Pro-independence campaign make spectacular breakthrough in new poll from traditionally No-friendly firm TNS-BMRB

TNS-BMRB have on average been the third most No-friendly out of the six BPC-affiliated firms.  Their last poll showed Yes on a new high watermark for the campaign - but even that was only 41.4% with Don't Knows excluded, as compared to the highs reported by ICM, Panelbase and Survation of between 47.1% and 48.3%.  It was also only fractionally higher than the figure that Yes had been hovering around in the firm's polls for several months.  However, just when we were beginning to wonder if Yes were ever going to make a telling breakthrough with TNS, it's finally happened tonight - and in quite some style, with the No lead slumping by 6% after Don't Knows are excluded.

Should Scotland be an independent country?

Yes 44% (+3)
No 56% (-3)

With Don't Knows left in, these are the figures -

Yes 32% (n/c)
No 41% (-5)

On the latter measure, Yes remain on the new high watermark for the campaign that they reached in the previous TNS poll - before that poll, they'd never been higher than 30%, or at least not since TNS introduced a major methodological change.  By complete contrast, the No campaign's 4% increase in the last poll (which was partly an illusion caused by rounding) has been more than wiped out, and they're now back down to a vote share that equals their all-time low.  Although on face value this looks like a straight swing from No to Don't Know, it's more likely that the last poll was simply an outlier in showing a big decrease in the undecideds.  That would mean there has been genuine movement from No to Yes at some point over the last few weeks.

As you'll doubtless have spotted already, the headline No lead is just 9% - the first time it has hit single figures with TNS in the campaign so far.  That's a 5% drop on the last TNS poll, and is 3% lower than the previous record low shown by the firm.

I've pointed out before that swings to Yes reported by Ipsos-Mori and TNS are considerably more important than swings reported by other firms, for the simple reason that they're the only two active referendum pollsters that actually seek out a fresh 'real world' sample in every poll, rather than rely on volunteer online panels.  Until now, both have tended to be firmly on the No-friendly end of the spectrum.  That isn't to say they're necessarily more accurate than others, because there are plenty of advantages to the online approach as well - it's more anonymous, for starters, and respondents are more likely to give honest answers.  But it's nevertheless highly encouraging to see an old-fashioned face-to-face pollster like TNS produce the kind of numbers that have previously only been reported by online firms.  Indeed, as things stand TNS are showing a slightly higher Yes vote than ICM (after Don't Knows are excluded), which is quite a turn-up for the books.

On the other hand, the perennial problem with TNS is that there's always quite a long gap between fieldwork and publication, meaning that their figures are slightly out-of-date by the time that we see them.  In this case, that will complicate any attempts to use this new poll to resolve the mystery of the contradictory trends we've been seeing recently from different firms.  The TNS fieldwork took place between the 25th of June and the 9th of July, meaning that it had the same start-date as the most recent YouGov poll, but a much later end-date.  So it certainly adds to the weight of evidence that the slight drift towards No reported by YouGov recently has probably just been margin of error "noise", and it leaves open the possibility that there has been movement to Yes since YouGov's fieldwork concluded.  The fieldwork for last week's Survation poll overlapped with the tail-end of the TNS dates, which is useful, because both firms are in agreement that Yes have reached a new high.  However, the most up-to-date poll remains the ICM poll published at the weekend, which finished a couple of days after TNS had stopped interviewing.  Confusingly, that poll showed a small increase in the No lead - but the changes were consistent with margin of error noise, and the headline figures remained well within ICM's normal range.

So although the TNS findings dramatically increase the chances that there has been real and substantial movement to Yes recently, they still don't constitute absolute proof.  As ever, we'll just have to await the next poll for more information.  One thing we can certainly say, though, is that there's no credible evidence at all of a swing against Yes - shouldn't that be happening by now, if the predictions made last year by certain "experts" (and Ian Dunt) have any validity?

*  *  *

UPDATE : Even better news - the STV website is reporting that the TNS figures for respondents who say they are certain to vote are as follows...

Yes 37% (+2)
No 46% (-3)

No word yet on the figures with Don't Knows excluded, but a rough calculation suggests that they're most likely to be Yes 45% (+4), No 55% (-4).  That would certainly explain why Yes Scotland tweeted that this poll showed Yes at 45%, rather than 44%!

*  *  *

REQUIRED SWINGS

Although I'm going to persevere with the Poll of Polls, I'm coming round to the idea that it may not be the best way of summarising the state of play, given the huge ongoing disparity between the figures produced by different firms - it's just providing an "imaginary middle".  So I'm going to start giving the swings that would be required for a certain number of pollsters to either show Yes in the lead, or a dead-heat.  For simplicity, these are based on the rounded numbers used for publication, and with Don't Knows not excluded.  (It wouldn't be possible to use unrounded numbers across the board anyway, because the notoriously secretive firm YouGov never reveal their unrounded results.)

Swing required for 1 out of 6 pollsters to show Yes in the lead or level : 1.5%

Swing required for 2 out of 6 pollsters to show Yes in the lead or level : 2.5%

Swing required for 3 out of 6 pollsters to show Yes in the lead or level : 4.5%

Swing required for 4 out of 6 pollsters to show Yes in the lead or level : 5.5%

*  *  *

SCOT GOES POP POLL OF POLLS

This update of the Poll of Polls essentially reverses the small boost for No that was seen last time round - the No lead with Don't Knows taken into account is now back below 11%.

MEAN AVERAGE (excluding Don't Knows) :

Yes 43.6% (+0.5)
No 56.4% (-0.5)

MEAN AVERAGE (not excluding Don't Knows) :

Yes 36.8% (n/c)
No 47.7% (-0.8)

MEDIAN AVERAGE (excluding Don't Knows) :

Yes 43.4% (+1.3)
No 56.6% (-1.3)

(The Poll of Polls is based on a rolling average of the most recent poll from each of the pollsters that have been active in the referendum campaign since September 2013, and that adhere to British Polling Council rules. At present, there are six - YouGov, TNS-BMRB, Survation, Panelbase, Ipsos-Mori and ICM. Whenever a new poll is published, it replaces the last poll from the same company in the sample. Changes in the Poll of Polls are generally glacial in nature due to the fact that only a small portion of the sample is updated each time.)

And here are the long-term trend figures, with updates prior to Easter recalculated to remove the inactive pollster Angus Reid ...

The No campaign's lead in the Poll of Polls mean average (not excluding Don't Knows) :

Sep 2013 - 21.6%
Sep 2013 - 21.4%
Sep 2013 - 19.4%
Oct 2013 - 18.8%
Oct 2013 - 18.4%
Oct 2013 - 18.2%
Nov 2013 - 18.4%
Nov 2013 - 18.0%
Dec 2013 - 17.0%
Dec 2013 - 16.8%
Dec 2013 - 16.4%
Jan 2014 - 14.4%
Jan 2014 - 14.2%
Jan 2014 - 14.2%
Jan 2014 - 15.2%
Feb 2014 - 15.0%
Feb 2014 - 15.5%
Feb 2014 - 15.5%
Feb 2014 - 13.7%
Feb 2014 - 13.3%
Feb 2014 - 14.2%
Mar 2014 - 14.2%
Mar 2014 - 14.5%
Mar 2014 - 14.5%
Mar 2014 - 14.7%
Mar 2014 - 13.8%
Mar 2014 - 13.0%
Mar 2014 - 12.5%
Apr 2014 - 12.5%
Apr 2014 - 12.7%
Apr 2014 - 12.7%
Apr 2014 - 12.3%
Apr 2014 - 11.4%
May 2014 - 11.2%
May 2014 - 11.2%
May 2014 - 11.5%
May 2014 - 13.3%
Jun 2014 - 12.1%
Jun 2014 - 12.1%
Jun 2014 - 11.3%
Jun 2014 - 9.9%
Jun 2014 - 10.3%
Jun 2014 - 10.7%
Jul 2014 - 11.0%
Jul 2014 - 11.0%
Jul 2014 - 11.7%
Jul 2014 - 10.9%