Nate Silver's US election prediction owes it all to Tunbridge Wells
IT WAS a punt that made a 34-year-old maths geek from Michigan a star, but the prediction that Barack Obama would retain the White House owes everything to Tunbridge Wells.
Nate Silver correctly forecast Obama would win the US presidential election, when other commentators were saying it was too close to call. While his achievement has been lauded around the world, the theories behind his feat were dreamed up 250 years ago, thousands of miles across the pond.
The Rev Thomas Bayes may not be a household name, but it is his conclusions on how to establish a mathematical basis for probability, a kind of magic formula for taking some of the guessing out of guesswork, which influence everything, from predicting race winners to life insurance and health data analysis.
For if you can factor in everything you know about a subject, then you have a better chance of predicting how things will turn out, whether it's life expectancy for an insurance policy or the next president of the United States.
The Equitable Life Assurance Company was first to use Bayes' Theorem of Subjective Probability to put its life insurance on a scientific basis. When it closed to new business in 2007, his notebook was found in its archives.
Bayes believed that, by gathering all available information and weighing up the current state of knowledge on a given subject, statisticians could use logic to work out a likely outcome.
But while Mr Silver had the benefit of high-speed 21st century technology at his disposal when he correctly nailed the outcome of voting in all 50 US states, after running thousands of computer calculations, Bayes had to work out his theories on his own, with only a notebook and pencil to assist his sums.
However, the two do appear to have shared one skill across the centuries – gambling. Before he became a media star, Mr Silver made his living as a professional poker player, while Bayes is reputed to have had a keen interest in putting his money on whatever was on offer in 18th century Tunbridge Wells, whether it was dice, cards or horses.
David Bushell, who wrote an article on Bayes for the Royal Tunbridge Wells Civic Society last year, said: "My interest grew from working in the insurance industry, and when I heard that his notebook had been bought by the Institute of Actuaries, I went to have a look. It contains a considerable amount of mathematical work, including discussions on probability, trigonometry, geometry, solutions of equations and differential calculus."
In fact, concluded Mr Bushell: "I think he enjoyed maths more than his religious role as a local Presbyterian minister."
As well as influencing engineering and gene-sequencing, it has been suggested that Bayes's work was used in cracking the Enigma Code in the Second World War and in tracking Russian submarines in the Cold War.
"His name still crops up regularly," said John Goss, of Coutts Bank in Tunbridge Wells, a member of the Bayesian Club for business and professional people, which uses Bayes's name.
Bayes was born in 1702, the oldest of seven children of a Hertfordshire Presbyterian minister. He was educated to join the church and arrived in Tunbridge Wells in 1731 as minister at the Mount Sion Chapel. He lodged at 69 London Road, on the corner of Church Road, a spot marked today by a heritage plaque.
He devised his theorem in the 1740s, but it was not widely recognised until after he died in 1862.