York Outer – 0.44% (Lab 27%)
Romsey & Southampton N: 0.46% (Lab 10.9%)
Cheltenham: 0.66% (Lab 11.8%)
Eastleigh: 1.12% (Lab 21.1%)
Westmorland & Lonsdale: 1.7% (Lab 7.8%)
Hereford & N Herefordshire: 2.4% (Lab 10.2%)
Carshalton & Wallington: 2.93% (Lab 17.3%)
Taunton Deane: 3.3% (Lab 12.1%)
Chippenham: 4.7% (Lab 16.8%)
Leeds NW: 5% (Lab 31.9%)
Torbay: 6% (Lab 14.5%)
Sutton & Cheam: 6.2% (Lab 11.8%)
Camborne & Redruth: 7.1% (Lab 25.6%)
Richmond Park: 7.1% (Lab 9.25%%)
Cheadle: 7.4% (Lab 9.9%)
Portsmouth S: 8.0% (Lab 22.4%)
Southport – 9.2% (Lab 12.8%)
Newton Abbott: 10.5% (Lab 11.4%)
Brecon & Radnorshire: 10.2% (Lab 15%)
Devon North: 10.7% (Lab 8.9%)
St Austell & Newquay: 12.4% (Lab 13.8%)
Mid Dorset & N Poole: 13.1% (Lab 10.1%)
Oxford W & Abingdon: 13.4% (Lab 15.9%)
What’s this doing to the election mathematics?
According to UKPollingReport the above Lib Dem held seats are part of the 114 constituencies which the seat calculators say the blues need to win simply to get a majority of one. The notional majorities from the Rallings & Thrasher projection are shown. Listed as well are the R&T projections for the Labour shares.
Yet it’s clear that the battles in these 23 seats will be totally different from LAB>CON marginals where the general election will effectively be a referendum on Labour and Mr. Brown.
Even more than in previous elections I contend, Labour supporters will switch from their allegiance to stop the Tories winning in THEIR constituency.
We had a hint of the scale of that in October’s Bedford mayoral by-election where the Labour vote totally collapsed in an area where they hold the main Westminster seat. Stopping the Tory became the paramount objective.
It follows that if the Tories are going to find it harder against Lib Dem incumbents then they need to be doing disproportionately better in the LAB>CON marginals where bigger swings are required.
There’s another element that affects the polls. What are Labour supporters in LD-held seats and targets saying if invited to take part in voting intention surveys? My guess is that many are sticking with their many allegiance even though it’s not the party they will end up voting for.
According to one ICM survey on the eve of the 2005 general election 30% of all Lib Dem voters nationally said that their vote was not for their first choice but against the party that they wanted to stop in their constituency.
This is one of the factors, I suggest, that contributes to Labour over-statement in the polls when tested against real elections. Their voters have been much more likely to vote tactically than Tories and very little of this is picked up this far out.
In the New Year I’m hoping that the PB/Angus Reid relationship might be extended to polling the key marginals in order that we can get a more refined picture. We are also planning to include some LD>CON seats to find out how much they differ from the national swing.