Conversely, substantial Lab/LD margins in Scotland and Wales may be sufficient to put Lab ahead for UK, even if they do come third to LDs in England.

There is more than a subtle difference between speculation and assumption ! While I agree that Stuart Gregory’s Lab 15% / LD 30% is far-fetched to say the least, I do not think it impossible for LDs to overtake Labour in terms of votes – but not seats – at the next election.

Historically, we have typically seen a pattern of movement in polls from start of campaign to polling day which generally shows a change of 2-3% in the Con – Lab differential to benefit of Con. However, if one looks more closely, this is typically not down to a discernible shift direct from Lab to Con, but more to a shift of 2-3% from Lab to LD. On that basis, it is quite plausible that if the Lab / LD margin at the start of the campaign is under 4%, then LDs could emerge ahead of Lab in final tally of votes.

This is most likely to happen if, as we have seen in a number of recent polls, both Lab and LD are in the low 20s. [My expectation is that both Lab and LD will be in the 20-25% range at next GE, and that LDs will be (just) ahead of Lab in England, but not across UK].

Note however that LDs overtaking Lab in % vote is not of itself going to make them the offical opposition since they will still be a considerable way behind Lab in terms of seats. If there is a bias in the system such that Cons need to be 5% ahead of Lab to win more seats, and up to 10% ahead for an outright majority, the same is true in spades for LDs.

That will be especially true at the next election where LDs are likely to rack up several million votes in 150+ English seats where they are second to Cons and where Lab are nowhere in sight.