I meant to start the year with a general round up of where we are in the polls. Instead I've been distracted by an outbreak of red-wallism because of the Deltapoll survey in the Mail on Sunday.

The Red Wall as a concept is generally rather ill-defined. James Kanagasooriam originally used the term in a twitter thread in 2019 to refer to those seats that were demographically similar to Conservative seats (older population, lots of homeowners, etc) but where the Conservatives underperformed. These were disproportionately post-industrial seats in the North and Midlands, particularly in former mining areas. That historical antipathy to the Tories had provided Labour with a "red wall" of seats.

James Kanagasooriam @JamesKanag

The first is a huge "red wall" stretching from N Wales into Merseyside, Warrington, Wigan, Manchester, Oldham, Barnsley, Nottingham and Doncaster. When you talk about cultural barriers to voting Tory – this is where it is. This entire stretch shouldn't be all Labour but is (9/16)

Over the subsequent elections there have been massive swings to the Conservatives in some of these areas and in others - like Merseyside - it has held firm. Part of that swing will be historical factors fading away - the generation that experienced deindustrialisation moves on, old pit heads are replaced by new build estates. It is probably also the partly the impact of Brexit, upsetting and resetting old identities.

When people talk about the red wall now - and especially when they poll it - they tend to define it differently. It's often used as a term to refer to seats the Tories gained from Labour in the North & Midlands in 2019. This is definitely NOT the same as James's original definition. There are plenty of seats in James's definition that didn’t go Conservative, plenty of Lab>Con wins that weren't in James's list. However, looking at the seats the Conservatives did win in 2019, they are disproportionately in the North and Midlands and do at least share some of the characteristics of James's original Red wall seats.

Since then the so-called Red Wall has become something the political commentariat and politicians themselves have obsessed over. How will this go down in the red wall? How will Labour win back the red wall? What will voters in the red wall think of that? This tends to be based upon a perception of what the author’s idea of a stereotypical working class Conservative voter would think, rather missing the point of James’s original hypothesis that voters in those areas were actually demographically similar to more Tory areas. (For what it’s worth, such statements are normally rubbish anyway - opinions on climate change and culture war issues in the so-called “red wall” seats are generally typical of in Great Britain as a whole, if a little more negative towards immigration.)

The Deltapoll survey in the Mail on Sunday today isn’t actually a poll of the “red wall” seats - it’s a standard GB poll, but with a deliberate oversample of Tory gains at the last election. The “red wall” figures in the Mail are based on the crossbreak for the 57 seats that the Tories gained in 2019. Full tables for the poll are here.

For what its worth that suggests Labour are doing significantly better in those seats the Tories gained in 2019. It shows 9.5 percentage point swing from Con>Lab nationwide, but a 13 swing in seats they gained at the last election. This would be smashing news for Labour… but to me it looks too good to be true for Labour. The poll suggests almost no swing at all in non-Tory seats, with all the change in votes concentrated in Conservative seats, especially those won at the last election. Given it’s just an oversized crossbreak in a GB poll, I’d be a bit wary before taking that as gospel. If anything, I would expect Labour to do worse in those seats they lost in 2019 than in comparable marginals… and that is part of the reason why we are all probably paying more attention to the so-called “red wall” seats than we should.

As things stand the Parliamentary boundaries are likely to be redrawn before the next election. For the time being though we can only base it upon the boundaries we have, and looking down the list of Labour's top target seats by required swing, only around half are retaking those seats the Tories gained in the North, Midlands & Wales (the “red wall”). Almost a quarter are in the South, East or London. Those other seats are just as important to the next election as the so-called “red wall”.

On top of that, even among seats with comparable majorities, the "red wall" seats are not necessarily the easiest seats for the Labour party to gain. Historically parties have over-performed in seats they gained at the most recent election because of the incumbency effect (the Labour party will have lost any personal vote their former MP had, the new Conservative incumbent will have built one up).

Secondly the whole point of James's original "red wall" hypothesis is that these were seats that for cultural reasons were less Conservative than you would have expected given their demographics. To some degree that has unwound in some areas. There is probably not an easy way for Labour to rebuild that reluctance to consider voting Tory in places where it has collapsed. It is also worth considering whether it has even fully played out... it may be there is further realignment to go.

Thirdly, the so-called “red wall” gains the Tories made in 2019 tend to be Brexit voting seats, voters who the Conservative party are very conscious of trying to deliver for. The more southern, more graduate-heavy, less-Brexity seats on Labour’s target list may be the more winnable.

In previous election cycles we’ve often seen marginal polls, surveys of those seats that have a majority of less than 10%, or the seats that party X would need to win to form a government. These aren’t ideal - the seats with the lowest majorities are not always the most likely to fall - but it at least better than assuming the next election’s battleground will be same as the last election’s battleground.

The old cliché is that generals end up making mistakes by preparing the fight the last war. That is just what we're doing by endlessly fussing over the "red wall".

The ideal is to identify and look at those seats where the Tories are most vulnerable and/or Labour have the best chance of winning.

The second best is to look at those Conservative seats with the smallest majorities, and those seats where Labour need the smallest swing to win. Traditional "marginal seat" polls.

It is the very definition of fighting the last war to instead keep on obsessing over the seats that happened to change hands at the last election.

Anthony Wells

Stop obsessing about the Red Wall