For a polling junkie like myself, the news in 2010 that YouGov would be polling daily for The Sun was excellent news. Gone were the times of waiting days and days for the latest poll- an instant snapshot after every single political event was now available.
Of course, the fieldwork is always a few days behind, meaning that it’s not really instant at all. And while it’s fun for some basic analysis, do the daily polls really add much value to the overall polling picture? In my opinion, they can be vital.
This idea first struck me a few days ago, when Labour’s lead hit 15% on February 4th. If a general election were called tomorrow, I would be utterly stunned if Labour got anywhere near a 15% lead. I would be surprised if there was any Labour supporter anywhere who believed that this was a true reflection of the current picture.
The leads since then have been much more consistent, and arguably more realistic: 10, 11, 8, 9, 11, 11 and 10. The lead seems to have settled into these kind of figures, and 15% is seen as the outlier. It is interesting to note then, that if these polls did not come at such a quick rate, a very different scenario could occur.
Imagine that 15% lead came, then no polls for a few days, and then the 8% poll is released. Suddenly Labour have dropped 7 points from their lead. Pundits would be scrambling to find reasons why Labour have been doing badly, or what the Conservatives had been doing well, when in reality, hardly anything has changed.
Therefore, the value of the daily YouGovs is clear- follow the trend, not any one specific poll. A 15% amongst a raft of 8-10% leads is probably too good to be true, much as it would be for the Conservatives if they were at level pegging in tomorrow’s poll. With such a constant stream of data, outliers are always going to be likely, but the trends will always prevail.
As this blog continues, I will be trying to examine as many of the regular pollsters as possible- it is a shame that there has not been much more (public) polling on Eastleigh, and I am surprised that more of them do not see it as almost a competition to see who can be the most accurate to the final result. Perhaps we will see some more as the campaign reaches its final week.