Average of 8 latest polls, taken 14-22 Nov (incl today's BMG, Opinium, Panelbase, YouGov):
Lib Dem 15%
Brexit Party 4%
Projected Con majority 94
(All polls take account of Brexit Party standing down; Scotland projected from Panelbase & YouGov, 9-25 Oct)
— John Rentoul (@JohnRentoul) November 23, 2019
But the range is still huge, from Con majority of 0 on a 10-point lead (Panelbase) to one of 210 on a 19-point lead (Opinium)
— John Rentoul (@JohnRentoul) November 23, 2019
What is the scale of the Conservatives’ task? They have kept a clear lead over Labour and the Lib Dems seem to be flatlining at best. Tory spirits are buoyant. Pundits are turning to consider not just whether the Conservatives will have a majority but how big that majority will be.
Is this warranted? Here is a list of the top 150 Conservative targets organised by swing required to take them, with the best odds on the Conservatives taking those seats. There’s a general assumption that the Conservatives will lose 20 or so seats to the Lib Dems and the SNP, so they need to get quite a way down this list. (That assumption might very well be wrong – indeed it’s my hunch that it is – but let’s go with conventional wisdom for now.)
The polling at the top of the thread looks comfortable for the Conservatives. The average polling represents a swing from Labour to the Tories of 5.5% (achieved entirely by a move of voters away from Labour). If the Conservatives have indeed got a swing from Labour of that size, they would take every Labour seat up to Bridgend on a uniform swing. That’s 51 seats – not all that far down the list really. If they lost 20 seats to the Lib Dems and the SNP, this would represent a net gain of 31 seats on the 2017 result – 348 in total. That would give the Conservatives an overall majority of 46.
Let’s have a look at this same 150 seats organised by the Conservative best price odds. Immediately we can see that Labour seats have risen to the top of the table. Noticeably, however, bettors are far from gung ho about the Conservatives’ chances. Only 47 seats on the table are rated 5/6 or better (the bookies’ evens). In ten of those, the Conservatives are rated 5/6. If they win the 50/50 balls and lose 20 seats elsewhere to the Lib Dems and the SNP, the Conservatives would get 344 seats. If they lose the coin tosses, they would get 334 – an overall majority of just 18.
Moreover, bettors are notably cautious even about how many safer bets are out there. In a further 18 seats, the Conservatives are odds on but best priced at a longer price than 1/2. If they lost all of those, they would be back to the 316 seats: one less than they got in 2017. When you look at the constituency betting, bettors are evidently scarred by the experience last time.
The caution extends both ways. The Conservatives are best priced below 2/1 in a further 30 seats in the top 100 targets. If they won all those, they would have an overall majority of 98 (again, assuming the loss of 20 seats elsewhere – though that assumption looks shakier the more seats that the Conservatives win).
Those short prices are not justified by current polling. The Conservatives would need to pull ahead of Labour considerably from where they are now or we would need to see a breakdown of past tactical voting against the Conservatives to see the Conservatives make those kinds of gains.
I’ve also colour-coded the far left hand column so that you can see how enthusiastic bettors are about specific seats. The bluer the seat, the more punters expect the Conservatives to outperform in that seat relative to uniform national swing. The browner the seat, the longer the odds on the Conservatives taking the seat relative to their starting position.
You can immediately see that bettors have strong views about the Conservatives’ relative chances in specific seats, and not just by reference to the party currently holding it. Few seats are in the lighter shades indicating that they are close to their uniform swing. Many seats are in the deepest shades, indicating that they are 50 or more places away from where you would expect to find them. Scottish seats, Lib Dem seats and three way marginals predominate among the seats out of favour. That makes sense, although in the first two cases I wonder whether this has been a bit overdone.
On the other side of the fence, very Leavey seats (with the exception of Don Valley, where for no reason I can really follow people seem to think that Caroline Flint’s Brexit position will help her) have leapt up the charts. Hemsworth, West Bromwich West, Coventry North West and Newport East have amazingly short prices, apparently on the belief that their voters are clamouring to get Brexit done.
You would need sensational local information to justify some of the prices on the Conservatives in such seats. I’m a simple soul and it seems to me that the evidence for them is singularly lacking. Maybe they will come home, but I’d want better prices to join in the party. I’d rather be betting on Keighley, Canterbury, Peterborough and Don Valley. In all of those seats the Conservatives are clear challengers to Labour and there is no great reason to expect the national trend to be bucked in them. With the exception of Don Valley, the odds don’t really send a shiver of excitement up the spine.
The price that stands out to me on this table is in Westmorland & Lonsdale. 7/2 to overturn a majority of 777 on a programme of getting Brexit done? Surely that has a chance of resonating in such a large rural constituency. It looks shorter than a 7/2 shot anyway. I’m on.