(Above, the current poll averages extrapolated into a prediction, courtesy of Electoral Calculus)
Back in 2017 I made my predictions for that snap election. Here’s what I predicted and what actually transpired:
Conservative 306 (down 24) - Actual result: 318
Labour 240 (up 8, inc seats in Scotland) - Actual: 262
Lib Dem 36 (up 28, inc seats in Scotland) - Actual: 12
UKIP 0 (as is) - Actual: 0
SNP 50 (down 6) - Actual: 35
Greens 2 (up 1, gaining a Bristol seat) - Actual: 1
So Tories and Labour both did better than my instinct, Lib Dem and SNP worse. It was Labour who I most underestimated two years ago. How will it change this time? Well, here’s my knee-jerk, top-of-the-head prediction for December 2019:
Conservative 298 (down 20)
Labour 232 (up 10)
Lib Dem 50 (up 38)
SNP 50 (up 15)
Brexit Party - 0 (but splits Tory votes)
UKIP 0 (stands in hardly any seats)
Greens 1
Plaid 2 (down 1)
DUP 9 (down 1)
Sinn Fein 7
This will, as always, be wildly inaccurate (in fact it doesn’t add up properly, which is a bit of a cop out). If it happened, it could put Labour in charge of a coalition government with the SNP or Lib Dems, or the Tories in charge of a coalition with DUP and Lib Dems.
I look forward to being proved wrong in all regards, and whatever happens you can confidently predict it’ll be totally unpredictable.
UPDATE: 13th Dec - ha ha ha, this must be the most incorrect prediction I've made since 1983. I'm currently looking at the Tories on 363, up 66, heading for their biggest majority since 1987. I know nothing.