISMS® software where they can be readily retrieved.
  • Finally, at the point of disposal, we again deconstruct returns into the same four buckets and compare with: (i) the return profile at acquisition; and (ii) individual team members’ return forecasts and comments.

This process brings clarity and accountability to our investment decisions and I highly recommend it. It allows you to work out the extent to which actual returns differed from your base case model due to skill, mistakes or market forces. By the way, in recent years the analysis has not flattered: fully 70-80% of EG’s recent performance has been due to market forces.

And for those bold enough to claim prescience when it comes to market forces, we have that (all too inconvenient) record of their return estimates at the time of acquisition and their comments as well. It turns out that mere mortals are not all that talented at predicting major market movements, though self-attribution bias means we tend to remember when we get it right, but readily forget when we get it wrong. Human, all too human.

So Tacitus and JFK got it right: victory has a thousand fathers, but defeat is an orphan.

Ask and You Shall Find

So the next time you catch up with your investment manager, make a point to compliment them on their recent out-performance. And then listen carefully to their response.
If they sheepishly claim all the credit, lean in and whisper (a dictum that my brother, Shane Geha, was fond of repeating to me ad nauseum): “You are never as smart as people think you are when you succeed – and never as dumb as people think you are when you fail.”

Then, if just for sport, step back and watch your investment manager wrestle with the shadowy demons of their self-attribution bias.

Read on LinkedIn

Read more of Adam’s Insights here