Projected effects of the boundary change
I've now had chance to plug the English Boundary Commission's provisional recommendations into Excel and calculate some proper notional results for the proposed seats. My projection is that the provisional English boundaries would reduce the number of Conservative seats by 5, the number of Liberal Democrat seats by 7 and the number of Labour seats by 18. On the new boundaries, the Greens would not have won a seat in Brighton (but see the caveats below).
Labour do worst in London, the West Midlands and Yorkshire, where they would have won 5 fewer, 4 fewer and 3 fewer seats respectively. The Liberal Democrats would have won 2 fewer seats in the South West and the North West, and one seat less in the South East, London and the North East. The Conservatives only do really badly in the North West.
Clearly the Conservatives have come out the best, which is to be expected (boundary changes always favour the Conservatives, because of the general population movements in the UK - population in Labour voting areas like Northern cities is falling relative to Tory suburbs), Labour lose the most numerically, the Liberal Democrats lose the most as a proportion of their seats (though, again, see the caveats at the bottom of this piece)
As I've said before, the notional change in seat numbers doesn't tell the whole story. We'll get a better understanding when we get chance to look at the changing patterns of marginals - are there more or less winnable marginals for each party on particular swings, do the swings needed for each party to win change to any significant degree? We won't have full answers to that until we have the Scottish and Welsh boundaries, but I'll try to have a stab in the coming days.
I'll also put up some more detailled analysis of regions, which seats are shifting hands and which MPs are losing seats in coming days. It's taken time because of the extent of change - in many places the pattern of seats in a county has changed to the extent that it's hard to say which seat is the successor to another.
For those interested in the mechanics of the projections, I've used a method that is basically the same as that used by Rallings & Thrasher for the "official" notional figures used by the BBC, ITN and Sky when boundaries change (see here for the detailed explanation from when I did the same calculations at the last boundary review).
There are a couple of important caveats about notional results - they are an attempt to show what would have happened if the votes cast at the last election had been counted on the new boundaries. They are NOT a prediction of what would happen in an election tomorrow on these boundaries - clearly Labour have gained support since then and the Lib Dems have lost support. Neither are they a prediction of how people would have voted in an election in May 2010 on the new boundaries, as some people would have voted differently if the tactical position in their seat was different, and parties would have campaigned differently. This is particularly important for the Liberal Democrats, notional projections can often paint too bleak a picture for them because their support and campaigning is so focused upon winnable constituencies, some of the apparently bleak areas brought into the constituency may in reality have potential for them. The same applies to the Green party in Brighton.
The final caveat is, of course, that these are only provisional recommendations. Boundaries recommendations can and do change after the consultation stage.
