Notes on The Success Equation

…great success combines skill with a lot of luck. You can’t get there by relying on either skill or luck alone. You need both.

– Michael Mauboussin, The Success Equation

Since my notes on The Outsiders: Eight Unconventional CEOs and Their Radically Rational Blueprint for Success generated a whole lot of reader interest and were featured on Market Folly, I thought you might be interested in more condensed book wisdom. This time, The Success Equation: Untangling Skill and Luck in Business.

With the exception of victims of severe self-serving bias, everyone is aware that success/failure is a combination of skill and luck. The hard part is distinguishing the two in different activities. The book goes a long way in helping us do precisely this.

The Success Equation

  1. Definitions
    • Luck out of your control and unpredictable, “luck is a residual: it’s what is left over after you’ve subtracted skill from the outcome.” This follows from the definition of skill.
    • Skill – luck is a small part of the outcome, when the process of making decisions is highly correlated with the outcome.
  2. Start by placing the activity on the luck-skill continuum. Chess (close to pure skill) vs roulette (close to pure luck). Can you lose on purpose? Being here, you are most likely interested in investing. So, it is good to keep in mind that the author’s research puts investing closer to the luck side than to the skill side of the spectrum.
    • Take sample size into accountthe variation of the mean is inversely proportional to sample size. For skill-dominated activities, you don’t need as large a sample to draw reasonable conclusions as you do for luck-dominated activities. “A small number of results tell you very little about what’s going on when luck dominates, because the bell curve will look fatter for the small sample than it will for the overall population.”
    • Skill = permanent; luck = mean-reverting
  3. Five things to know about statistical prediction:
    • Base rate – prior information about such cases
    • Specific evidence – info about the case at hand
    • Expected accuracy of the prediction – how precise you expect to be given the info that you have. If expected accuracy is low (luck is high) – weigh the base rate more. If expected accuracy is high (skill is high) – weigh the individual case more.
    • Correlation ≠ Causation – A preceding B doesn’t mean A causing B. Stories do exactly this. They tell us that A causes B and we feel smart and satisfied. Beware of stories in investing! “…when your undertaking involves a dose of luck, the link between cause and effect is broken.” This is what spurious correlations look like.
    • Still, correlation is important – “Perhaps the most important idea is that the rate of reversion to the mean relates to the coefficient of correlation. If the correlation between two variables is 1.0, there is no reversion to the mean. If the correlation is 0, the best guess about what the next outcome will be is simply the average.” However, “the most obvious hazard is that correlations are not stable in many fields.”
    • Statistical significance – Another issue, very familiar to data scientists, is called data mining. As long as you examine enough relationships, you are bound to find some that are statistically significant, but significant by sheer chance.
    • 4th quadrant (black swan domain) – Statistical methods don’t work for events with extreme outcomes and complex payoffs.
  4. Find the useful statistics by examining the coefficient of correlation.
    • Persistent (reliability in statistic speak) – past highly correlated with present
    • Predictive (validity in statistic speak) – outcome highly correlated with goal
    • For example, in investing earnings growth is a predictive statistic. You can earn high returns if you can predict earnings. But, earnings growth is not persistent, making it useless. Sales growth, on the other hand, is much more persistent, but it is less correlated with returns.
  5. Build skill and acknowledge luck
    • Deliberate practice – best way to succeed in activities that are dominated by skill.
    • Focus on process – best way to succeed in activities that are dominated by luck.
    • Acknowledge luck – it plays a part in every endeavor and there is nothing you can do about it. Acknowledge this and move on – either with deliberate practice or polishing your process. Don’t let it fool you into thinking you are so smart/skilled. This one is very pertinent in the ongoing bull market.
  6. The Paradox of Skill – with more skill luck becomes more important. When participants are matched on skill, it would be luck that determines the outcome. This is particularly relevant for stock market participants. Exactly because there are plenty of skilled participants, the market is mostly efficient most of the time and investing lies more towards the luck side of the continuum.
  7. Simple models work for independent events (like tossing a coin). “You don’t know what the outcome will be, but you do know all of the possible outcomes.”
  8. Dependent events (probably all events with a social element) require what Howard Marks calls second-level thinking, because participants influence one another’s choices. “The process of social influence and cumulative advantage frequently generates a distribution that is best described by a power law. … One of the key features of distributions that follow a power law is that there are very few large values and lots of small values. As a result, the idea of an “average” has no meaning.”
  9. The Equation
    • Estimated true skill = Grand average + shrinkage factor (observed average grand average)
    • Just plug the correlation as a shrinkage factor, the grand average for the activity and the observed average in the case you are considering. As you can see from the formula, with a factor of 1 the equation simplifies to skill = observed average, i.e. skill dominates and your best guess for the next outcome is the previous outcome – no reversion to the mean at all. When the factor is 0, the equation simplifies to skill = grand average, i.e. full reversion to the mean and your best guess for the next outcome is the average outcome.

Here are some more notable quotes from the book.

“If you become skilled in a physical or cognitive task, your body knows what to do better than your mind, and thinking too much about what you’re doing can actually lead to degradation in performance. In these activities, intuition is powerful and valuable.”

– Michael Mauboussin, The Success Equation

Those of you, who have spent some time perfecting a sport, know this feeling of your body being on autopilot – it does all the right moves without a second of thought. The closest to this we have in investing are Buffett and Munger’s 5-minute decisions on whether to invest or not.

The book also deals with an issue I have considered writing about but haven’t yet gotten around to doing it. I am talking about another investor favorite – Good to Great by Jim Collins. Other book authors and bloggers have pointed to the poor performance of the companies featured in the book in the years after its publication. It is a classic example of survivorship bias. The companies that used the same strategies and failed were left out of the sample.

“No one questions that Collins has good intentions. He really is trying to figure out how to help executives. And if causality were clear, this approach would work. The trouble is that the performance of a company always depends on both skill and luck, which means that a given strategy will succeed only part of the time. So attributing success to any strategy may be wrong simply because you’re sampling only the winners. The more important question is: How many of the companies that tried that strategy actually succeeded?”

– Michael Mauboussin, The Success Equation

For an example of a great company with a brilliant strategy that failed nonetheless, read more on the story of Sony’s MiniDisc.

The Success Equation is a must read for those striving to better understand the game (any game) and gain a good advantage over the less well-informed players.

“The argument here is not that you can precisely measure the contributions of skill and luck to any success or failure. But if you take concrete steps toward attempting to measure those relative contributions, you will make better decisions than people who think improperly about those issues or who don’t think about them at all. That will give you an enormous advantage over them.”

– Michael Mauboussin, The Success Equation