How election maths got complicated

What will the 2019 general election be remembered for? Will it forever be the Brexit election, the first day of Britain’s socialist decade, the start of the implausible Tory-SNP coalition, somehow all of the above?

Whatever happens in a few weeks, it seems fair to say that one thing will be discussed by academics in decades to come when they look back at 12^th December: tactical voting.

It has been the phrase on everyone’s lips since the Commons finally voted to dissolve itself, and it currently seems that every seven minutes or so, a new website gets launched to tell people who they really should vote for.

If anything, asking someone which party they are planning to back next month now feels obsolete; it is all about the seat they are voting in, how every party is polling in that seat, whether that person’s candidate stands any chance of winning or retaining that seat, whether there is any other candidate that person should be voting for instead, and so on.

How did we even get here?

GROT!

It probably feels worth pointing out that tactical voting really, really isn’t new. Take the 1997 election – you know the one, historic Labour landslide, Tories on approximately eight seats, RIP Michael Portillo, yadda yadda.

What may have slipped through your net is the huge campaign urging people to vote tactically that year in order to finally get the Conservatives out of office. GROT (Get Rid Of Them) was the main group behind it, and newspapers published things like this:

“It requires only around 12,000 to vote tactically in key constituencies to give John Major his P45 from Downing Street. A further 80,000 tactical votes could give Tony Blair a working majority. In a spirit of public interest, the Independent on Sunday offers tactical advice.”

This was followed by a list of 20 seats, along with strict instructions on who to vote for based on results from 1992. According to several studies published some years later, tactical voting in ​‘97 gave the Labour party between nine and 14 extra seats, and the Liberal Democrats between 10 and 21 seats it otherwise wouldn’t have gained. Major was on his way out anyway, but this is what gave his party the spanking of a lifetime, and it only relied on between 8.5% and 10% of voters switching their preferences.

1997 wasn’t a one-off either. With the 2001 election came the first tactical voting sites; ​“Internet-literate voters have at least half a dozen websites giving a breakdown of party strengths, some even offering to facilitate a ​‘swap’ of anti-Tory votes”, the Guardian wrote at the time. Still, according to them, ​“the kind of people likely to use the internet are probably already fired up and motivated to vote tactically anyway.”

Those nerdy precursors clearly had an impact in the end, as coverage of the 2005 election once again featured a number of guesses and predictions on just how many people would vote tactically that time around.

In short: none of this is new. As political scientist and Queen Mary academic Philip Cowley explains, ​“Tactical voting has been steadily on the rise for decades. I am not sure it’s much more prominent in the media now than in (say) 2005, when it was talked about a lot, or even 1997.”

So why does it feel bigger now than it ever did?

TRIBES

According to professor Will Jennings, who teaches politics at the University of Southampton, ​“One reason tactical voting is on the agenda is the strengthening Brexit identities of many voters, and the direct consequence of the election for Brexit itself. This matters especially for remain voters who face the risk of their vote being split between the Lib Dems, Greens and Plaid Cymru in Wales – and Labour too for Remainers who want a second referendum.”

As opposed to previous elections, tactical voting this time around is not specifically about keeping the Conservatives out for the sake of it, but rather about stopping the one thing they are planning to do if they get a majority.

This changes the dynamic as the goal is clear – stop Boris Johnson’s Brexit – but it does not mean the politics of it becomes easier. For a start, Remainers are split on what they want: is it revoking Article 50 altogether? Having a second referendum? Renegotiating a softer deal? A combination of those?

On top of this, voters are likely to have their own policy preferences, and while some of them only care about stopping Brexit on 12^th December, a majority will count Brexit as only one of several priorities.

What this means in practice is that it is not as easy as saying ​“if you want the Tories out, vote for the candidate most likely to win against the Tories in your seat”. For some, the Liberal Democrats’ time in coalition means that voting for them would be a sacrifice too far; for others, Labour’s Brexit policy isn’t straightforward enough for them to overlook Corbyn’s flaws; for some remainers up north, stopping Brexit is not as important as Scotland staying in the UK, and so on.

Then there is the fact that the Labour party, the Lib Dems and the SNP currently belong in different parts of the political spectrum; being against the Conservatives is one thing, but drastically moving towards the left or the centre to stop them is another.

And even if you were ready to overlook all of the above and simply do what you must to stop Brexit, your way forward would not be immediately obvious. With so much (competing and contradictory) information online about tactical voting, who can you trust?

TECH MELTDOWN

“The world of data and politics is like the Wild West these days,” says Jennings. ​“Some of the tactical voting guides seem to match up fairly closely to what you’d expect – given everything we know about the national polling and more regional variations. However there is a serious problem with a lack of transparency over data and methods. Some are simply not replicable, and some of these guides use national polls to model local races far beyond what the data would allow with any degree of confidence.”

In more human language: some websites will take those UK-wide polls saying that, for example, the Tories are on 40% (+2) overall, Labour 30% (+3) and Lib Dems 15% (-1) and take that to mean that whatever seat you’re looking at, these figures will apply for your sitting MP and their two opponents. Which is, to be clear, not the case.

While some of the tactical voting projects may well be disingenuous and unfairly slanted towards one party, the reality is simply that there is now too much data. Between national polling, regional polling, the results of the 2017 general election, the results of the European elections earlier this year, and heaven knows what else is floating around, it can be hard to know what to base your calculations on.

The democratisation of political analysis also means that anyone with basic skills and good intentions can launch their own website full of their own figures, which doesn’t always help anyone.

Finally, there is one deeper issue that no amount of professional work can solve: Brexit has broken the party system. One interesting example of this came from the Lib Dems selecting a candidate in Canterbury, Tim Walker, to stand against Labour MP Rosie Duffield.

The move led to a number of pro-Remain activists to condemn the party, as Duffield has always been a very vocal anti-Brexit MP, and with Canterbury being a Labour-Conservative marginal, a split remain vote could lead to the seat going back to the Tories.

On the other hand, some argued that while Duffield herself was ardently pro-Remain, her party is still more moderate on the Brexit question than the Lib Dems, so someone wishing to vote only based on the strongest anti-Brexit stance should have the ability to vote for Walker.

This will be a conundrum for voters in a number of seats: how best to get your preferred outcome? Vote for, say, a Labour MP who you know will fight in your corner within a party not fully aligned to your views, or a Lib Dem whose position you’re entirely behind but who has less of a chance of winning? Or: should you vote for your local Labour MP if you know they back a harder deal and have said repeatedly they would oppose a second referendum if the party at large would deliver one?

Tactical voting works best when parties can be seen as monoliths, and reduces elections to a simple question of numbers: there are 650 cards and if you and your tribe can flip enough of them, you win. It does not matter what the fine print is on your card; you are just part of a wider game.

Maybe it worked in previous elections, but as with so many other things in our endlessly bleak political landscape: things now feel a bit more complicated than that.

(Still, if you really do feel like you want advice on how to vote tactically next month, there is a now a website comparing all the recommendations from the different tactical voting sites. In nearly 80% of cases, the different sites agree on which candidate is best, but if you live in one of the 130 seats where there is conflicting advice, it may be worth getting stuck into the details yourself and decide which candidate you want to back. Good luck!)