Here’s a lesson from the U.K. election: follow the money, not the polls.
David Cameron confounded polls showing his Conservatives were neck and neck with Ed Miliband’s Labour Party and won a second term as prime minister, winning a majority in the House of Commons. As far back as February, betting firms made Cameron favorite to win.
“Punters were piling into a Tory majority at 10/1 on the day, when we thought it had no chance and some political scientists put it at exactly a zero percent possibility,” said Matthew Shaddick, head of political odds at Ladbrokes Plc.
Ladbrokes paid out 2 million pounds ($3.1 million) to clients who placed bets on outcomes such as Cameron winning a majority, winning most seats or becoming prime minister. Meanwhile, the British Polling Council set up an inquiry into what went wrong after Cameron publicly scorned the polls as he celebrated with party workers in London on Friday.
Apologizing for a poor performance, Stephan Shakespeare, chief executive officer of YouGov Plc, wrote in a Twitter post that it had been a “terrible night” for the polling company.
“We need to find out why,” he said.
YouGov shares fell 0.4 percent in London trading Friday, even as U.K. shares rallied the most among western-European markets after the Conservatives won an overall majority.
The last day there was more than a 1 percentage-point gap between the Tories and Labour in the daily YouGov poll was April 25. The vote share after all 650 seats had been counted was 36.9 percent for the Conservatives and 30.4 percent for Labour.
“The final opinion polls before the election were clearly not as accurate as we would like,” the Polling Council said on its website. “The fact that all the pollsters underestimated the Conservative lead over Labour suggests that the methods that were used should be subject to careful, independent investigation.”
One possible explanation is that pollsters are finding it harder to get the right survey sample, said Nate Silver, who founded the stats-oriented news website FiveThirtyEight.com.
“Polls, in the U.K. and in other places around the world, appear to be getting worse as it becomes more challenging to contact a representative sample of voters,” Silver said in a blog post Friday. “That means forecasters need to be accounting for a greater margin of error.”
A second challenge for pollsters is translating vote shares into parliamentary seats. Under Britain’s seat-by-seat, winner-take-all system, small shifts in vote share can mean big moves in the number of seats.
Some pollsters argue that they weren’t that wrong after all, but the media over-interpreted the numbers to infer a tied race. Vote shares are subject to a margin of error of plus or minus 3 points, according to Andrew Hawkins, chairman of polling firm ComRes.
“Some commentators have been very quick to put the boot into pollsters for calling it wrong,” Hawkins said. “The truth is that pollsters, when they stick to their knitting, measure vote share. Most of the polls from most of the pollsters were within the margin of error.”
It could also be that maybe the polls weren’t that wrong after all, but voters switched sides or made their final choice at the last moment, according to Tim Bale, a professor of politics at Queen Mary, University of London.
“Maybe they got to the polling booth and began to worry about Miliband,” said Bale. “Or they balked at the possibility of SNP in government.”
Gamblers did better. Last month, most betting websites narrowly made Cameron favorite over Miliband.
While the favorite tag moved back and forth between Cameron and Miliband in the home stretch, many bookmakers cut the odds on the Conservative leader as money flooded in.
Betfair, for example, cut Cameron’s odds to 5/6 from even in London. That means a successful 6 pound bet won 5 pounds; previously the same bet would have won 6 pounds. Paddy Power Plc, Ireland’s largest bookmaker, also cut Cameron’s odds after taking a 30,000-pound bet on the Conservative leader.
“The vast majority of money we took was on Cameron to win, even if we took a slightly opposite view,” said Shaddick at Ladbrokes. “Punters were much more confident and right on the number of seats and Cameron winning than the pollsters were. That may be because they are more in touch with sentiment on the ground, it may be gut feeling. It may just be pure luck.”