Is Stock Picking A Game Of Skill Or Luck?
Photo by Amanda Jones on Unsplash
“A major industry appears to be built largely on an illusion of skill [...] The evidence from more than fifty years of research is conclusive: for a large majority of fund managers, the selection of stocks is more like rolling dice than like playing poker.”
Daniel Kahneman, Thinking, Fast and Slow
Is active investing a game of luck, or skill, or a combination of the two? This question leads to all sorts of finger pointing and soul searching amongst managers. Indeed, since the dawn of investing, narratives have been constructed to justify the role of the stock picker.
Most people buy into the view that luck plays an important role in generating alpha, but the fact remains that humans are generally poor at understanding the dynamic interplay between luck and skill. Because of this, the relative contribution of these interrelated forces remains shrouded in doubt and a healthy dollop of controversy.
The beautiful game.
We actually embarked upon this article with the intention of exploring the crossover between asset managers and football managers. Indeed, both games are beautiful. They involve huge amounts of pressure, lucrative financial rewards (often with a lack of true alignment), an emphasis on style, ferocious competition and of course, heated debate as to the ability of one person to “win” over sustained periods.
Luck plays a major role in winning football matches, just as it impacts investment returns. Skill matters, too. And both domains exhibit an uncanny propensity for that great leveller, mean reversion. Just look at what’s happened to Man United in recent years, or the once fêted Mauricio Pochettino at Spurs. Analogies in the investment landscape are legion and too numerous to mention.
Instead of focusing on José Mourinho’s staggering £15m annual salary at Spurs (which follows his inglorious departure from perhaps the world’s most famous club with a lacklustre 42% win ratio in all competitions), we want to explore the nature of luck so we can better understand how it functions in real-world scenarios. We’re learning as we go, so please feel free to contribute your own thoughts by getting in touch via email or Twitter.
A side order of theory.
A couple of important definitions, so we know exactly what we’re dealing with:
Luck is the force that causes things, especially good things, to happen to you by chance and not as a result of your own efforts or abilities.
Skill is the ability to do an activity or job well, especially because you have practised it.
Twitter co-founder and fasting advocate Jack Dorsey has opined that “success is never accidental”. But it’s self-evident that some activities involve more luck than others. Roulette (certainly not the russian variety) requires more luck than ultra-marathon running; juggling live chainsaws involves more skill than winning £105 million on the EuroMillions. Still, arguments rage regarding the relative contribution of luck in predicting future outcomes.
There is, in fact, a simple way of establishing whether an activity involves skill, one proposed by research analyst Michael Mauboussin – just ask if the player can lose on purpose. If they can, it’s a game of skill (to a degree). If they can’t, luck persists (again, to a degree).
But we can’t talk about the role of luck and skill in investing without name-checking that eternal adversary of fund managers, mean reversion.
Past performance vs future outcomes.
In his influential book, The Little Book of Common-Sense Investing, Vanguard founder Jack Bogle monitors the performance of the top 20 funds between 1982-1992 and 1995-2005. In both periods, the top 20 funds were in the top quartiles based on their performance, but in the subsequent years after their outperformance, these funds dropped to the bottom quartile, becoming the worst performers. Thus the template disclaimer thrust upon managers by regulators rings true.
If we assume that investing is mean reverting, it follows that it involves luck. In his fascinating (and incredibly long) memo “Untangling Luck And Skill” (which later become a book with a similar name) Mauboussin describes skill as “a drag on the reversion process”. The investment management industry has normalised a narrative whereby mean reversion is a drag on the skill of managers, but what if it’s the other way round? What if skill is a drag on the omnipresent force of mean reversion?
People are bad at understanding the interplay between luck and skill because they are bad at understanding mean reversion and statistical uncertainty. Human beings have evolved by extrapolating future outcomes from past events. But the things that make us well-adapted biological organisms don’t necessarily make us good investors. “I burned my hand on the hob, so I will be careful to avoid touching the hob after cooking in future” makes sense from the standpoint of self-preservation. But it doesn’t follow that “Today I bought a stock and its price increased by 10%, so I will buy the stock again tomorrow.”
The inconvenient truth for investors is that extrapolation is a fallacy. Four decades ago, Amos Tversky and Daniel Kahneman identified a decision-making bias that they called “belief in the law of small numbers”. This is the incorrect (but nonetheless widely held) belief that small samples resemble the population from which they are drawn. Mauboussin has observed that the magnitude of this fallacy grows larger as the luck-to-skill ratio rises. And it’s absolutely rife in the world of investing, with PMs seemingly obsessed with press releases and earnings calls that reveal little about where stock prices and markets are going.
The focus on extrapolation amongst institutional asset allocators is another case in point – a 3-year track record for emerging managers is seemingly sacrosanct. Capital flows freely to those who have done well in the past, with little or no consideration for the role of luck in short-term performance (or at least, a lack of rigour in attributing performance to skill rather than luck).
The paradox of skill.
It’s worth considering the role of competition in markets and how it modulates skill.
The theory goes that as competition increases, investment managers become more skillful and levels of expertise converge, luck plays a bigger role in determining outcomes. This implies that the role of luck in active investing should increase going forward due to better education, access to uniform information, commoditised operational infrastructure and other competitive forces that are levelling the playing field for managers.
Marc Rubenstein highlighted “the paradox of skill” when he tweeted that:
“To many players of fantasy football #FPL it feels like the game is getting harder. They’re right. It’s due to the *paradox of skill* as explored by @mjmauboussin.”
Over the past 10 years the number of people playing Fantasy Football has increased from 1.95 million to 5.91 million. Rubenstein went on to explain that:
“With that demand comes an increased supply of information via Twitter, websites and podcasts. Via those resources skill has been democratised. Given that performance is a function of skill and luck, if the variance of skill goes down, then more of the result will stem from luck [... ] The pattern is the same in other competitive pursuits like baseball and the stock market.”
It turns out that skill doesn’t scale very well. Researchers at Stanford Business School have found that a typical mutual fund manager is persistently skilled and that top performers are especially skillful, but that the market is so saturated that skill is competed away as money flows into emerging funds. We also need to factor in the unintended consequences of past success and how this might impact subsequent performance. As AUM grows, the investable universe of stocks that a manager can trade shrinks due to a range of factors, making it more difficult to generate returns.
Parsing luck from skill.
How do we disentangle luck from skill?
Some sports are easier to analyse than others. Major League Baseball, for example has a long season consisting of 162 games. NBA basketball teams play half that number of fixtures. That’s why quantitative analysis of baseball is more effective as a management technique in baseball (as evidenced by Michael Lewis’ masterpiece Moneyball) than it is in basketball and other lower frequency sports. And it’s why we often feel a little short changed at the end of the FIFA World Cup, when one football nation is crowned the world’s greatest after only playing 7 games, having avoided some of the strongest teams in the tournament.
The same is true of investing, but with a twist. Yes, some funds have limited track records, but not all fund managers approach markets in the same way. Perhaps it is easier to discern the respective contributions of luck and skill for high-frequency managers who trade every day, than it is for those who buy and hold for long periods in concentrated portfolios with low turnover. As markets become more efficient, higher frequency and tech-enabled, the old school approach adopted by allocators since way-back-when will need to adapt. When crypto markets are open 24/7 and trades can be executed at breakneck latency, an equivalent 3-year track record can be compressed into a much shorter time frame.
Parsing luck from skill is possible, but it takes huge amounts of data. Newfound Research has calculated that it takes 27 years of data in order to establish whether a portfolio manager has what it takes to beat a 2% hurdle rate with the same 95% degree of confidence. How many fund managers do you know that have remained faithful to the same shop (or even the same industry) for the best part of three decades? Here’s a clue – Citywire reckons that only 1% of UK-based based fund managers have more than 20 years’ experience.
Inputs over outputs.
Andrew Mauboussin and Sam Arbesman have analysed mutual fund streaks over the past 40 years. Their research shows that only a small subset of the investing population is skillful, and that the percentage of funds that are skillful is declining, consistent with the market becoming more efficient and competitive. Luck is playing a bigger role in determining success, whether we like it or not.
But that doesn't mean that luck is everything. Consider another game that we like to play – poker. To wheel out an old cliché, you play the hand you’re dealt, within the context of every other player at the table. Your skill in playing that hand, relative to other players, can improve your odds of winning the pot.
In sports, skill is not a static thing. It rises as athletes become stronger, more coordinated and experienced, but it then falls as they age and their physical conditioning deteriorates. But in analytical and creative industries like investing (and yes, we do believe that investing is a creative industry), skill does not decompose – it compounds. Whilst football players tend to hit their peak in their late 20s, writers, artists and yes, investors tend to realise their potential later in life.
John Milton observed that “Luck is the residue of design.” Rather than extrapolating future returns based on past data that may or may not be repeatable, it makes sense for allocators to adopt an approach that focuses on inputs rather than outputs. The same goes for us as managers – it’s why we assess a stock’s prospects based on a range of factors, not simply how it has traded in the past. Corporate performance also exhibits strong tendencies for mean reversion. Could it be that the philosophy and process that has been refined by a management team says a lot more about a company’s chances of success than its previous 4 quarters of earnings? Could it be that a company’s people – their expertise, talents, expertise and relationships – are a better predictor of success than its historical performance? And could the same be true of fund managers?
A game of inches.
Aside from skill and luck, here’s another quality to look for in professional investors – humility. In his book Good To Great (which itself is pretty great), Jim Collins notes that the most successful leaders often attribute their success to luck. The same goes for fund managers. The best managers have great skill, but they also admit that luck has played a role in their success.
Stephen Jay Gould famously described baseball legend Joe DiMaggio’s 56-game hitting streak in 1941 as “a matter of extraordinary luck imposed on great skill.” That’s how we like to think of investing, too. By focusing on minimising downside risk we seek to create upside optionality (otherwise known as luck) via our ideas. We’ve been lucky, over the years, and we’ve been unlucky, too. Mean reversion might be the natural law of financial markets in aggregate, but that doesn’t make it inevitable for individual funds. As active managers we must we fight every day to resist the tyranny of statistical inevitability, inch by inch, to the end.