Imagine it’s New Year’s Eve 1999, and you are considering an investment in Microsoft. You have a high level of conviction (say 90%) that Microsoft will benefit from the coming internet revolution. And it turns out you’re right about that: since 2000, Microsoft stock is up over 900%, making it one of the best performers of the past two decades. But do you have the nerve to hold on to your investment when it loses 65% during the first year as the dot.com boom turns to bust?
Have you been unlucky? Or unskilful? Probably both – as Duke notes, “just as we are almost never 100% wrong or right, outcomes are almost never 100% due to luck or skill.”
However, as in all scientific endeavours, the sample size matters. Results will eventually revert to the mean, so it makes sense to evaluate the process irrespective of the end outcome. The risk in relying on outcomes as a feedback mechanism is that random, improbable results could wrongly lead us to change our framework or beliefs. Decision quality should therefore be judged on repeatability. The objective of decision analysis is to delineate luck from skill; to focus on the signal, whilst ignoring the noise.
Duke argues the best way to do this is through probabilistic thinking. She suggests envisaging different possible outcomes as the branches of a tree, with the size of each branch determined by both its likeliness and the impact of the potential outcome.
What does this mean for an investor? Nipping back to 2000, I have 90% confidence that Microsoft will benefit from the internet revolution, but I deem that outcome already fully priced into the shares. That complicates the risk/return picture – is it worth investing? Probably not, for now.
Duke also points out that, by deliberately considering alternative scenarios, we reduce our susceptibility to hindsight bias(seeing outcomes as inevitable once they’ve happened). The journey to today was once uncertain. What we struggle to recall are the branches lopped off by the passage of time. The past seems deterministic, whilst the future is probabilistic.
Counterintuitively, striving for accuracy in decision-making contradicts how our brains have evolved. Our prehistoric tendencies prioritise efficiency over accuracy. If you encounter a sabre-toothed tiger, don’t think, run – and hope someone else is slower to react. But, when making decisions, accuracy is vital: pick the poisonous berry, and your decision-making ability may be permanently impaired. Thinking in Bets seeks to marry these two otherwise competing modes of thought.
To think in bets means entering a state of perpetual learning. Challenging our existing tenets by seeking objectivity through truthseeking and belief calibration. Changing our minds as the facts change. Only then are we more likely to improve our chances of favourable outcomes. And favourable outcomes can compound, allowing the skilled poker player – or investment manager – to prevail over time.
