Thread by @centrist_phone

Last night's post-debate polls by YouGov showed a narrow Johnson win.

Johnson 52%
Corbyn 48%

Good news for Johnson, right? Well... no.

Let's look into the raw data and uncover some key points hidden by the headline numbers.

1/9

Firstly, who took part in the poll?

We have a sample of 1322 people, weighted to 2017 voting intention, EU voting intention, etc.

Looking at the numbers, Leavers however may have been over-weighted and 18-24's under-weighted. This will make the sample lean toward Con.

2/9

We should then consider the current popularity of the leaders and voting intention.

Net satisfaction ratings:

Johnson: -20
Corbyn: -40

According to the last YouGov poll, VI is

Con: 42%
Lab: 33%

In the context of who 'won', this is important.

3/9

People went into this debate:

a) Prefering Johnson to Corbyn, by a margin of 20 points

b) Planning to to vote Con over Lab, by a margin of 9 points

This should mean, if the perceptions going into the debate did not change, a very comfortable Johnson victory.

4/9

So two things must have happened for the result to be this close:

1) Many people with a negative opinion of Corbyn were pleasantly surprised

2) Many people who are not currently planning on voting Labour thought Corbyn won

But this only tells a fraction of the story

5/9

On a less divisive question of who 'won', YouGov also asked people how 'well' or 'badly' each performed.

Here, Corbyn scored a clear victory:

Johnson
62% well
38% badly
NET: +24

Corbyn
69% well
31% badly
NET: +38

This is remarkable when you consider favourability ratings

6/9

Comparing favourability ratings before the debate, with performance ratings, shows the difference:

Corbyn
Favourability: -40
Performance rating: +38
Net gain: +78

Johnson
Favourability: -20
Performance rating: +24
Net gain: +44

A net advantage of +34 to Corbyn

7/9

So, who did Corbyn need to impress?

2017 Lab voters, LD's, and Don't Knows. These are crucial for Labour to gain on Con in polls.

Here, we see an emphatic Corbyn victory:

Lab 2017
Well: 90%
Badly: 10%

LD 2017
Well: 81%
Badly: 19%

DK's
Well: 75%
Badly 25%

8/9

The question of who 'won' the debate is sensibly defined by who gained the most from it.

It's clear in the data that Corbyn out-performed Johnson relative to VI and favourability ratings going in.

It's also clear that Corbyn impressed the voters he needs to close the gap.

/END

Additional thought (1/1)

I've noticed comments about 'YouGov bias' and conspiracy theories.

Let me say this clearly - this is utter nonsense.

Differences in methodology are not bias. Pollsters do however, tend to herd. That's driven by fear of being wrong, not pro-Tory bias.