As voting commences, here's WorldView's guide to the three big questions about polling.
Why are there no exit polls during the day?
Exit polls are generally considered one of the most accurate ways to gauge the result of a vote before the final results come out. These polls get their name as they are conducted as voters exit polling stations. This means that rather than ask someone how they might vote in the future, you are asking them how they just voted moments ago. In theory at least, there's no longer any room for the respondent to change their mind.
However, Britain has strict laws on what information can be published on election day (for example, the BBC is required to stick to simply retelling factual events). Exit polls are no exception.
Under Section 66A of the Representation of the People Act of 1983, it is unlawful to publish “any statement relating to the way in which voters have voted at the election where that statement is (or might reasonably be taken to be) based on information given by voters after they have voted.” This means that in theory, anyone who publishes exit poll data before polls close at 10 p.m. local time could face a fine or even jail time.
This doesn't mean, however, that exit polls can't be collected before 10 p.m. So, expect a lot of data to come out at 10 p.m. — along with our first real sense of how the election has gone.
What have the pre-election polls been saying so far?
It's complicated. The chart above shows a moving average of British polls, compiled by the group Britain Elects, since Theresa May took the prime minister's office last July.
As you can see, for a long time the polls showed May's center-right Conservative Party had a comfortable lead over its traditional rival, the Labour Party; on April 18, the day May called for an election, the Conservatives were ahead by an average of 16 points. Many attributed this lead to the faults of Labour's divisive leftist leader, Jeremy Corbyn, with some going so far as to suggest Britain risked becoming a one-party state.
Perhaps thanks to Corbyn's underdog status and May's missteps on the campaign trail, that remarkable lead has now evaporated. Britain Elect's average for Tuesday showed only an eight percentage-point lead. Some individual polls were far lower: One conducted last week gave them just a one percentage-point lead. It is important to note that not all polls look quite so bad — two released on the eve of the election gave the Conservatives a 12- and 10-point lead, respectively — but May's once easy-seeming path to victory doesn't look assured right now.
Complicating matters is Britain's first-past-the-post voting system, whereby Britain's citizens vote for their local politician rather than directly for a prime minister. To put it simply, this means that raw percentages don't always mean that a party will be able to win the most seats in Parliament. Estimating the number of seats is far harder than estimating raw votes.
Right now, the Conservatives hold 330 of Parliament's 650 seats, meaning they have a slim majority of just five seats. May called the election to add to that majority. Some polling firms have suggested she could still do so — ComRes predicted that she could end up with 362 seats, the biggest majority since the days of Margaret Thatcher. But another polling firm, YouGov, has released estimates based on a complicated big-data model that suggest the Conservatives could drop to 302 seats.
Such a result would be a major shock to Britain, resulting in a hung Parliament where no one party has a majority. Under this scenario, Corbyn could feasibly form a coalition with Britain's third party, the Liberal Democrats, and other smaller parties to become prime minister.
Aren't British polls always wrong anyway?
Over the past few years, Britain's polling companies have acquired an ugly reputation for inaccuracy. Many U.S. readers will remember the surprise after the Brexit vote, as many polls had failed to correctly gauge the momentum of the anti-E.U. movement.
An even bigger miss occurred the year before, however, when polls completely underestimated the Conservatives Party's support, some instead predicting a victory for Labour's then-leader Ed Miliband. This wasn't the first time Labour's chances had been overestimated: Historically, British polling companies have been accused of missing Conservative voters — something often referred to as the “shy Tory” factor.
Most polling companies conducted big overhauls to how they operate — with many coming to the conclusion that they had given too much weight to young voters, who tend to vote less than older Britons. Many have tinkered with their methodology in the hope that they can correct what once went wrong. YouGov even offered two different polling methods — the aforementioned big-data model and a more traditional poll — with very different results.
The exit polls released Thursday evening should be more accurate. But with such varied figures coming from the pre-election polls, only one conclusion is certain: Someone will be wrong.
More on WorldViews