Read on Twitter
The last 17 opinion polls, by 10 different companies, all give the Conservatives a double digit lead.
Of the 9 companies to have polled at least twice, 7 have the Con lead higher than in their first poll of the campaign (one static, and one, by Panelbase, lower, by one point).
Of the 9 companies to have polled at least twice, 7 have the Con lead higher than in their first poll of the campaign (one static, and one, by Panelbase, lower, by one point).
The other polling is equally dire. Most recent poll in Scotland has Cons to lose just one seat, but polls in Wales and London have them to make gains. Most constituency polling not good for Labour either.
At this point, at almost any other election in the last 50 years, you'd conclude it was basically all over, and all that remained to be decided was the size of the majority. And maybe that's right this time too.
But if I was a Labour supporter, here's what I'd be clinging to...
But if I was a Labour supporter, here's what I'd be clinging to...
1. The polls can still change. Plenty of things can happen in two and a bit weeks. Plus, if the Lib Dem vote continues to fall, that should benefit Labour, whereas the Cons probably have less scope for further squeezing the Brexit Party vote.
2. The polls could be wrong. Even by the end of the campaign in 2017, Labour trailed by an average of 8 points in the polls. But the polls under-estimated Labour's vote by five percentage points. It could happen again.
3. The Brexit Party haven't gone away. The national polls have them polling c.3%. But they're only standing in <50% of all seats. In the seats the Cons need to take, and where the Brexit Party is standing, they must be polling c.6%, and Cons prob doing less well.
4. Still a decent number of DKs in the polls. We didn't use to care too much about Don't Knows, because when they made up their minds, they used to break pretty much like those who had already decided. That didn't happen in 2017, with ‘Don’t Knows’ going disproportionately Lab.
5. Labour can win without winning. They don't need a majority, just to stop the Conservatives from getting one. The Curtice line about Labour having no chance of winning a majority has been widely misunderstood; he wasn't saying there was no chance of a Labour government.
All of these are far from certain. If polls aren't moving in Labour's direction so far, why will they start now? Undecided voters falling in number week by week. Pollsters think they've corrected their errors from last time (and anyway polls normally err in a pro-Con direction).
It wouldn't take much for things to look a little bit more competitive, although time is running out for Labour.
The latest ICM poll is the first in 10 days not to show a double digit Tory lead.
Yet don't over-react. It's Tory lead of 7 points, exactly the same as ICM had in their first poll of the campaign. We'll need a few more before we're sure the lead is tightening.
Yet don't over-react. It's Tory lead of 7 points, exactly the same as ICM had in their first poll of the campaign. We'll need a few more before we're sure the lead is tightening.
Ditto for young @roger_scully's polling in Wales, with that +9 Labour jump. It's impressive, but even now that poll has Labour on course to lose seats in Wales, albeit only just.
But still, all shows how it isn't quite as done and dusted as you might have thought a few days ago.
But still, all shows how it isn't quite as done and dusted as you might have thought a few days ago.
The last 5 UK/GB opinion polls (by 5 different companies) have all shown the Conservative lead narrowing (by between 1 and 7 points).
BUT
Of these five, four still have Con leads in double figures - and four still have the Con lead larger than their first poll of the campaign.
BUT
Of these five, four still have Con leads in double figures - and four still have the Con lead larger than their first poll of the campaign.
The last seven polling companies to have produced GB/UK opinion polls have all shown the Conservative lead narrowing.
Of the last four to do so, three now have the lead in single figures.
Sunday's polls are going to be very interesting...
Of the last four to do so, three now have the lead in single figures.
Sunday's polls are going to be very interesting...
The widespread assumption of a Tory victory - helped by the YouGov MRP figures - combined with the opinion polls actually tightening, is close to the worst of all worlds for the Conservatives.
"Close to", because it's always better to be an average of 8-9 points ahead, rather than behind, but if I was a Conservative, I'd be beginning to get nervous.
Take the nine polling companies that have done multiple UK/GB polls since the Commons voted for an election...
Their first polls had a range of Con leads from 6-16 points, with an average of 10.
Their most recent polls have a range of 6-15, and an average of 10.
Their first polls had a range of Con leads from 6-16 points, with an average of 10.
Their most recent polls have a range of 6-15, and an average of 10.
This isn't to say that nothing has changed.
The vote shares have - with the Lib Dems down.
This period also includes the Brexit party's decision to stand down; so we would have expected the Con lead to grow, and it hasn't.
The vote shares have - with the Lib Dems down.
This period also includes the Brexit party's decision to stand down; so we would have expected the Con lead to grow, and it hasn't.
Survation's latest poll changes little of the above.
The first polls after the HoC votes for an election saw the Con vote range 35-42. The same 9 companies now range 39-46.
For Labour, the shift has been from 26-31 to 31-34.
The average gap is 10, exactly as it was then.
The first polls after the HoC votes for an election saw the Con vote range 35-42. The same 9 companies now range 39-46.
For Labour, the shift has been from 26-31 to 31-34.
The average gap is 10, exactly as it was then.
