Posts Tagged ‘predicting elections’

In the last 24 hours or so polling conducted for The Guardian has shown the “Leave” campaign in the British EU referendum to have jumped into a small lead over the Remain campaign.

This has excited much comment, cheering from the Brexit mob, and even a run on the pound. (Just like the good old days – Ed.)

Before the Little Englanders get too excited, though, they would do well to consider the following:

Betting market

By far the most accurate “poll” on upcoming elections has been shown – again and again – to be the betting markets. This is because they are made up of people who are not answering questions  on the phone or online, but people actually wagering real money – their own money – on their considered analysis of the likely result, assuming the risk for their opinions.

This is not to say that the betting market is always 100% accurate – that would be an over-statement – but it very accurately measures trends. So what is really interesting is always the MOVEMENT in the odds, just as the TRENDS in opinion polls are what are extremely accurate, not any individual poll.

The political betting market in the UK is very large, which also increases its accuracy.

These odds show a small movement towards “Leave”, but nothing like enough to see them winning the vote in three weeks’ time. The general consensus is still in the region of 4-1 (at least) ON a Remain victory, and as wide as 3-1 AGAINST (or higher) on a Leave win.

This small shift towards Leave is also picked up by the FT poll of polls, which shows Remain winning despite the ICM poll. It should also be noted that a cursory glance at the poll of polls show that ICM have consistently rated Leave higher than other polls, suggesting there is something in their methodology that is fractionally favouring the Leave side. Of course if they are proven right they will be cock-a-hoop, and good on them if it proves to be the case, but that rarely happens in such examples.

What really WOULD put the cat among the pigeons would be if the slight drift to Leave was picked up by other polls, or became a growing trend. That is not yet the case.

One thing that is clear is that the Leave campaign is polling far better than Prime Minister David Cameron would be happy with, and it may well be that a sizeable “out” vote will undermine his already eroding position within the Conservative Party.

He was urged by many (including then Lib Dem leader Nick Clegg) not to play around with a referendum to de-fang his Eurosceptic right but loftily chose to ignore that advice. As things stand, it may be that any victory in the election will be close enough that he, personally, is stripped of personal authority and mortally wounded. It may turn out to be a gross miscalculation.

When true leadership was required, he squibbed it. And the penalty for that is always, inexorably, inevitably, political death.

cruz2We have an enduring fascination with trying to pick election results before they happen.

And we are nearly always spot on, or at the very least, very close.

So here we go.

Turnout will be very high, and will be the first news story of the night.

A high turnout might suggest a sudden last minute surge for Sanders in the Democrat primary, but we believe the Democrat primary will be won by Hillary Clinton, shading the insurgent Sanders. Thetrump hands more overblown predictions of a Sanders victory will, we think, be shown to be wishful thinking by his supporters and progressives generally, as enough Iowans consider who can actually defeat the Republicans in a general election.

In this, Donald “Bogeyman” Trump and the perennially unpleasant Ted Cruz are actually helping the Democrat establishment to encourage people to coalesce around Clinton.

Hillary-AngryBut Sanders will do well amongst students and young people particularly, and cannot be counted out entirely yet.

Of course last time round she got a nasty shock when she surprisingly got beaten by Barrack Obama.

With Iowans, anything’s possible, but the talk afterwards will be, we think, of how the Democrats folded in behind Hillary when push came to shove. Sanders will, however, fight on, so that the Democratic nomination isn’t just a coronation.

On the Republican side we expect Trump to beat Cruz. That said, Cruz will poll more strongly in the evangelical and country areas where a lot of Iowans live, and where caucusing is a particular social and well-regarded event. So don’t discount a win by Cruz, even though we think Trump will do better. You might see a city/country split between these two, which would be interesting in its implications for the whole Republican race and the ultimate general election.

We expect the big story of the night on the right, though, will be a better than expected showing by Marco Rubio, who we think (have thought for some time) has an excellent chance of being the party’s eventual pick. We wouldn’t be at all surprised to see him poll 15-20%, or even, at a pinch, higher. And interestingly Cruz has turned to criticising Rubio in recent days, perhaps a reflection of a calculation that he is leaking votes to him while Trump’s supporters are staying mainly true to him. But this will be the day that Rubio really “arrives”.

And Trump and Cruz? What would they take out of the night when one of them wins?

Well, just ask Rick Santorum and Mike Huckabee how much significancesanders2 winning the Iowa caucuses really has.

Anyhow, that’s our best guess.

We’ll see if we were right soon enough.

Pay close attention to Bremer County. This particular collection of 24,000 Iowans have chosen the right President in the last 9 Presidential elections.

There are any number of good ways to follow the results as they come in, if you are a fellow election tragic. This is one of the better ones, and not being American it will hopefully be a bit less biased and a bit more informative.

http://www.theguardian.com/us-news/ng-interactive/2016/feb/01/iowa-caucus-results-live-county-by-county-interactive-map