Go figure: Why nothing is really news at all
Seen the news today? It's all about what happens. In his final Go Figure column, Michael Blastland wants to know about what didn't.
Say it's reported that candy floss doubles your risk of dying suddenly. Sounds bad.
Now flip this risk around so that it's expressed as the chance of nothing happening.
The first way of looking at it is a 100% increase in risk.
The second might mean a fall in your chance of nothing happening of 0.0001%.
This is because the actual daily risk of sudden death from accident, violence or poisoning and the like is about one in a million. Double it and you get two in a million. That's your 100%.
Meanwhile, the chance of being OK might fall from 999,999 in a million, to 999,998 in a million. Suddenly, it doesn't sound so bad.
This characteristic of risk, to see things only in terms of what happens, or might happen, rather than what doesn't, is actually a bias. Sometimes what doesn't happen is as important a way of seeing the world as what does.
In a sketch by comedians Mitchell and Webb, a filmmaker (Mitchell) is interviewed about his oeuvre after a clip from his latest work - Sometimes Fires Go Out. In the film, a small fire in the kitchen - yep, you got it - goes out. That's it.
The interviewer (Webb) says the film has been reviewed as "unrelentingly real", "a devastatingly faithful rendition of how life is" and "dull, dull, unbearably dull".
He introduces another: "The Man who had a Cough and it's just a Cough and he's Fine." Two Victorian lovers meet on the station platform. The man, spluttering, looks more pallid and doomed with each encounter.
"It's just a cough," he says, stoically. Except that it is - just a cough. In the last scene, he's dandy. It is one of the finest comic sketches about probability you'll ever see. But then, not much competition.
Stories are about what happens - they're not about what doesn't. Anton Chekov said: "If in Act I you have a pistol hanging on the wall, then it must fire in the last act." If nothing happens in a story, it's a joke. But the boring truth in real life is that, usually, nothing happens. Usually, the gun isn't fired.
Likewise, a cough is not a statistically significant event, but if a man coughs in an episode of the hospital drama Casualty, it's a triple heart bypass.
Jerker Denrell teaches at Oxford Business School. He describes hearing a presentation about the attributes of top entrepreneurs. Writing in the Harvard Business review, he said the argument went as follows: "All of these leaders shared two key traits, which accounted for their success: They persisted, often despite initial failures, and they were able to persuade others to join them."
The only trouble was, said Denrell, these selfsame traits are necessarily the hallmark of spectacularly unsuccessful entrepreneurs.
The difference is that the successful ones are still around and they're the ones we look to for examples. The ones for whom success didn't happen have gone - and are often ignored.
Denrell wrote that some studies have shown a failure rate of 50% of all new businesses during their first three to five years. After rapid growth in the US tyre industry for example, the number of firms peaked in 1922 at 274. By 1936, more than 80% were gone. That is, usually, boringly, big success doesn't happen.
"Anyone studying the industry in the 1930s," said Denrell, "would have been able to observe just a very small sample of the population that had originally entered."
If you want to study success, you have to pay as much attention to those for whom it didn't happen as to those for whom it did to see if your explanations for success are exclusive to success. But business bestsellers are not known for stories of those who never made it.
One last example. In his book Picturing the Uncertain World, Howard Wainer describes the apparent success of small schools, bringing massive charitable funding to the cause of making schools smaller.
And it's true that small schools were more often at the top of the leagues than you'd expect. About 12% of the top 50 schools for maths scores came from the smallest 3% of schools overall. The only problem was that small schools were also more often at the bottom of the leagues than you'd expect - about 18% of the bottom 50.
Wainer's explanation for the small schools phenomenon was that the ability of children in small schools just bounces around a lot more from year to year because it's a smaller sample.
But if you're looking for the keys to success, maybe you don't look at the bottom. Lack of success might strike you as a non-event, if it strikes you at all.
This is the last Go Figure. It's about to become a regular non-event. Hearty thanks to all who've followed us.