Honorary doctorate for Nate Silver, the man who made statistics cool
TIME magazine counts him among the hundred most influential people in the world. He correctly predicted the outcome of the last two United States presidential elections. On Saturday, 14 December, the American statistician Nate Silver was in Leuven to deliver the annual Christmas Lecture organised by the Faculty of Science. He was also awarded an honorary degree on the occasion.
The interdepartmental Leuven Statistics Research Centre (LStat) celebrated its 25th anniversary with a capacity audience in the Maria Theresia College – coincidentally, 2013 is also the Year of Statistics. The guest of honour was Nate Silver (35), praised by many for bringing statistics to the general public. The honorary doctorate from KU Leuven and LStat is his first honorary doctorate from a university outside the United States. "I've received two honorary doctorates from American universities, but an honorary doctorate from one of the world's oldest universities, with such a large student body, is something extraordinary."
Nate Silver first captured the baseball world with his PECTOA algorithm designed to predict Major League Baseball players’ performance. He then founded the award-winning political website FiveThirtyEight.com, named for the number of electors in the electoral college. He correctly predicted the results of the 2008 and 2012 presidential elections on that site. In 2008, his projections were correct in all but one state. In 2012, his predictions were flawless.
In his book, The Signal and the Noise: Why Most Predictions Fail - but Some Don't, Silver uses examples from everyday life to explains why some subjects lend themselves better to predictions than others and, more importantly, how predictions become more accurate as you eliminate incorrect variables. Thus far, he has focused his statistical projections on sports and politics. Which does he prefer? "Sports. People in the United States are often too passionate about politics. Sports predictions are better for your blood pressure."
FiveThirtyEight.com was recently acquired by ESPN and now you are expanding your team. Why?
"Statistical research – especially in combination with journalism – can be applied to many area. Aside from sports and politics, we are now focusing on economics, culture and science. So I've brought in people who are good in those fields. Statistics are everywhere. I think that journalists need to learn to make better use of those statistics and learn to analyse them more critically. We want to lead the way in that."
Sports and lifestyle broadcaster ABC is a sister company of ESPN. Does that mean you'll need to focus more of on those domains?
"Not necessarily. We focus on the five issues that I mentioned. ABC produces the Oscars, but its winners are pretty difficult to predict. Both ABC and ESPN belong to the Walt Disney Company, but that does not mean that we will be able to predict whether a film will do well at the box office. We think more along the lines of tips and advice for everyday use. For travelers, for example, we find out which airline offers them the best chance to get from point A to point B on time and what company will give the best deal."
"I always think it's important to know who my audience is. By talking to different people and adjusting accordingly, your story changes. This is important for one’s own research work. I would hate to have to just be able to speak on behalf of a political elite, for example. My background has always been in statistics, writing and journalism."
According to TIME magazine you are among the one hundred most influential people in the world. What do you think of that?
"I think that we will have to recalculate that (laughs). I'm not sure that I trust that. It's flattering, but people see you differently. It brings a lot of attention, and after a while that tends to feed itself. But it also has a very positive side: I now use my personal renown – before it wanes; and it will – to expand the team around me and continue working toward the longer term."
If everything is predictable, where's the fun in life?
"Many people do not realise that many things are predictable, so we don't have to worry about that. Moreover, there are still a lot of things we can't predict. After last year's presidential elections, some had the impression that we can predict everything perfectly, but elections are more the exception than the rule in that respect. For other events, it is more a story of failed predictions than successful ones."
"Some projections predict as little as 20% of the relevant variables. The rest is luck. But even if statistics can give only very rough projections, they're still usually better than conventional wisdom or gut reactions."
Luc Vander Elst
This is a translation of the original Dutch-language article.
Rector Rik Torfs presents Nate Silver with an honorary doctorate. | © KU Leuven – Rob Stevens
The Christmas Lecture and the presentation of the honorary doctorate took place in the Main Auditorium of the Maria Theresia College. | © KU Leuven - Rob Stevens