Weekend Summary

The clocks have gone back but the polls have marched onward. Here’s the master graph!

(I’ve updated the format a little – the line has now been removed, as this misleadingly overemphasised outliers and arbitrarily gave one poll undue prominence when several were on the same day. I’ve replaced it with markers to show the mean number of seats gained for a given party in each poll.)

Looks like further narrowing this week. Zooming in, here’s the last two weeks:

A steadier week this time around after last week’s uncertainty, but that’s not much comfort for Labour, who are now pretty steadily below the winning post. There’s plenty of narrowing, but interestingly Labour hasn’t lost much – the narrowing is due to a Conservative gain, quite possibly linked to the UKIP results settling back down as they drop out of the news cycle somewhat. (I suspect you’ll be able to watch for them jumping back up with Mark Reckless’s increase in coverage as the Rochester and Strood by-election approaches.)

Lastly, here’s the predicted result of running 1000 simulated elections on the latest UKPR average (currently CON 32, LAB 34, LD 8, UKIP 16, GREEN 5).

Clear advantage for Labour, who would take power in basically all runs (outright win in 18% of runs, hung Parliament in the remaining 82%, and Lab-LD was the only feasible coalition in 99.9% of those, “no obvious coalition” making up the other 0.1%). On average they’d be short by about 10. UKIP have dropped back a bit into a clear fourth place, but still a respectable number of seats; everyone else rumbles on much as usual.