Sometimes we get stuff wrong. Well, quite a lot of the time actually. But speculation - even inaccurate speculation - is part of how we learn. So it is with the UK Labour Party's loss of Copeland in Cumbria, and the swing against the party recorded in Stoke-on-Trent Central. We said that Labour probably wouldn't lose either seat. History was against such an outcome. The data said that serving governments just don't win by-elections against the official Opposition. They certainly don't in the seventh year of an austerity regime that has just spent months tearing itself apart over Brexit. You know what? Theresa May's Conservatives did it. They had a phenomenally good night, not only taking Copeland for the first time since the 1930s (above), but blunting the United Kingdom Independence Party's challenge in Stoke too. There shouldn't be any champagne left within a five-mile radius of Conservative Central Office. We said it couldn't happen. It did. Sorry.
It's time now, therefore, to turn to what those results mean. First, Labour's good news. UKIP probably aren't going to smash their way through Labour's heartlands in the same way that the Scottish National Party have done in Scotland. They had a pretty torrid time of it in Stoke, with their leader battered by a string of allegations that he had been, well... not entirely measured in everything he had said. Their vote went up only a tiny bit, in a seat which had voted heavily for Leave in the referendum on Britain's membership of the European Union. Their vote fell in Copeland. They're not going to be taking many Labour seats any time soon. Their brand is just too toxic. The entire reason for choosing them on the ballot is rapidly diminishing as Mrs May steals all their ideas. Their pivot from seeking Conservative votes, to wooing Labour defectors, just seems too crude. Her Majesty's Opposition is going to be Labour's title for as far as the eye can see.
The really depressing thing for Labour comes when we look at why that is. It's because the Conservatives did so well. UKIP weren't able to persuade Conservatives to vote tactically for them in Stoke, where Labour's vote share was actually lower than in Copeland. The Conservatives outstripped their recent performances in both seats, confirming that they really are riding as high in the polls as it appears. In fact, given that the Copeland swing was bigger than that implied by polls (part of the reason we thought that Labour would probably cling on there), opinion polling might still be rather underrating them.
Now Copeland is a special place. The nuclear industry is critically important to its economy, and Trident is important in the south of the seat. No doubt that hurt a party led by Jeremy Corbyn, with decades of opposition to the nuclear industry and nuclear weapons behind him. But the really scary thing for Labour is not the industrial mix in Copeland. It's the class mix. It's quite a working-class seat, and the party that sees itself as standing up for those people's interests and ideas just lost it - badly. Working class voters have been drifting away from Labour since the early 2000s, of course, but the trend of disillusionment and disengagement now seems to be accelerating. Take a look at the very latest YouGov opinion poll (opens as PDF), and you'll see that the Conservatives lead Labour in the C2DE social class category by ten percentage points (38 per cent to 28 per cent). Labour are just nine points ahead of UKIP with these voters. They're filing out of the Labour tent. Gradually, sometimes imperceptibly, but steadily, lots of traditional Labourites are just switching off. The new model Labour Party always tried to claim that it might be able to win elections by raising turnout. That looks a distant, forlorn prospect right now. Reader, we're prepared to bet that turnout in (by then ex-) Labour areas will be low indeed by 2020.
What did we say, back in January? Well, this: if Labour did lose one, and especially both, of these seats 'they probably face a defeat that cannot even be described as a catastrophe. Such a result would be a sign of impending cataclysm: the opening of a political black hole from which there may be no escape'. Now that might be a bit overcooked, actually, especially given UKIP's complete failure to move forward on any front. But there is now at least the possibility, at one end of the bell curve, of an actual Labour meltdown. Mr Corbyn is in all likelihood going nowhere. He's entrenched as leader, and he can't be removed. He's ludicrously unpopular overall, and he's dragging down his party, though among better-off Britons Labour isn't doing quite as badly as it is amongst those aforementioned C2DEs, and it's at least holding out in London and some other big cities. Eyewitness accounts in Stoke back up the numbers: among young people and students, Corbyn-Labour is pretty popular. Unfortunately, there aren't enough of such voters to prevent a really thumping marmalisation in a General Election, and in the age of Individual Electoral Registration, all-expat voting and a rapidly ageing population, their significance is about to depreciate rather than increase.
This is not the likeliest outcome, by any means, but a 1931-style rout cannot now be ruled out. Flying blind with a comms team that seems able only to speak in a weird kind of pseudo-English fixated on 'bargain-basement tax havens' (whatever that means), and which spends its limited time on this earth threatening parents with their own children's death, is... sub-optimal. As they put up utter neophytes for trial by interview, and go on being 'directed' by a 'Shadow Cabinet' totally unmatched to the task before them, the risk of Labour's complete annihilation seems to grow by the day, if not by the hour. A General Election campaign might be fronted by Mr Corbyn losing his rag with reporters and John McDonnell shouting at imaginary plotters, backed up by amateur dilettantes unable to get even the most basic press release out in a timely fashion. If that does happen, the Conservatives could well end up with a 1930s-style clean sweep. Copeland is a warning: a clear siren of danger in the blue skies and the sharp air of February. All democrats - all citizens who believe in a viable multi-party system, in electoral competition, and in good governance - should listen. Heed the warning, before it's too late.