Just for the sake of completeness, here are the three most recent Scottish subsamples from Britain-wide polls -
ICM: SNP 34%, Conservatives 32%, Labour 21%, Greens 4%, UKIP 4%, Liberal Democrats 4%
YouGov: SNP 37%, Labour 29%, Conservatives 23%, Liberal Democrats 7%, UKIP 4%
Opinium: SNP 35%, Conservatives 32%, Labour 24%, Greens 6%, UKIP 1%, Liberal Democrats 1%
In a sense these are in line with the full-scale Scottish poll from Panelbase, because they all show the SNP in the lead, and they all show the SNP well ahead of Labour, who until recently had looked like the main challenger. There have now been twenty-three subsamples since the election, and fourteen have put the SNP ahead. Seven have shown a Labour lead, and only two have shown a Tory lead.
If we buy into the theory that there was a Labour surge during the summer which has since subsided, there's one huge mystery that has yet to be solved. How do we explain the significant swing from SNP to Labour in the very recent Cardonald and Fortissat by-elections, which took place at roughly the same time as the Panelbase poll was in the field? Perhaps there were local factors at play, and perhaps it's just coincidence that more or less the same thing happened in two different places at once...but we should probably keep an open mind until we have more information.
Note : I'm out of the country for a couple of weeks with intermittent internet access, so blogging may be light.