Thread by @centrist_phone

The YouGov MRP.

Terrible news for Labour? No.

Dig a little deeper, and all is not what it seems.

There are numerous trends and data points that indicate that Labour are, in my view, LIKELY to deprive Johnson of a majority.

A thread - with supporting charts/data.

1/12

Let's start with the YouGov 'house effect'.

Every pollster has a different approach to crunching the data.

YouGov have consistently reported the lowest Lab % in 2019.

See attached example from this year. They were the only pollster not reporting a Lab lead for MONTHS.

2/12

And now let's look at some of the MRP data they didn't release to the public.

A leaked photo from the YouGov presentation yesterday shows a decline in the Tory lead over the last week.

I have cut off the Y axis at 7% as this is the level most agree = hung parliament.

3/12

This decline in the Tory lead has had a corresponding impact on their predicted majority.

A week ago, the MRP was predicting a 96 seat majority.

As of yesterday, that was 62 (not 68).

But a similar drop in their lead over the next week would have a dramatic impact...

4/12

The YouGov model shows a LOT of tight marginal seats.

This matters. Small increases in the Labour vote share has a big impact.

If Labour closed the gap by another 4%, as they did last week, they'd win back 28 seats.

This would reduce the Con majority to 4.

5/12

However, it seems likely, based on the trends, that they will close the gap by more than 4 points.

Con have swallowed Brexit Party. There are no more votes for them to consume.

But Labour are still devouring Don't Know, Lib Dems, and Greens - just like 2017 (image below).

6/12

So far, the polling trends mirror 2017

1) An initial Con bump as leavers unite, then flat-lining
2) A steady but sure Labour surge as Don't Knows, Lib Dems, and Greens switch

Con won't go higher. But DK's, LD's, and Green voters are switching to Lab with plenty more to go

7/12

Other fundamentals mirror 2017 too.

One is leadership ratings.

The more public see of Corbyn, the more they like him. The opposite is true of Johnson and Swinson.

This trend bodes well for Labour, and will concern the Tories who are banking on an 'anti Corbyn' campaign.

8/12

Voter registration and the interest people express in the election in polls indicates that turnout will be high.

High turnout favours Lab. Because occasional voters are more likely to vote Lab.

4.8m people registered to vote in the last two months. 3.2m were under 35.

9/12

To give you an example of the impact of turnout on the polls.

A recent Kantar poll had the Tories 11 points ahead. But if you applied 2017 levels of turnout, it dropped to 4%.

High registrations and a likely high turnout favour Labour.

10/12

In concluding, the MRP shows how fragile the predicted Tory majority is.

Remove the YouGov 'house effect' = no majority
Apply the 7% lead seen in other recent polls = no majority
Labour continue to eat DK/LD votes = no majority
Have a high turnout election = no majority

11/12

But, if you add a combination of these factors:

Remove the house effect, Labour continue to move up due to eating up DK/LD, having a high turnout election, etc.

Then it's reasonable to say that it seems v tough for Con to win a majority. And Lab could be largest party.

END