The nominations are in!
A reminder about polling...
... And its significance in this election.
In 2017 YouGov got it right. They did two types of poll. One was a general poll which was done on regional polling. Early versions of this during the campaign discounted the don't knows. Later ones guesstimated how the don't knows would vote. This polling turned out to be close to the result but not exact.
The other poll you Gov did was on a constituency level. It was right before the election and it proved to be the most accurate of all, until we saw John Curtice's exit poll (which was spot on).
This time around YouGov have just switched to a constituency version of their polling because its much more complex this time with various pacts in action. They will be promoting respondents on the basis of who is standing in their constituency.
I'm not aware of other pollsters and their methodology but YouGov is interesting because of how close they were to the result last time.
This time around we are also seeing the active use of polling to lead voters, rather than necessarily reflect it. The Lib Dems and Remain have done a lot in what they see as key marginals to aid their credibility as realistic challengers. It's a more sophisticated version of their infamous, 'Only the LDs can beat X here' barcharts of shame. But it's unlikely they will be the only ones to try and use the technique. They probably will just be a little more transparent about it.
John Curtice has gone on record as saying there are only two realistic outcomes for the election: A Tory Majority or a Hung Parliament.
For the Tories to win they need a significant lead in the polls. To be sure probably 10% lead because of the regionality and constituency anomalies. Anything less than 6 or 7 percentage ahead and it tips to a hung parliament. YouGov currently have them on 13pt lead... BUT that's without fully accounting for the 1/5 of voters who are currently undecided. Last time around those who decided at the last moment tipped heavily in favour of Labour rather than the Conservatives.
Who stays at home, or who spoils a ballot could have particular significance this time around as disenchanted voters are made up of a higher number of voters who do usually vote than usual and a broken tribalism. Thus making it more difficult to predict than ever before.
So be a bit wary of polls and what they show - and what they don't show...