Publications
Partners
As investors, what are we doing? We’re trying to predict the future; something that is amazingly hard, yet we attempt it as if it’s the easiest thing in the world.
Trying to determine a company’s revenue growth going forward, profit margins, tax rates and more is what investing is about.
Throw in local and global events such as three finance ministers in five days last December, currency gyrations, Fed interest rate decisions, local/global GDP, and we really have a complicated set of data that we need to use to make our predictions.
Becoming a predictor
With this in mind, I have been reading books on predicting and forecasting.
The first I read was by Nate Silver titled The Signal and the Noise: Why so many Predictions Fail – But Some Don’t.
He’d 100% predicted all the electoral college votes for the 2008 US presidential election, earning himself fame and a New York Times blog titled 538 (the number of US electoral college votes).
I enjoyed the book but didn’t feel I learnt a lot, so I went in search of another and found Superforecasting: The Art and Science of Prediction by Philip E. Tetlock and Dan Gardner.
I loved this book and feel it really helped me understand forecasting as a concept and, importantly, how we can improve our ability to do so.
But first, a disclaimer: No person is able to consistently make correct predictions.
There are simply too many moving parts in the world, especially the financial world, to get it right all the time. But we certainly can improve our forecasting and if we apply it in our investing process, it should be able to improve our returns over time.
I took three main points from Superforecasting:
1. The precision of a forecast:
Typically an investor’s forecasts are binary.
You either expect something to happen or you don’t. But life is not black and white – it is lived in the grey areas in-between and, as such, so is investing.
So you need to give exact percentage chances of something happening.
For example, if you are analysing a stock and you expect it to beat the overall market over the next five years, we need to refine that. Rather than saying that you expect the stock to beat the market, you must determine and allocate a percentage.
For example, you can work out that there’s a 65% chance of that stock beating the market over five years.
That defined percentage is a more accurate description of what is likely to happen, although nothing is certain.
How you can determine this percentage is detailed in the book – I won’t go into specifics here – but it is certainly possible to work out things such percentages if you’re prepared to put in the work.
2. Keep adjusting:
The 65% chance you have today is not fixed, rather, it is fluid as information changes. So as that information changes, make sure you analyse the new information and adjust your percentage, if necessary.
Do this as often as needed. Don’t see making adjustments as an admission of being wrong – information has changed and your forecast needs to reflect those changes.
3. Focus on time frames:
Beating the market over five years is very different to beating it over one year. Be cognisant of the time frame you are using and, of course, keep adjusting your prediction as time marches on.
A five-year prediction is very different to a five-day prediction and our predictions need to reflect this.
The book is recommended reading and while we can’t always be right, we can certainly be above-average forecasters and this will help us to be better investors.
This article originally appeared in the 25 February 2016 edition of finweek. Buy and download the magazine here.
