ICM online poll suggests the EU referendum race is absolutely level

This may help to steady the ship slightly for the Leave campaign, after a couple of recent polls that made some commentators wonder if the referendum was gradually ceasing to be competitive.

Should the United Kingdom remain a member of the European Union or leave the European Union?

Remain 45% (+2)
Leave 45% (-2)

In spite of the tie for the lead, the percentage changes in this poll look perfectly consistent with the modest movement to Remain suggested by the Opinium poll at the weekend.  They're quite hard to interpret, though, because the last ICM poll was unusually good for Leave, so even if the underlying state of play had remained static over the last week, we might have expected some kind of reversion to the mean in the new poll.  Certainly if the wheels were coming off for Leave to quite the extent reported by the ORB series of phone polls, it's likely that ICM would have shown a clear Remain lead today.  Perhaps the ICM online methodology is just proving slow in picking up a pro-Remain trend that will eventually become apparent in all polls, but at least for the time being there are grounds for the Leave camp to feel somewhat reassured.

(UPDATE : It turns out that ICM have made an important methodological change, which entirely explains the small swing to Remain - without it, Leave would still have been 4% ahead.  That's incredible - it suggests that ICM haven't picked up any sort of movement over the last week at all, in spite of the hype over the Remain surge in the Ipsos-Mori and ORB phone polls, and to a lesser extent in the Opinium online poll.)

In the wake of the ORB poll yesterday, there was a lot of talk about the supposedly decisive shifts in opinion among pensioners and Tory voters.  Well, you wouldn't know that from the ICM poll, which gives Leave a more than 2-to-1 lead among over-65s, a slender lead among people who would currently vote Tory, and a healthy 9% lead among people who actually voted Tory in the general election last year.  The reason for the conflicting results is of course the well-established divergence between online and phone polls, which if anything may be growing wider.  The trouble for the Leave campaign, though, is that it's beginning to look like they may need online polls to be completely right and phone polls to be completely wrong - somewhere in the middle probably isn't going to be enough.  And even if online polls are bang-on accurate, Remain would still have at least a 50% chance of winning.

*  *  *

SCOT GOES POP POLL OF POLLS

50/50 ONLINE/TELEPHONE AVERAGE :

Remain 46.5% (+1.0)
Leave 40.9% (-0.1)

ONLINE AVERAGE :

Remain 42.4% (+0.5)
Leave 42.7% (n/c)

TELEPHONE AVERAGE :

Remain 50.5% (+1.5)
Leave 39.0% (-0.2)

(The Poll of Polls takes account of all polls that were conducted at least partly within the last three weeks. The online average is based on nine polls - four from ICM, three from YouGov, one from TNS and one from Opinium. The telephone average is based on six polls - two from ORB, one from ICM, one from YouGov, one from Ipsos-Mori and one from ComRes.)