New improved model to see if Corbyn can win a General Election

The other day I posted a model to see if a Corbyn-led Labour Party could win a general election. I had several comments via Facebook and Twitter, some with suggestions on how to improve the model, and some questioning the assumptions. So in the spirit of the title of this blog, I have made some changes to see if more can be known, even if nothing can be proven.

Here goes with the run down on the revised model. For ease I repeated much of what I said in the original post but added a few bits in where I’ve tweaked the model. I italicised the bits where I have amended the model from last time.

1. The turnout in the 2015 general election was 66.1 per cent.
2. You can find the general election turnouts since 1945 here and you’ll see that turnout was only above 80 per cent twice: 83.9 per cent in 1950 and 82.6 in 1951. Originally my model has assumed a Corbyn-led Labour Party could get turnout up to 75 per cent, but a few people pointed out at nearly a 10 point increase this was unrealistic. So I’ve changed it to 71 per cent. But you can change this parameter yourself on the model.
3. It would be wrong to assume all the new voters because of increased turnout would vote Labour, some would be motivated to vote Tory to “stop Corbyn.” Originally I assumed 85 per cent of new voters would vote Labour. Again many people said this was unrealistic as the evidence is non voters tend to split on similar lines to the other party votes. Fair enough but it is clear Corbyn is engaging with the disaffected so it is plausible to assume the non voters who do turn out are more likely to be Labour voters, and I’ve set this at 66 per cent. You can change this parameter yourself on the model.
4. A Corbyn-led Labour Party would undoubtedly bring some disgruntled lefties who voted Green back to Labour. I have assumed 75 per cent of Green voters would switch to Labour. But not all will, as Green voting is clearly an expression of identity. This stays the same as the other model
5. My model only looks at seats in England and Wales as Scotland as things are very different due to issues around nationalism. There is no doubt a Corby-led anti-austerity Labour Party would win back many voters in Scotland, but nationalism is a strong force in Scotland, so it would be fanciful to assume all seats would be won back. I’ve assumed Labour would win 25 seats. But again you can change this if you want as the model allows.
6. In my original model I assumed all other votes remain the same. But some comments I received were that Corbyn could win back some working class UKIP voters, there is now a parameter for this. I have assumed 20 per cent of all UKIP voters will switch to Labour. You can change this parameter yourself on the model.
7. Finally, this is based on the 2015 boundaries and the Tories intend to reduce the number of seats from 650 to 600 and they will seek to get boundaries redrawn in such a way that it benefits them. So whatever the results this model shows, it will be harder for Labour if the boundaries are redrawn.

And now, on to the model. See below. Enter your values in the red boxes and the graph will change.

If you want to use it on a phone or tablet you need to double tap one of the red cells to get it to bring up the keyboard.

It shows we are in hung parliament territory, with even a Lab / SNP / Plaid / Green coalition not enough to form a majority administration. Maybe the Lib Dems (if they are still around) might be able to help out.

As I said in my original post, the Tories could be ousted but it is going to need a lot of effort to get those non voters out: they are called non voters for a reason. It seems very unlikely a Labour majority government could be formed. But then again given the pounding Labour took at the last election maybe it is too much to expect a majority Labour government in 2020 whoever is the leader.