Betting on a UKIP by-election upset in the absence of hard polling
More from Aby Tomlinson – the energetic LAB campaigner who tweeted about the responses she was getting in Oldham pic.twitter.com/idnhv91aa9
— Mike Smithson (@MSmithsonPB) November 30, 2015
Alastair Meeks gives his view
Oldham West & Royton should have been a spectacularly boring by-election. It is a previously-safe Labour seat held at the last general election by a leftwing MP with a thumping majority and an absolute majority of votes cast. To gain it, UKIP would require a swing of 17.1%. Such a swing would be remarkable against a party of government, never mind against a party of opposition. In normal circumstances, you would expect a very comfortable Labour hold.
These are not normal circumstances. Labour seem to be testing to destruction the concept of a core vote by stage managing a procession of calamities and internal feuding to climax just in time for the by-election vote. As a result, a string of nervous accounts of canvassing have been percolating out from Labour activists via twitter and newspaper articles. Might Labour actually manage the previously unimaginable and lose this seat?
We do not have any published opinion polls to clutter our thinking (which is perhaps just as well given how they led us astray in the general election) so we are thrown back on first principles.
By chance, we have four recent by-elections in the vicinity from the last Parliament as comparison points: Oldham East & Saddleworth, Manchester Central, Heywood & Middleton and Wythenshawe & Sale East. The first three of these constituencies are adjacent to Oldham West & Royton – there must be something in the water round there. In all four, Labour actually increased its share of the vote at the by-election. Turnout dropped in all four: least in Oldham East & Saddleworth (which had been a marginal), where turnout was at 78% of its general election level, and most in Wythenshawe & Sale East, where turnout was 51.9% of its general election level.
Clearly if Labour increases its share of the vote in this by-election it will hold the seat. No one seems to think that it will do anything like that well. It has been suggested that at least half of its general election voters will not turn out for it on 2 December. However, while worse than in any of the four by-elections referred to above, that wouldn’t be all that bad. If Labour retain only 50% of their votes from the general election and UKIP turn out two thirds of their vote from the general election, UKIP would still need to find almost as many votes again to overtake Labour.
So UKIP would need to secure direct defections from Labour, improve the turnout of its own general election support and harness tactical Conservative votes in its cause. It is important to grasp that UKIP need to make substantial numbers of new converts come what may – if they turn out every last one of their general election supporters and get no new supporters, they are left hoping that 37% or fewer of Labour supporters in May turn out. Even for a dark wet day in December against a party led by a man who has not yet gained the demeanour of a vote winner, hoping for such a spectacular vote strike looks like a losing strategy to me.
UKIP have no doubt been particularly focussing on getting direct defections from Labour simply because there are more Labour voters to go at in this constituency. If Labour turnout is 50% of their May vote and UKIP turn out two thirds of their vote, UKIP will snatch the seat if they can persuade a third of the former Labour supporters to back their cause and twist the arms of a quarter of the Conservative support to lend them their votes.
Is that doable? Maybe, but make no mistake, it would be a landmark success for UKIP if it were done. Conservative voters don’t have much of a track record of tactical voting but equally Jeremy Corbyn’s polling levels with Conservatives are dismal even for a Labour leader, so who knows? Perhaps fewer than 50% of Labour’s May voters will turn out. If that figure drops towards the low 40s, Labour look in deep trouble.
Enough of the numbers, what is actually going on? No one really seems too sure, which is perhaps not that surprising in a constituency that will have not been subject to extensive canvassing in the past. Labour will be reliant on their experience in running a postal voting operation while UKIP, unusually, have the benefit of the experience of a seasoned local campaigner in Joe Fitzpatrick (who when still with Labour masterminded Phil Woolas’ subsequently-overturned victory in 2010 in Oldham East & Saddleworth).
The consensus of the various field reports seems to be that Labour will hold with a small majority; some have mentioned a majority in the hundreds. That smacks of herding rather than insight to me, given the sketchiness of the information.
Current best prices are Labour 1/3 and UKIP 5/2 (both prices are available with a range of bookies). You can get 11/4 on Betfair as I write. Labour are rightly favourites and might yet win with a decent majority but the chances of a UKIP win on the information that we possess look a bit better than one in three. I’m already on UKIP at 11/4 and longer and I’m not topping up, but if I were starting from scratch I’d be backing UKIP now.
Given that this by-election is not going to help choose the next government and given that Jeremy Corbyn is reportedly not attracting the admiration of the typical Oldham voter, turnout is likely to suffer. Ladbrokes previously offered an over/under line at 44.5%, which seemed very generous. This has now been updated to 37.5% (5/6 over or under, according to your preference). This would place the drop in turnout from the general election very close to the average of the four nearby by-elections in the last Parliament. If like me you think that turnout is going to drop by more than usual, take the under side of this bet.
Whatever happens, this by-election is going to prove educational. If Labour win well, we should take careful note of Labour’s efficient expectations management. Any other result is grim for the red team. We can rest assured that in that case the reasons will be debated at length.