Politics: manic

By Richard North - June 20, 2024

I have spoken to several people over the last few days who tell me they have given up following the election coverage in the media.

They do right, and if anyone had any doubts about whether this was the correct thing to do, yesterday’s crop of headlines should convince all but the most hardened of media-watchers to declare a vow of abstinence.

I have already written of the media frenzy which seems increasingly to have lost any grip with reality but the descent into madness has now become manic.

Exhibit one, which will be dragged out by historians in the years to come to illustrate the collapse of objective reporting is the front page (online) of the Telegraph graced by two words: “Tory wipeout”.

This is linked to a report headed: “Sunak to lose seat in Tory wipeout, major poll predicts”, which heralds a sub-heading which further informs us: “Exclusive survey for The Telegraph also shows party on track to slump to 53 seats, with around three-quarters of Cabinet voted out”.

Once again, we are dealing with that exotic beast, the MRP analysis, this one produced by Savanta which estimates that Labour could win 516 seats and the Conservatives 53, with a Labour majority of 382.

The projection, we are told, suggests that the Conservatives are in for a close fight with the Liberal Democrats (50 seats) to become the Official Opposition in the next parliament, but it has disappointing news for Reform UK and the Greens, neither of which pick up any seats.

Something that doesn’t come over from the Telegraph headline, though – even if it is buried in the text – is the Savanta observation that 175 seats are “too close to call”, including a raft of senior Conservative seats such as Sunak’s Richmond and Northallerton seat and James Cleverly’s Braintree.

In other words, the methodology is insufficiently sensitive to enable a result to be predicted within the margin of error, meaning that you might just as well stick a pin in one of the parties, for each of these constituencies, with about as much chance of getting it right.

Nevertheless, this does not stop Savanta going for the headline buzz, claiming that “many potential Conservative leadership contenders could lose their seats, including Penny Mordaunt and Grant Shapps. But the word “could”, here, is doing some extremely heavy lifting.

Expressed in percentage vote share, we are told that the “implied” vote shares from this model are Labour 44 percent, Conservative 23, Reform UK 13, Lib-Dems 12 and Greens 4. As with all the other MRP polls to date, the “others” don’t get a look in, being awarded less than one percent.

Only from the small print do we learn that, with so many seats so close, and most of them with the Conservatives in 2nd place, this means that if Sunak is able to have a positive final couple of weeks of the campaign and regain even just a few percentage points, then the model would likely project the Conservatives reaching a three-figure seat tally.

Even then, this overstated the utility of the projection. It is based on an online poll of 17,812 respondents, conducted between 7 and 18 June, which means that individual constituency results rely on an average of 27.4 respondents.

Yesterday, I picked up on the Ipsos caveats, which aren’t repeated to anything like the same degree by Savanta.

But the Ipsos warning stands, that with relatively few respondents per constituency, the projection is unlikely to be able to capture the full local context, especially where there are unique political dynamics. Taking into account unique local dynamics is also difficult.

If I have been able to convey nothing else over the past few weeks, it is that – in this election – there are an unusually high number of seats which have “unique political dynamics”, the complexity of which render their outcomes unreadable.

Incidentally, Savanta appears to contradict itself, saying that Reform doesn’t pick up any seats, then saying that it could take nine, based on probability estimates. The Telegraph, however, settles on zero seats, suggesting that this will mean an eighth defeat in a row for Farage.

Since The Times is also playing the MRP game, there is better news for Farage in its report, where the paper relies on YouGov to award Reform five seats in the Commons.

The main thrust of the report, though, is “Labour on course for biggest landslide in 100 years”, with only the sub-heading telling us that Reform is further eroding Conservative.

Here, the headline figures are 425 seats for Labour, 108 for the Tories, 67 for the Lib-Dems, plus the five for Reform and only two for the Greens. The “others” are not even categorised. Vote share is put at 39 percent for Labour, 22 percent for the Conservatives, 15 percent for Reform, 12 for the Lib-Dems and 7 for the Greens.

Significantly, since YouGov previously – and with much fanfare, put Reform one point ahead of the Tories, now has the “insurgent” party back in third place, seven points behind the Tories.

Looking at the two MRP analyses side-by-side, one wonders if the polling companies were looking at the same electorate, when it comes to the Labour result. Savanta gives Labour 44 percent, YouGov 39. Other figures are closer: the Tories respectively are awarded 22 and 23 and Reform 13 and 12. Only the Lib-Dems, on 12, get the same treatment from both pollsters.

But the comparisons between seats is even starker. Savanta gives Labour 516, YouGov 425. Tories get 53 from Savanta, 108 from YouGov, the Lib-Dems 50 and 67 respectively. The top-line variation for Reform is between zero and five. These are not slight differences. There is nearly a hundred difference in Labour seats between the two analyses.

Yet, for all that, for madness to verge on insanity, one must go to social media, where Matt Goodwin and his “People Polling” company lurks. Via Election Maps he has come up with Westminster voting intentions which give Labour 35 percent, Reform 24 percent, the Tories a mere 15 percent and the Lib-Dems 12.

To give Election Maps its due, it calls this result “an outlier of epic proportions”, expressing complete disbelief. But it does translate the results into seats, estimating 389 for Labour, 109 for Reform, 72 for the Lib-Dems and a pitiful 28 for the Conservatives. That George Galloway seems to take the results at face value perhaps tells you something of the man.

Needless to say, Galloway is not alone, with multiple Reform supporters all over the Goodwin poll like a bad smell, and the party itself parading the result, even claiming that the party might top Labour in the polls.

We have, of course, been there before, with predictions that Farage would lead a band of 20 MPs and the Conservatives would be reduced in number to 26. But memories are short and temperatures febrile.

Still, there seems to be no end of speculation that can be generated by what is probably a spurious poll, with the Telegraph already discussing the shape of the Tory leadership contest should Sunak be ousted.

Perhaps a far more relevant statistic though – of far more importance than anything generated by an opinion poll – came yesterday with the news that 882 illegal immigrant crossed the channel into England in 15 Dinghies. To date, 12,313 illegals have now crossed the channel this year in 249 dinghies – up nearly 20 percent on last year’s numbers.

These are the figures which might end up driving the election.