Make no mistake the polls point to the IndyRef being on a knife-edge and so much depends on turnout
More than 80% of postal votes have now been returned
The big unknown from this election is turnout – something that all the pollsters seek to measure and apply when working out their final vote shares.
Thus the 2% NO lead ICM phone poll used the firm’s standard turnout adjuster of attaching a 50% discount to those who didn’t vote in previous parliamentary elections.
But in the context Thursday’s totally unprecedented election we don’t know whether that’s a valid approach or not.
The other major phone pollster, Ipsos-MORI, only includes those saying they are 100% certain in headline figures. Again how valid will that be given the immense interest and enormity of Thursday’s vote.
We do know that more people, an estimated 97%, are on the Scottish electoral registers and that there has been a huge effort to get this up to that level in advance of the referendum. But how many of the new names will cast votes on Thursday and are the more marginal ones more likely to be YES supporters?
The current postal vote turnout is in excess of 80% pointing to and overall postal vote turnout of 85%+. The assumption is that NO will have the edge amongst these voters largely because they are older. Will that hold?
So many imponderables and so much of this is emotional which is hard to measure.
I am maintaining my split betting position so I win the same whatever