Labour's polling recovery is not enough to save it

I've started to notice a really irritating trend on my Facebook recently. People trying to point out that during the election campaign, Labour's share in the opinion polls has risen.


There a couple of problems with this fact. Firstly, opinion pollsters in this country have traditionally over-rated Labour. In the last thirty years, for example, the margin of error between Labour's eve of poll rating and their actual vote share has been on average 3.2% points:

1987- Final polling day average of 33.5%. They won 30.8%, so a gap of 2.7% points.
1992- Final day polling average 39.3%. They won 34.4%, so a gap of 4.9% points.
1997- Final day polling average of 45.3%. They won 43.2%, so a gap of 2.1% points.
2001- Final day poll put Labour on 47%. They won 40.7%, so a gap of 6.3% points.
2005- Final day poll put Labour on 38%. They took 35.3%, so a gap of 2.7% points.
2010- Final day polling average of 27.6%. They took 29%, so an underestimate of 1.4% points.
2015- Final day polling average of 33.0%. They took 30.4%, so a gap of 2.6% points.

Only once have the opinion pollsters underestimated Labour's vote share. Every other time, it has been overstated. This has led to some embarrassing mis-predictions. In 1987, the BBC forecast from their polling that Mrs Thatcher's majority would be cut from 145 to 20; she got a majority of 101. In 2015, Ed Miliband was as stunned as the rest of us at his defeat. And most famously of all, in 1992, all the opinion polls missed a late surge in Tory support, at Labour's expense. Even in the exit polls, voters wouldn't admit to voting for John Major, and so the predicted hung parliament never materialised, and Labour went down to a record fourth election defeat.

The problem here is that, even with a current polling average of 32%, Labour are likely to do slightly worse than that on the day itself. They could do significantly worse, especially if Theresa May and the Brexit deal are activating a case of 'Shy Toryism.' Nothing the opinion pollsters have done with their methodology has managed to correct this flaw, yet.

The second problem with Labour's rise in the polls is that they are not alone. Since the campaign started, another party has enjoyed a polling bounce. The Conservatives. The collapse of UKIP support, from the low teens down to below 5%, has seen those voters largely migrate to the Conservatives. The Tories are looking at taking nearly half the vote. I'll be amazed if they don't take above 45%, a feat last achieved by Ted Heath in 1970 (Incidentally, the last Conservative leader to stand with a clear commitment to changing Britain's relationship with Europe, although Heath is doubtless turning in his grave at what is going on).

It is highly likely that this government will get the best electoral mandate of any UK political party since the 1960s or 1950s. When he added 0.8% points onto the Tory vote share in 2015, Cameron was lauded as an electoral wizard. And he had presided over five years of dull but effective government (at least for his supporters). God knows what that makes Theresa May, who has achieved this surge in support whilst presiding over the upending of half a century of foreign policy, amidst an extremely divisive and inflammatory atmosphere.

Even if both party vote shares are rising, the Tories are still miles ahead of Labour. Whereas a share of 32% for Labour would be pretty decent, and enable them to hold a Conservative party on 35%-37% to a virtual tie, a rise as high as 35% would be utterly meaningless when the Conservatives have gobbled up nearly half the electorate. Some Labour candidates may gain ground and lose their seats simultaneously, even sitting MPs. But Labour's boats will not be raised by the tide when there's a tidal wave sweeping through the harbour at the same time.

And I'm not even going to engage with the idea that if Corbyn matches Miliband's vote share he should stay. That would tear Labour apart. But that's for another day.

So, sorry to disappoint, people who seem to think that Labour getting to 32% in the opinion polls is something to get excited about. It should be, and I'm glad they're doing a bit better. But there is an avalanche coming down the mountain towards us. Don't kid yourself your snowman will reduce you from it.

Edward Heath, entering Downing Street as Prime Minister, June 1970. Theresa May, who was 13 at the time, may be about to be the first PM since then to be elected with more than 45% of the vote.