Thursday, 16 January 2020

If we elect a Remainer as Labour Leader, we will lose in 2024

One of the most widespread myths about the 2019 general election is that the Labour Party lost because it lost far more Remain voters than Leave voters. Commentators frequently point to the 8pt decline in Labour’s popular vote share, and the 6pt rise in the combined vote share for the Lib Dems, SNP and Green Party, and draw the conclusion that most of Labour’s lost votes went to parties that were unambiguously pro-EU. Therefore, they say, Labour should become a party that supports re-joining the EU.

This would be entirely the wrong lesson to learn from an election in which, as I will argue, Labour’s lost Leave voters were the main reason for its defeat. It is true, yes, that Labour’s vote share declined by 8pts from 2017-19, and that the Lib Dems, SNP and Greens saw their combined vote share rise by 6pts. But these headline figures do not tell us the whole story: beneath the surface, the Tories lost Remain voters to the Lib Dems but gained Leave voters from Labour, whilst Labour lost some of its 2017 Remain voters but lost far more Leave voters. In this article I will demonstrate that Labour lost far more Leave voters in 2019 than Remain voters, and I will conclude that any Labour leader who embraces the cause of re-joining the EU would be unelectable in 2024.

Part 1: Correlation is not proof of causation 

In the 2019 general election, the net changes in vote share within Great Britain were as follows:

Con: +1.2pts
Lab: -8.1pts
LD: +4.2pts
SNP: +0.8pts
Grn: +1.1pts
BXP: +2.1pts 

At first glance, the cause of Labour’s landslide defeat would appear to be simple. Labour’s vote fell; the combined vote of the pro-EU parties rose. Labour’s neutral position on Brexit would seem, therefore, to have cost it the votes of millions of Remainers, and thus the party lost seats – even in areas that voted Leave – as a result of losing Remain voters. The solution to this, some Remainers say, is for Labour to elect a passionate Remainer as Leader who should then oppose Brexit at every opportunity and argue for Britain to re-join the EU once we have left.

But, as with most simplistic interpretations of election results, this interpretation of the 2019 result fails the most basic statistical rule: correlation is not proof of causation. Just because Labour’s vote fell by 8pts and the Remain parties’ vote rose by 6pts, that does not mean that Labour mostly lost votes to Remain parties. Rather, the opposite occurred.

The graphs below show the results of polls by Lord Ashcroft that looked at how different demographics voted in the 2017 and 2019 elections. As you can see, amongst Remain and Leave voters, the changes were as follows:

[Remain voters]
Con: -5pts
Lab: -4pts
LD: +7pts
Oth: +2pts

[Leave voters]
Con: +13pts
Lab: -9pts
BXP: +4pts
Oth: -5pts



































As you can see, amongst Leavers, there was a net swing to the Conservatives of 11%, with Labour’s vote declining by 9pts and the Tories’ vote rising by 13pts. Amongst Remainers, however, the Conservatives declined by 5pts, whilst Labour declined by 4pts. The net effect was that Labour’s lead over the Conservatives (29pts) was larger in 2019 than it was in 2017 (28pts). There was a 1% swing to Labour amongst Remain voters.

But what does all of this mean in terms of overall vote share? Well, as Lord Ashcroft provided detailed data tables, we can analyse his data and see how many Remain and Leave voters were lost in net terms by each party.

The graphs below show the political composition of the Labour vote and the Tory vote in 2017 and 2019.















 

















These figures were calculated by looking at many Remainers/Leavers voted for a particular party (e.g. 51% of Remain voters supported Labour), and seeing what percentage of the overall total those voters were. For example, Remain voters were 49% of Lord Ashcroft's sample in 2017, therefore Labour voters who voted Remain thus made up 26% of all voters. Thus we can conclude that out of Labour’s 41pt total vote share in 2017, 26pts came from Remain voters.

As you can see from the graphs, the Tories did lose Remain voters – 3pts’ worth of Remain voters, in fact – but they also gained 4pts’ worth of Leave voters, meaning that they made a net gain of +1pt.

Labour, meanwhile, lost 3pts’ worth of Remainers, gained 1pt from those who did not vote, but lost 6pts’ worth of Leavers, for a net loss of 8pts.

Thus, whilst on the surface it appears as if Labour lost votes entirely to the Lib Dems and other parties, in fact the Conservatives lost votes amongst Remainers but gained votes amongst Leavers; Labour, meanwhile, made a net loss of around 935,000 Remain voters (3% of all voters), but made a net loss of 1.9 million Leave voters (6% of all voters).

In short, Labour lost far, far more Leavers than Remainers. And because Labour’s lost Leave voters switched their vote directly to the Tories, in Tory/Labour marginals these lost Leave voters had twice the impact of losing Remain voters to the Lib Dems.

This leads us into the next part of the article – Labour’s lost seats.

Part 2: Labour lost seats in Leave areas, not Remain areas

In 2017 - not including Chorley, which is now the Speaker's seat - Labour won 254 seats within England and Wales. 157 of these seats (62%) voted Leave in 2016, whilst 97 voted Remain (38%).

In 2019, Labour won 201 seats within England and Wales (-53). Labour lost a total of 54 seats to the Conservatives, and of these 54 seats, 52 of them voted to Leave the EU in the 2016 referendum. In other words, Labour lost 54 of its 157 Leave seats (33%) but just 2 of its 97 Remain seats (2%).

The average Leave vote in the 54 seats that Labour lost to the Tories was 60.1% - 8pts higher than the national Leave % in 2016. It is an irrefutable fact that Labour lost the election because it lost votes in Leave areas, and as we have seen, Labour lost more Leavers than Remainers overall. The fact that Labour lost more Leavers than Remainers in net terms thus had a greater impact in these seats, as the % of Leave voters in these seats was higher than in the rest of the country.

The table below lists these 54 seats, along with the changes in vote share for each party.


The graph below shows the average swing to the Conservatives in constituencies that Labour won in 2017, grouped by the size of the Leave vote in 2016. As you can see, the higher that the Leave vote was in a constituency that Labour won, the larger the swing to the Conservatives.
















Losing Leave seats was always a risk for the Labour Party if we alienated Leave voters, as the overwhelming majority of Labour’s most marginal seats voted Leave in 2016. Of the 90 seats in England and Wales that Labour won in 2017 by a margin of 20pts or less, 73 of them (81%) voted Leave in 2016. The average Leave vote was 55.9%. The table below lists these seats

For comparison, the 165 seats in England and Wales that Labour won in 2017 by a margin of 20pts or more were far more evenly split: 85 of them voted Leave in 2016 (52%), whilst 80 voted Remain (48%). The average Leave vote was 48.8%. The table below lists these seats.


Losing a Leave voter to the Tories in Great Grimsby (where 71% of voters voted Leave in 2016, and where Labour had a majority of 2,565 votes) thus had far more of an impact than losing a Remain voter to the Greens in Bristol West (where 79% of voters voted Remain in 2016, and where Labour had a majority of 33,215 votes).

In short, the seats that Labour lost to the Tories were more pro-Leave (with an average Leave vote of 60.1%) than the country as a whole, and Labour's most marginal seats were more pro-Leave (with an average Leave vote of 56%) than its relatively more safe seats (with an average Leave vote of 49%).

Losing more Leave voters than Remain voters thus had a magnified effect in these marginal seats.

Part 3: is a Remainer really electable in 2024?

As we saw in Part 1 of this article, Labour cannot win over 40% of the vote without winning at least 25% of Leave voters, which means that we need to win back the 1.9 million Leave voters that we lost in 2019. And as we saw in Part 2, nearly all of the seats that we lost to the Tories voted Leave, with the average Leave vote in our lost seats being 8pts higher than the country as a whole. Despite this, many commentators argue that Labour will only win the next general election by becoming a party that supports re-joining the EU, and by winning back lost Remain voters.

But the numbers simply aren’t there for this strategy to have any hope of succeeding. To become the largest party in 2024, Labour needs to gain around 80 seats from the Conservatives in England and Wales. But of the 80 most-marginal Labour target seats won by the Tories in 2019, 67 of them (84%) voted Leave. The table below lists these seats.

Not only that, but the Remain voters that Labour lost in 2019 did not vote for the Conservatives; as we saw in Part 1, the Tories also declined amongst Remainers. Focusing on Remain voters will not reduce the Conservatives’ share of the popular vote, which in 2019 was a historic 45% of the vote.

Even if the next Leader wins back Labour’s lost Remain voters, Labour would still only win 36% of the popular vote (+3pts) and 218 seats (+16). Because Labour did not lose Remain voters to the Tories, gaining Remain voters does not reduce the Tory vote share by a single vote; the Tories would still win by a 9pt margin, and would win a majority of 50 seats.

By contrast, if Labour won back its lost Leave voters, Labour would win 39% of the popular vote (+6pts), winning 266 seats (+64). Because Labour would be gaining Leave voters directly from the Tories, the Tory vote share would fall to 39% (-6pts), leading to a hung parliament in which Labour would be able to form a minority government.

The only way to actually win the next general election is by winning back the support of the Leave voters that we lost in 2019, and winning new Leave voters directly from the Conservatives. Alienating Leave voters further by seeking to overturn the democratic choice that they made in 2016, and focusing on winning the support of an increasingly small group of quixotic Remainers, is not a route to victory. It is a recipe for defeat.

Conclusion

In conclusion, let’s review the facts:

  • We lost twice as many Leave voters (-6pts) as Remain voters (-3pts) in 2019
  • 52 of the 54 seats that we lost to the Tories in 2019 voted Leave in 2016
  • Winning back the Remain voters that we lost in 2019 will still result in a defeat
  • 84% of the seats that Labour needs to gain from the Tories voted Leave in 2016

It is difficult to conclude, in light of this evidence, that electing a Remainer or a re-joiner as our Leader will lead to victory in 2024. Rather, for Labour to focus on winning the support of a small group of enthusiastic Remainers will simply have the effect of leading us to another defeat

By 2024, the question of whether we are going to Leave the EU or not will have been definitively settled. The debate will be over. It would be a mistake for Labour to elect a Leader who will try (and fail) to re-open that debate; and it would also be a mistake to elect a Leader whose major contribution to the 2019 election was a 2nd referendum policy that led to us losing 1.9 million Leave voters (and thus losing the election).

As much as many Labour members might want to stay in the EU, it is very clear now – after Leave won a referendum, a general election, a European Parliament election and another general election – that the country does not. It is time that we finally respected that democratic decision, allowed the country to move on, and articulated a vision for a progressive future outside of the EU.

Wednesday, 8 January 2020

How accurate were the polls in the 2019 election?

British opinion polls have had a difficult few years. On average, they predicted the wrong outcome in the 2015 general election, the 2016 EU referendum and the 2017 general election. It was probably unsurprising, therefore, that many commentators - myself included - expected the election result to be surprising in one way or another.

As it turned out, the most surprising thing of all happened: the polls were actually right.

On average, the polls in the final week of the election (9th-12th December) showed the following results:

Tories 43%, Labour 34%, Lib Dems 12%, SNP 4%, Greens 3%, Brexit Party 3%

And on December 12th, the actual election result within Great Britain was:

Tories 44.8%, Labour 32.9%, Lib Dems 11.8%, SNP 3.9%, Greens 2.8%, Brexit Party 2.1%

The average polling error for the three major parties in the final week was plus or minus 1.03pts, the lowest average polling error since 1955 - so, in other words, the polls were more accurate in 2019 than in any election for over 60 years. That's impressive. But what was even more impressive were some of the results for individual pollsters. Let's go through the results for individual pollsters and see how accurate they were.

Opinium was the most accurate

Many people were surprised by the large Conservative leads that Opinium was showing throughout the campaign, but in the end, they were proven right - exactly right. As you can see in the graph below, Opinium's final poll predicted 45% for the Tories, 33% for Labour and 12% for the Lib Dems. They estimated that the Tories would win by 12pts; the Tories won by 11.9pts.

Opinium's average error for the top 3 parties was therefore just plus or minus 0.2pts, the lowest average polling error of any pollster in 2019.

The table below shows the final poll from each pollster, and the average error for the top 3 parties. Note that I have calculated the average error for Qriously by comparing it to the UK-wide result, as unlike other pollsters, Qriously polled the whole of the UK.




















Opinium, Ipsos MORI, Number Cruncher Politics and Kantar Public all had average errors of less than 1%. All pollsters estimated that the Conservatives would win - the only difference between the pollsters was in how big the Tory margin of victory was estimated to be.

Overall, the pollsters performed very very well, and the average of polls overall was more accurate than in any election since 1955.

Tuesday, 3 December 2019

Polling round-up: the 2019 general election, week 4

This article was made possible by our Patreon supporters! Special thanks in particular go to our amazing $5 supporters: Vincent Calabrese, Scott Folan, Josuke Higashikata, Chloe Hopkins, Harry Jackson, Stephen Kaar, Bert Rothkugel, sn, Alan Spence, Alex Wilson

To support Stats for Lefties on Patreon, visit https://www.patreon.com/leftiestats


--

The 2019 general election is nearly over. Just 9 days remain until Britain will vote in its 3rd election in just 4 years, continuing a journey that began with the unexpected 2015 election result and was followed by the unexpected 2016 referendum result, which itself was followed by the unexpected 2017 election result. In all three cases, most opinion polls completely failed to predict the actual result: the polls predicted a hung parliament in 2015, a victory for Remain in 2016, and a Tory majority in 2017. None of those things happened.

Week 4 of the 2019 election campaign just ended, and Labour clearly did well last week. After only increasing their vote share by 1pt in Week 2 and Week 3, Labour's average vote share in polls rose by 3pts last week. It is now averaging 33% - exactly the same as in Week 4 of the 2017 election campaign. The Tories, meanwhile, appear to have finally peaked at 43%, and are doing worse than at this point in 2017 (when they were averaging 46%).

So, without further ado, let's look at the numbers.

What the polls looked like in Week 4 (25th November-1st December)

In the week of 25th-1st December, the average of polls was (changes since the previous week): 

Tories 43% (-), Labour 33% (+3), Lib Dems 13% (-2), Brexit 4% (-), SNP 4% (-), Greens 3% (-)


According to my seat estimate, the seat result would be (changes since the previous week):

Tories 348 (-12), Labour 219 (+13), SNP 41 (-), Lib Dems 19 (-1), Others 23 (-) 

The Tories' estimated majority, which last week was 70 seats, has now fallen to 46 seats. Labour have increased their vote share and estimated seat total (the only party to do so).

The Liberal Democrats, meanwhile, have collapsed further, falling to just 13% in the polls and 19 seats in my seat estimate. It is worth bearing in mind that, 7 months ago, the Lib Dems' estimated seat total was 60 seats. It is now just 19 (-41). Amongst Remain voters, the Lib Dems have collapsed still further, whilst Labour is winning the support of almost 50% of Remainers.

Amongst Remain voters, voting intention in Week 4 of the campaign was:

Labour 48% (+4), Lib Dems 23% (-3), Tories 18% (-), SNP 5% (-), Green 3% (-1), Brexit 1% (+1)

Labour's share of the vote amongst Remainers (48%) is the party's highest in a weekly average since the week of 28th February-3rd March, whilst the Lib Dems' share of the vote (23%) is their lowest in a weekly average since 20th-26th May.

Amongst Leave voters, voting intention in Week 4 of the campaign was:

Tories 70% (+1), Labour 16% (+2), Brexit 7% (-1), Lib Dems 3% (-1), Grn 2% (+1, SNP 1% (-1)

How is Labour doing compared to this point in 2017?

MPs voted to hold the 2017 general election on 19th April 2017, so the period of time most comparable to Week 4 of the 2019 campaign was the week of 15th-21st May 2017.

At this point in 2017, Labour was averaging 33%. The Tories’ average poll lead was 13pts, and their estimated majority was 76 seats. By comparison, the Tories’ average poll lead this week was 10pts, and their estimated majority was 46 seats.

Over the following two weeks, Labour support rose steadily until the party was averaging 37%, as shown in the chart below. Labour then outperformed these polls and won 41% within Great Britain in the 2017 election.

Whilst the 2017 campaign was 1 week longer than the 2019 campaign, the final week did not make any difference to the polls; in fact, Labour's average vote share in the final week (36%) was 1pt lower than in Week 6 of the 2017 campaign (37%).


Thus far in 2019, Labour's vote share week-to-week has exactly matched its vote share week-to-week in 2017, as shown in the graph below. If this trend continues, then we should be averaging 35% by the end of this week, and 37% by 12th December. Even if the Tory vote share does not change, this should put us within 6pts of the Conservatives by election day. This would almost certainly result in a hung parliament, with the possibility of Labour becoming the largest party if the polls are slightly wrong (as they were in 2015, 2016 and 2017).


Tuesday, 26 November 2019

Polling round-up: the 2019 general election, week 3

This article was made possible by our Patreon supporters! Special thanks in particular go to our amazing $5 supporters: Vincent Calabrese, Scott Folan, Josuke Higashikata, Chloe Hopkins, Harry Jackson, Stephen Kaar, Bert Rothkugel, sn, Alan Spence, Alex Wilson

To support Stats for Lefties on Patreon, visit https://www.patreon.com/leftiestats


--

The 2019 general election is now in full swing. Both major parties have released their manifestos, the debates have begun, and the minor parties are collapsing.

The Brexit Party, in particular, has almost completely collapsed; this is partly due to the pollsters adjusting their methodology to take account of the Brexit Party's decision not to stand in the 317 seats that were won by the Tories in 2017. Unsurprisingly, the Tories have gained the most from the Brexit Party's collapse, but with 2-and-a-half weeks to go the Tories are running out of Leave votes to gain and are doing worse than at this point in 2017.

Labour's policies, meanwhile, have been recieved very positively - and its Brexit policy is actually the most popular policy of all the parties. This election is still wide open, and either major party could win. So let's take a look at the polls.

What the polls looked like in Week 3 (18th-24th November)

In the week of 18th-24th November, the average of polls was (changes since the week of 11th-17th November): 

Tories 43% (+2), Labour 30% (+1), Lib Dems 15% (-), Brexit 4% (-2), SNP 4% (+1), Greens 3% (-)


According to my seat estimate, the seat result would be (changes since the week of 4th-10th November): 

Tories 360 (+3), Labour 206 (-2), SNP 41 (-), Lib Dems 20 (-1), Others 23 (-)

The Tories’ gains last week were again almost entirely at the expense of the Brexit Party, partly because pollsters made changes to their methodology to account for the fact that the Brexit Party is not standing in the 317 seats that were won by the Tories in 2017. Overall, both major parties gained votes over the past week, and largely stood still relative to one another.

Amongst Leave voters, voting intention in 11th-17th November was, on average, as follows (changes since 4th-10th November):

Tories 69% (+2), Labour 14% (-), Brexit 8% (-3), Lib Dems 3% (-1), SNP 2% (-), Grn 2% (+1)

Amongst Remain voters, voting intention was:

Labour 44% (-), Lib Dems 26% (-), Tories 18% (-1), SNP 5% (-), Green 4% (-1)

Whilst the Tories are increasing their vote at the expense of the Brexit Party, Labour also continues to increase its support. The party is now averaging 30% in polls; in the week before MPs voted for an election, the party was averaging 23%.

The Lib Dems have not had a good week. Their average vote share in polls remains 15%, their lowest vote share in a weekly average since 20th-27th May (when they averaged 13%), and their estimated seat total has fallen to just 20 seats – which would represent a net gain of just 8 seats compared to 2017. Labour is now estimated to win 208 seats to the Lib Dems' 20.
For Jo Swinson to fulfill her ambition of becoming Prime Minister, she would now need to increase the Lib Dems' support by over 10 million votes in less than three weeks. That simply isn't going to happen. Either Boris Johnson or Jeremy Corbyn will win this election. Nobody else can.

Labour's policies are popular

Last week, Labour released its 2019 manifesto - and just like in 2017, the party's policies were recieved favourably by the public. DeltaPollUK polled the public to ask what they thought of particular policies, and these were the results:


All of the Labour policies listed in the poll had the support of a plurality of voters, and all but two (free broadband and Labour's Brexit policy) had the support of a majority of voters. The party's pledge to increase health spending by 4.3% a year is the most popular policy, winning the support of 81% of voters. Notably, its policy of raising taxes for those earning over £80,000 a year (whilst freezing taxes for everyone else) is also overwhelmingly popular.

Labour's Brexit policy, which has often been dismissed by centrists as a fudge that pleases nobody, is in fact the most popular Brexit policy out of all three major parties - and is the only policy that has more supporters than opponents.


Surprisingly, Labour's Brexit policy is actually more popular amongst Remain voters than the Liberal Democrats' pledge to revoke Article 50.


How is Labour doing compared to this point in 2017?

MPs voted to hold the 2017 general election on 19th April 2017, so the period of time most comparable to Week 3 of the 2019 campaign was the week of 8th-14th May 2017.

At this point in 2017, Labour was averaging 30%. The Tories’ average poll lead was 17pts, and their estimated majority was 110 seats. By comparison, the Tories’ average poll lead this week was 13pts, and their estimated majority was 70 seats.

Over the following few weeks, Labour support rose steadily until the party was averaging 36-37%, as shown in the chart below. Labour then outperformed these polls and won 41% within Great Britain in the 2017 election.


Summary

In last week's article, I wrote that:

"This week will see the publication of the 2019 manifesto and the Leaders’ debates, and all we need this week to continue on the 2017 trajectory is a gain of 1pt – and if the manifesto doesn’t deliver at least a 1pt increase in Labour’s vote share, I will be amazed."

Even though the manifesto usually takes about a week to affect the polls, Labour's vote share did indeed increase by 1pt this week, meaning that the party has already gained 7pts in three weeks. The party is performing just as well as it did at this point in 2017, and it is winning back Remainers at the speed of light, whilst the Tories are doing worse than they were doing at this point in 2017.

With two-and-a-half weeks left, we are closer to the Tories in the polls than we were at this point in 2017. The Lib Dems are collapsing, Remainers are flocking to the Labour Party, and the Tories have maxed out their Leave vote whilst alienating Lib Dem-leaning Tory Remainers.

We are going to win.

Monday, 25 November 2019

Remainers aren’t voting for the Lib Dems: they now support Labour

This article was made possible by our Patreon supporters! Special thanks in particular go to our amazing $5 supporters: Vincent Calabrese, Scott Folan, Josuke Higashikata, Chloe Hopkins, Harry Jackson, Stephen Kaar, Bert Rothkugel, sn, Alan Spence, Alex Wilson

To support Stats for Lefties on Patreon, visit https://www.patreon.com/leftiestats


--

We are the party of Remain”, said Jo Swinson when she was elected as Leader of the Liberal Democrats in June 2019.

But Remainers don’t seem to agree.

Ever since the 2019 general election began, the Liberal Democrats – running on a manifesto pledge to revoke Article 50 – have been rapidly losing the support of Remain voters, with Remainers overwhelmingly rallying behind the Labour Party’s second referendum policy as the best and most credible way to stop Brexit. Let’s take a look at the numbers.

1) Remainers support Labour’s Brexit policy

In a poll conducted on 21-23 November, DeltaPollUK found that more Remain voters support Labour's policy of holding a referendum on a renegotiated Brexit than support revoking Article 50.



Furthermore, in September 2019, YouGov asked voters in a poll how they would feel if Labour supported particular policies on Brexit. Amongst Remain voters, 59% said that they would feel delighted/pleased if Labour adopted a policy of holding a referendum on a renegotiated Brexit deal; just 17% said that they would be disappointed/angry.

2) Remainers support Labour now, not the Lib Dems

Following the European Elections, in which the Liberal Democrats came second with 20%, Remain voters increasingly began to support the Lib Dems in opinion polls. This support reached its highest in the week of 23rd-29th September, when 38% of Remain voters said that they would vote Lib Dem – a 4pt lead over the Labour Party. The following weeks saw Labour and the Lib Dems tied for first place amongst Remainers.


But once the 2019 general election begun, Remainers began to rally behind the Labour Party; Labour’s support amongst Remainers (44%) is now at its highest since the week of 8th-14th April (when it was 48%).


In the past four weeks, Labour’s support amongst Remainers has risen from 34% to 44% (+10). With two-and-a-half weeks left before polling day, and with Labour’s support amongst Remainers rising by an average of 3pts per week, it is looking increasingly likely that Labour will equal (or even surpass) its 2017 vote share of 55% with Remainers.

So much for the idea that Remainers will never vote for Corbyn!

3) Remainers are liking Corbyn more and more, and Swinson less and less

In the week before MPs voted to hold the 2019 election (21st-27th October) Jo Swinson had an average net approval rating with Remain voters of +13pts, whilst Corbyn had an average net approval rating of -30pts. In the weeks since then, this gap has narrowed. Here’s Corbyn’s average approval rating amongst Remainers over time


Meanwhile, here is Jo Swinson’s approval rating amongst Remainers:


Swinson’s net approval rating has fallen from +13pts to +7pts in just a few weeks, whilst Corbyn’s net approval rating has risen from -30pts to -15pts.

After this week’s TV debate between Boris Johnson and Jeremy Corbyn, meanwhile, Remainers were very positive about Corbyn. According to a YouGov poll of voters who watched the debate, amongst Remain voters:

-> 82% thought that Corbyn came across as more in touch with ordinary people

-> 77% thought that Jeremy Corbyn performed best overall

-> 73% thought that Corbyn came across as more trustworthy

-> 60% thought that Corbyn came across as more likeable

-> By a 16pt margin, they thought that Corbyn came across as more Prime Ministerial

-> By a 6pt margin, they thought that Corbyn performed best during the Brexit section of the debate

Conclusion

In this context, the argument that the Lib Dems are now the tactical vote choice in strongly pro-Remain areas is beginning to look very outdated. Remainers are overwhelmingly supporting the Labour Party now, not the Lib Dems; and rightly so, as only Labour can actually deliver a People’s Vote and give Britain the opportunity to remain in the European Union.

In 88% of the most marginal seats that were won by the Tories in 2017, Labour came second; Labour is currently averaging 30% in polls to the Lib Dems' 15%; most Remainers support the Labour Party, not the Liberal Democrats; and more Remainers prefer the Labour Party's Brexit policy than prefer the Lib Dems' Brexit policy.

All the available evidence shows that Labour is the party of Remain - not the Lib Dems.
If you're a Remainer, the choice is clear: either we leave the EU with Boris Johnson, or have a 2nd referendum with Jeremy Corbyn. There is no other possible outcome. The choice is yours.

Monday, 18 November 2019

Polling round-up: the 2019 general election, week 2

This article was made possible by our Patreon supporters! Special thanks in particular go to our amazing $5 supporters: Vincent Calabrese, Scott Folan, Josuke Higashikata, Chloe Hopkins, Harry Jackson, Stephen Kaar, Bert Rothkugel, sn, Alan Spence, Alex Wilson

To support Stats for Lefties on Patreon, visit https://www.patreon.com/leftiestats


--

The 2019 general election is underway! The Labour manifesto will be announced this week, and the Leaders’ debates are about to begin. This week will be very important. But for now, let's look at last week's polls.

Unsurprisingly, the Brexit Party’s decision to stand down in seats that the Tories won in 2017 led to the Conservatives gaining in the polls, but Labour gained votes too, and the Tories are still doing worse than at this point in 2017.

Centrists are once again trying to label Corbyn as unelectable before a single vote has even been cast. But they’re wrong.

What the polls looked like in Week 2 (11th-17th November)

In the week of 11th-17th November, the average of polls was (changes since the week of 4-10 November):

Tories 41% (+3), Labour 29% (+1), Lib Dems 15% (-1), Brexit 6% (-3), SNP 3% (-1), Greens 3% (-)


According to my seat estimate, the seat result would be (changes since the week of 4th-10th November):

Tories 357 (+19), Labour 208 (-11), SNP 41 (-3), Lib Dems 21 (-5), Others 23 (-)

The Tories’ gains this week have been almost entirely at the expense of the Brexit Party, partly because pollsters have made changes to their methodology to account for the fact that the Brexit Party is not standing in the 317 seats that were won by the Tories in 2017. Amongst Leave voters, voting intention in 11th-17th November was, on average, as follows (changes since 4th-10th November):

Tories 67% (+7), Labour 14% (+1), Brexit 11% (-7), Lib Dems 4% (-), SNP 2% (-), Grn 1% (-1)

Amongst Remain voters, voting intention was:

Labour 44% (+3), Lib Dems 26% (-3), Tories 19% (+1), SNP 5% (-), Green 5% (+1)

Whilst the Tories are increasing their vote at the expense of the Brexit Party, Labour also continues to increase its support – in Labour’s case, at the expense of the Lib Dems.

The Lib Dems have not had a good week. Their average vote share in polls has fallen to 15%, their lowest vote share in a weekly average since 20th-27th May (when they averaged 13%), and their estimated seat total has fallen to just 21 seats – which would represent a net gain of just 9 seats compared to 2017.

This is a massive loss of support for the Lib Dems. Just a few months ago, they seemed to be on the verge of a breakthrough. In the week of 10th-16th June, the Lib Dems were averaging 21% in polls and their seat estimate was 60 seats (they won 12 seats in 2017). Their support has now collapsed, and the party is losing more and more Remain voters every week, as the chart below demonstrates.


How is Labour doing compared to this point in 2017? 

MPs voted to hold the 2017 general election on 19th April 2017, so the period of time most comparable to Week 2 of the 2019 campaign was the week of 1st-7th May 2017.

At this point in 2017, Labour was averaging 29%. The Tories’ average poll lead was 18pts, and their estimated majority was 116 seats. By comparison, the Tories’ average poll lead this week was 12pts, and their estimated majority was 64 seats.

Over the following few weeks, Labour support rose steadily until the party was averaging 36-37%, as shown in the chart below. Labour then outperformed these polls and won 41% within Great Britain in the 2017 election.


What is notable about the 2017 campaign is that, whilst Labour’s support rose significantly over the course of the campaign (+10pts) it never rose by more than 3pts (on average) in any particular week. Indeed, aside from the week of 15-21 May 2017, Labour’s support never rose by more than 2pts.

Therefore, even if we are on course for another 2017-style Labour resurgence, we should not expect to see significant changes in any particular week. In fact, Labour’s gains in the week of 11th-17th November (+1pt) are exactly what we should be expecting to see if Labour’s 2017 gains are being repeated.

Summary

In short, Labour is gaining in the polls every week. It has gained 6pts in just three weeks, it is performing just as well as it did at this point in 2017, and it is winning back Remainers at the speed of light.

All in all, last week went okay. The Brexit Party’s decision to stand aside in seats won by the Tories in 2017 unsurprisingly led to an increase in the Tories’ vote share, but Labour’s vote has also increased. This week will see the publication of the 2019 manifesto and the Leaders’ debates, and all we need this week to continue on the 2017 trajectory is a gain of 1pt – and if the manifesto doesn’t deliver at least a 1pt increase in Labour’s vote share, I will be amazed.

We are going to win.

Wednesday, 13 November 2019

Polling round-up: the 2019 general election, week 1

This article was made possible by our Patreon supporters! Special thanks in particular go to our amazing $5 supporters: Vincent Calabrese, Scott Folan, Josuke Higashikata, Chloe Hopkins, Harry Jackson, Stephen Kaar, sn, Alan Spence, Alex Wilson

To support Stats for Lefties on Patreon, visit https://www.patreon.com/leftiestats


--
 
The 2019 general election has begun. With Parliament unwilling to pass his Brexit deal by his self-imposed deadline of October 31st, Prime Minister Boris Johnson called for a general election, and the Labour Party agreed.

The campaign has begun – and this time, Labour can win.

Let’s take a look at what the polls looked like just before the election began, and what’s changed over the past week (4th-10th November).

What the polls looked like before the election

In the week before Parliament voted for an early general election, the Tories had good reason to feel confident about winning a snap election. In the week of 21st-27th October, their poll lead was impressive: they led Labour by 15pts, which on a universal swing would result in a Conservative majority of 74 seats – the biggest Conservative majority since 1987.


However, this has rapidly changed.

What the polls look like now

In the two weeks since then, Labour has rapidly improved its share of the vote, whilst the Tories’ share of the vote (38%) has not changed.


In 14 days, Labour’s average vote share has risen from 23% to 28%, a change driven largely by gains amongst Remain voters. In the week of 21st-27th October, voting intention amongst Remain voters was:

Labour 34%, Lib Dem 33%, Tories 18%, Green 7%, SNP 6% 

But by the week of 4th-10th November, this had shifted to:

Labour 41% (+7), Lib Dem 29% (-4), Tories 18% (-), SNP 6% (-), Green 5% (-2) 

Labour has also increased its vote share amongst Leave voters from 10% to 13%.

As a result of these Labour gains, the Tories’ lead has fallen from 15pts to 10pts after just two weeks. The overall changes are:

Tories 38% (-), Labour 28% (+5), Lib Dem 16% (-2), Brexit 9% (-2), SNP 4% (-), Green 3% (-1)

As a result, the Tories’ estimated majority has substantially declined. In the week of 4th-10th November, the Tories’ estimated majority was just 28 seats – down from 74 seats just two weeks ago.

It is worth bearing in mind that, at a similar stage in the 2017 election campaign, the Tories’ estimated majority was an astonishing 114 seats. They went on to lose their majority just 5 weeks later.

How is Labour doing compared to this point in 2017?

MPs voted to hold the 2017 general election on 19th April 2017, so the period of time most comparable to Week 1 of the 2019 campaign was the week of 24th-30th April 2017. At this point in 2017, Labour was averaging 28%. The Tories’ average poll lead was 18pts, and their estimated majority was 114 seats.

Over the following few weeks, Labour support rose steadily, as shown in the chart below.


What is notable about the 2017 campaign is that, whilst Labour’s support rose significantly over the course of the campaign, it never rose by more than 3pts (on average) in any particular week. Indeed, aside from the week of 15-21 May 2017, Labour’s support never rose by more than 2pts. Therefore, even if we are on course for another 2017-style Labour resurgence, we should not expect to see significant changes in any particular week.

Instead, as long as Labour’s support is consistently rising by 2-3pts every single week, we should be in a very good position to win on December 12th.