By now, even the most casual sports fan may have an opinion on England’s chances of beating South Africa this Saturday and thus winning the 2019 Rugby World Cup.
Should you not happen to have any thoughts on the matter, however, and decide to canvass the views of a couple of rugby-loving friends, your ultimate conclusion may well depend on whether they use descriptive terms or answer with a numerical probability.
So, at least, runs the argument of US academics Robert Mislavsky and Celia Gaertig in their newly published paper Combining Probability Forecasts.
Its secondary title, ‘60% and 60% is 60%, but Likely and Likely is Very Likely’, offers an indication of its authors’ thinking but – and not for the first time this World Cup – let’s return to a rugby analogy to illustrate the point further.
Numeric probablity may sway opinion
Say you asked your friends what they thought the odds were of England captain Owen Farrell lifting the trophy this weekend and one replied 60% and the other 80% then, according to Mislavsky and Gaertig – and assuming you had no preconceived ideas of your own on the matter – you would typically process those answers as the probability being 70%.
You would, in other words, instinctively arrive at an average.
If, however, both your friends thought it ‘likely’ England would win the match – skipping over the fact that word can mean different things to different people (as we discussed in Communicating investment risk) – then here you would typically take both replies as a double endorsement.
In effect you would add the two positive signals together to conclude England’s chances were not ‘likely’ but ‘very likely’.
Not everyone likes sport
And just in case our sporting analogies leave you cold, we will also throw in this travel-oriented example, which Mislavsky and Gaertig themselves use in their paper.
“Imagine that you are purchasing a plane ticket for your next vacation and you check two websites, Kayak and Hopper, to see if they predict any future price changes,” they write.
“If both websites say that there is a 60% chance that prices will increase, you would typically average the two and also believe there is a 60% chance. However, if both sites say that it is ‘likely’ that prices will increase, you would act as if you are ‘counting’ each prediction as a positive signal, becoming more confident in your prediction and believing that a price increase is ‘very likely’.”
This has serious implications for investors
Now, if this paper’s conclusions are correct, it would appear to have all manner of implications for investors – not least with regard to how they go about picking stocks for a portfolio.
If you are forming your own views by carrying out your own research – as we do here on The Value Perspective – then you are giving yourself a good chance of avoiding falling into the psychological trap of the ‘double endorsement’.
If your investment process is largely based on the ‘buy’, ‘hold’ and ‘sell’ views of sell-side research analysts, however, then that is another matter.
In that case, if a lot of analysts view a particular stock as a ‘buy’, say, Mislavsky and Gaertig’s research suggests you could essentially process that information as cumulative and, with every positive recommendation, grow increasingly confident in that company’s prospects.
Indeed, processing information this way can feed into a number of the behavioural biases that are hardwired into the human brain and which investors must do their best to guard against.
Examples would include ‘base rate neglect’, which features in our checklist of the leading emotional biases, and ‘overconfidence’ and ‘social proof’, which both appear in our checklist of the leading cognitive errors investors can make.