Think probabilistically

Think probabilistically

Imagine that I would have asked you in 2012: will the French president Nicolas Sarkozy win re-election in the upcoming vote?  At the time, that would not have been an easy question.  In fact, there were three variables that you could consider in predicting the outcome.  

The first is incumbency.  Data from previous French elections indicate that an incumbent such as president Sarkozy, on average, can expect to receive 67% of the vote.  So based on that, you might forecast that Sarkozy is 67% likely to remain in office. But there were also other variables to take into account.  For a number of reasons, Sarkozy had fallen into disfavour among French voters, and pollsters had estimated (based on low approval ratings) that Sarkozy’s re-election chances were actually 25%.  Under that logic, there was a 75% chance that he would be voted out.  However, it was also worth considering that the French economy was not doing so well and economists guessed that Sarkozy would only win 45% of the French votes.  

So there were three potential futures to consider: Sarkozy could earn 67%, 25% or 45% of the votes.  In one scenario, he would win easily.  In another one, he would lose by a wide margin and the third scenario was a relatively close call.  The question now is: how do you combine those contradictory outcomes into one prediction? Well, you simply average your estimate based on incumbency, approval ratings, and economic growth rates.  And if there is no specific reason for treating one variable as more important than another, then use equal weighting. This approach leads you to predict that Sarkozy has a 46% chance of re-election.  And indeed, in May 2012, he received 48,4% of the votes and he was replaced by François Hollande.  

This example is an illustration of the most basic kind of probabilistic thinking.  It’s a simple example that teaches an interesting underlying idea: contradictory futures can be combined into a single prediction. However, we are not really used to thinking about multiple futures.  Even though, the future is always a wide range of possibilities.  And these possibilities often contradict one another.  However, if you combine them, there is a higher chance that your prediction will be better – or sometimes even close to perfection.  That’s the value of probabilistic thinking.  It is the ability to hold multiple, conflicting outcomes in your mind and estimate their relative likelihoods.  Probabilistic thinking will help you to become better at predicting the future and at making better decisions. 

Latest articles by Filip