Join our community of smart investors
OPINION

Rewarding luck

Rewarding luck
April 12, 2017
Rewarding luck

Romain Gauriot and Lionel Page, two Australian-based researchers, established this by looking at what happened in Europe's top five football leagues when shots hit the frame of the goal. They identified 13,066 such occasions over 10 years, in 2,387 of which the ball bounced into the net and in 10,679 it rebounded out. Which of these happened was, surely a matter of luck. It shouldn't, therefore, affect judgments of the players' ability.

But it did. Professors Gauriot and Page show that journalists gave significantly higher ratings to players who scored rather than hit the post. And even team managers, who should know better, judged scorers more favourably. They were more likely to pick them for the next game and gave them more time before substituting them.

All this is evidence of the outcome bias: we tend to judge performance by results even if they were due to luck rather than ability.

In fact, we have even more striking experimental evidence of this. Jordi Brandts and colleagues at the Autonomous University of Barcelona got a group of students to guess the outcome of five coin tosses. They then asked other students to bet on who of the first group could best predict future coin tosses. They found that 82 per cent of students were willing to pay to back the individual who had been most successful in predicting the first five tosses even though that success was obviously due to luck. This is not an isolated finding. Experiments by Warwick Business School's Nick Powdthavee have corroborated it.

Investors make a similar mistake. They tend to buy good performing funds even though past performance doesn't lead to future performance by any more than you'd expect from chance, and even though that good performance might well be due to dumb luck. Economists at the University of Mannheim conclude that investors "confuse risk taking with manager skill and are thus likely to over-allocate capital to lucky past winners".

The fact that football pundits and investors make the same mistake shouldn't surprise us. In both football and investing the ratio of noise to signal is high. Single matches can be decided by luck, and share price volatility means that even investors without skill can beat the market sometimes, perhaps even for an extended period.

What's true for fund managers might also be true of corporate managers. Alex Coad at Sussex University has shown that corporate growth is largely random. And Liverpool University's Charlie Cai has shown that investors tend to pay too much for stocks that have grown rapidly. Together, these findings suggest that stock-pickers back luck and not just skill: they pay for growth that might have been due to chance but under-pay for the ability to generate persistent profits. This is consistent with a point made by the Nobel laureate Daniel Kahneman: "Because luck plays a large role, the quality of leadership and management practices cannot be inferred reliably from observations of success."

I don't say this to complain about any injustice. Instead, the point is simply that in economic affairs as in sport the ratio of noise to signal is high. And even experts tend to underestimate this.