Well, haven't we all got egg on our faces...

When this hit my screen, I had to take a double-take, as it just didn't make sense.

Hardly anyone believed the exit poll when it first came out.

All the opinion polls suggested that we'd have the most hung parliament in 40 years.  The suggestion that the Tories would be within striking distance of a majority government was, to say the least, surprising.  We even saw a pledge by Lord Ashdown to eat a hat on air if the poll was right.

Before the first result came in, we saw an online poll conducted by YouGov, asking voters how they voted.  That poll showed very little change between the two parties, so I (among numerous others) started considering that as the true indication of how the election went.

That's two elections in a row now where we were mistaken in disbelieving the exit poll.

As of 07:30 BST on Friday morning, with 562 of 650 results declared, this is the prediction of where the result could end up:

The Conservatives are about 50-50 for an overall majority.

The Conservatives are slightly below the odds (about 1/3) of getting to the 326 seats needed to form a majority government.  However, when you consider that Sinn Fein MPs don't turn up, this reduced the number needed to 324, which the Tories are about 50-50 on achieving.

So what happened?  There is clearly something wrong with the methodology used by the pollsters, as even YouGov day-of poll was way off the mark.  The pollsters will, as they did in 1992, have to recalibrate and figure out what to change before the 2020 election.  As for the model, once all the results are known, I will test it with the actual results in each region and see how close it got.

Barring a non-confidence motion over the course of the next parliament, the next general election is now scheduled for Thursday 7 May 2020.