Nick Cohen makes a good point:

Political correspondents, who couldn’t find a big picture in a multiplex, buzz around Westminster hyperventilating about the number of letters sent to the 1922 Committee, and the replacement of minister X with junior minister Y, as if it mattered in the slightest.

The cliché is right. It is indeed easy to miss the wood for the trees. The fox (pdf), who knows many things, can indeed miss the big picture which the hedgehog (who knows one thing) sees. Those same political correspondents’ - despite or perhaps because of their knowledge of detail and minutiae - failed to anticipate Brexit and the rise of Corbynism.

It would, however, be wrong to single out political correspondents here. Nick’s point generalizes. Take, for example, finance. Advisors are often good foxes: they know many things. It’s better, however, to know one big thing – that active fund managers generally do not earn their fees (pdf) – and so simply hold market tracker funds. Fbuse

A similar thing works in forecasting. Sometimes, it’s better to know one big thing. For example, if you’d wanted to know where stock markets were heading this year, one fact would have told us – net foreign buying of US equities. When this is high (a low point on my chart) it predicts that markets will fall in the following 12 months*. High buying last year predicted that the market would drop this. And yes, this indicator also warned us of the 2008 financial crisis. Knowing one thing would have served us better than knowing many things.  

In saying this, I’m echoing Gerd Gigerenzer. Simple heuristics (pdf) – knowing one robust and relevant thing – beat knowing lots of little irrelevant ones. Big Facts matter. Knowledge, prediction and explanation are three different things: you can have one without the other two. 

The issue here is whether we can distinguish between noise and signal. A mastery of detail can mean that you have full knowledge only of noise. This is handy for pub quizzes, which is no small thing (I write as a member of the team that’s top of division two of the RIQL) but it can blind you to what’s important. You have only cargo cult expertise. One reason why I’ve paid little attention to the day-to-day developments in Brexit negotiations (progress might not be the mot juste)  is that I’ve not known what’s noise and what’s signal.

Rogue companies exploit this. When they are under investigation from regulators or the police they often boast that they are cooperating fully. What they mean is that they are handing over terabytes of data and hoping the authorities will miss the wood for the trees.

You might think that in siding with hedgehogs against foxes I’m disagreeing with Philip Tetlock who has shown that foxes make better predictors than hedgehogs. I’m not sure I am. We mean different things when we distinguish between the two. I’m thinking of the distinction between those who know lots of small things and those who know one big relevant one. He instead uses it to differentiate between empiricists and grand theorists. In that sense, I agree with him. The problem with that sort of hedgehog is that the one big thing they know is sometimes wrong. You might have heard of such people.

* The reasoning here is simple. Foreign buying is an indicator of investor sentiment (pdf): high buying betokens high sentiment. And when sentiment is unusually high, it tends to mean-revert downwards, thereby dragging prices down.