Do the pollsters know what they’re doing? Do respondents know what they’re being asked?

The Slog investigates why the Greek Referendum opinion polls are all over the place – and concludes that the predictions are at best misleading.

Last Sunday, Alco polling in Athens showed 57% of respondents ‘wanted a deal’ with the Eurogroupe, 29% didn’t. This was not “do you want to accept the terms of the last offer?’ which is what will be on the ballot paper. On a similar basis the next day, only 47% wanted a deal, 33% didn’t. But EFSYN 24 hours later showed a diametrically opposite result.

Fieldwork was then undertaken by ProRata research on Monday and Tuesday. On Wednesday, it reported that only 33% were in favour of accepting the deal, and 54% planned to vote against acceptance. (By this time, the bank closures were known about).

The same day, I posted about another poll showing a lead for No to the deal of between 7-12 points across all social groups except ND and PASOK supporters. (https://hat4uk.wordpress.com/2015/07/02/giving-a-false-impression-jeremy-hunt-the-eurogroupe-nigel-farage-and-dermot-oleary-which-is-the-odd-one-out/)

So Alco was entirely alone in its view of public opinion on the issue, while the trend was clearly moving Syriza’s way. But Reuters chose to ignore all non-Alco polls until the ProRata report. Their chief writer on Greece, Hugo Dixon, is solidly pro-Eurogroupe and has offered a consistently negative view of both Syriza support and the Greek economy. Reuters published the ProRata survey results, but said “the No lead is clearly narrowing”.

But only three days earlier, the site showed the YES campaign clearly ahead. Now the position had been reversed, but the NO lead was narrowing. So apparently, those keen to continue the baseball sessions with a YES had gone NO once the bank closure had happened, and then back to YES when they opened again. Unless the Greek electorate is masochistically deranged, that’s about as counter-intuitive as it gets;

Again completely against the trend, this morning Alco has published a repeat survey showing 44.8% Yes, 43.4% NO. So given the margin of error (around 3%) that’s Too Close to Call. Reuters once more chose to lead on the Alco study, calling it “a narrow lead” for the YES side – even though statistically it is no such thing.

As my own feedback from Athens, Crete, the Mani, Pyrgos and Kalamata reflects none of this seeming instability of opinion, I decided to look a little further into Alco’s survey track-record.

From 11th to 13th January prior to the General Election that brought Syriza to power, Alco gave the Party 32.4% of the vote compared to 28.9% for New Democracy. Eight days later on January 22nd, Alco gave Syriza 31.2% versus 27.1% for ND.

In the election itself, Tsipras’s Party got 36.3% of the vote, New Democracy 27.8%. What looked too close to call on margins of error (and let’s not forget how wrong, on similar figures, UK polls got the May 2015 election) turned into Syriza winning comfortably by 8.5%.

I’m not the only one who thinks the Referendum data rather odd. The Telegraph’s politics blog yesterday found it inexplicable….but did nail one very important point, with which I agree entirely: the Western press, the eurogroupe and almost all business websites are either being sloppy or pernicious in muddying the clear waters that lie between NO TO THE DEAL ON OFFER and NO TO THE EUROZONE. If there is one thing all the polls agree upon, it is that circa 70-75% of Greeks don’t want to be thrown out of the eurozone.

My own strong (and unpopular) view is that the idea of Greece remaining an equal partner in the ezone now and defying the eurogroupe is a fantasy based on the cake and eat it with chocolate sauce and ice cream on top principle. Whether one calls this Greek opinion naive or cheeky depends on one’s perspective on The Marathon. The point is, as an expectation it exists. And the usual media suspects are exploiting it.

Old friend Tony Taylor was spotted by The Slog in Syntagma Square this week showing his solidarity with the Yes vote. Tony’s a long-time europhile with broad experience of handling EU-based automotive clients, and good for him – he believes in it with a passion. But flicking Greek channels earlier this week, I was struck by the imbalance of coverage in favour of YES versus NO demos. An audit published yesterday confirms that gut feel: the Greek TV bias was a staggering 8:1 in favour of YES.

This is a steeply tilted playing field that Syriza supporters, sympathisers with Greece and indeed all those fearful of the EU’s anti-democratic tendencies must learn to accept. For me, the most dangerous thing of all is to ignore it in a rose-tinted episode of fluffy optimism. This is where the British Labour Party falls down (and then apart) over and over again, because it doesn’t get out enough.

I would recommend that any non-Greek visit Kalamata in all its intellectual and achitectural glory, and take in the self-interested, ruthless and ineluctably materialist views of those with fat heads and big bottoms who tend to predominate in this, the capital of Antonikis Samaras admiration. Everyone in Europe needs to grasp that, if its political elasticity is tested too crudely, then Hellas is going to dissolve into civil war.

In the meantime, I fancy a pillar of salt is required in relation to Sunday’s YESNO interlude. If the Greeks vote for more baseball bat pummellings, then that’s their funeral….and believe me, it will be. My gut empiricism says they won’t. But either way, one electoral negotiating ploy will not make a jot of difference to the eventual outcome: economically, socially and financially, the euro is a toxic enemy of democracy. It is doomed.