The answer is 42
Right - it's Sunday night and there's little else to do so let's engage in a little electoral prediction courtesy of the BBC's seat calculator. This device translates votes under the First-Past-The-Post electoral system into seats in parliament. If the exact 2001 pattern were repeated this time, the shares of the vote and seats for the 3 main parties and others would be:
Lab 41% 400
Con 32% 169
LD 18% 47
Oth 9% 9
Taking the minor parties out of the equation (for the moment - I'll probably do a post on UKIP later in the campaign), my prediction is that Labour will lose around 6 percentage points of that 41, due to a combination of anti-war defections to the Lib Dems, minor parties or staying at home, and the fact that Labour voters will be (slightly) less motivated than Tories. The Tories will do better than last time, but not spectacularly - say a 4% improvement. The Lib Dems will pick up a couple of points, largely on the back of Iraq. Fed into the BBC model this produces:
Lab 36% 344
Con 36% 219
LD 20% 52
Oth 9% 31
which is a Labour majority of 42.
The main point to make about this is that for a party to get a working majority on 36% of the vote would be pretty extreme, even for the UK electoral system. Some extra analysis I did shows that the Tories need about an 8 point lead even for a tiny overall majority! If I were a Tory and I saw these results I'd take to the streets at once, or at least go into coalition with the Lib Dems to get PR.
The bias in the system at the moment is unbelievable. I can even get Labour as the largest single party with the third largest share of the vote: try Lab 29%, Tories 31, Liberals 31, for instance. These kind of numbers make the Gore/Bush anomaly in 2000 look like a tiny incident.
The latest polls, showing a 4 to 7% lead, don't change my thinking on this - in fact quite the opposite. The comparable lead on the polls at this stage in the 2001 campaign was 16 points - 7 points higher than the final result. If today's ICM poll with a 7% lead has the same degree of inaccuracy as last time, the actual lead is zero - the two parties are running neck-and-neck. If I can get some betting action for a majority of 40 or maybe even a bit less, I might be going there. We'll return to this issue later in the campaign. A working week beckons... even for the guys in Longbridge, fortunately.
Lab 41% 400
Con 32% 169
LD 18% 47
Oth 9% 9
Taking the minor parties out of the equation (for the moment - I'll probably do a post on UKIP later in the campaign), my prediction is that Labour will lose around 6 percentage points of that 41, due to a combination of anti-war defections to the Lib Dems, minor parties or staying at home, and the fact that Labour voters will be (slightly) less motivated than Tories. The Tories will do better than last time, but not spectacularly - say a 4% improvement. The Lib Dems will pick up a couple of points, largely on the back of Iraq. Fed into the BBC model this produces:
Lab 36% 344
Con 36% 219
LD 20% 52
Oth 9% 31
which is a Labour majority of 42.
The main point to make about this is that for a party to get a working majority on 36% of the vote would be pretty extreme, even for the UK electoral system. Some extra analysis I did shows that the Tories need about an 8 point lead even for a tiny overall majority! If I were a Tory and I saw these results I'd take to the streets at once, or at least go into coalition with the Lib Dems to get PR.
The bias in the system at the moment is unbelievable. I can even get Labour as the largest single party with the third largest share of the vote: try Lab 29%, Tories 31, Liberals 31, for instance. These kind of numbers make the Gore/Bush anomaly in 2000 look like a tiny incident.
The latest polls, showing a 4 to 7% lead, don't change my thinking on this - in fact quite the opposite. The comparable lead on the polls at this stage in the 2001 campaign was 16 points - 7 points higher than the final result. If today's ICM poll with a 7% lead has the same degree of inaccuracy as last time, the actual lead is zero - the two parties are running neck-and-neck. If I can get some betting action for a majority of 40 or maybe even a bit less, I might be going there. We'll return to this issue later in the campaign. A working week beckons... even for the guys in Longbridge, fortunately.
3 Comments:
It isn't "bias", and it isn't even particularly a problem with FPP, it's the use of "swing" as the basis of the electoral calculations that says that Labour would win when in 3rd place on votes. Swing only represents half the combined change in voting % between the two parties, and assumes that what happened in 2001 is a valid basis for what is to happen on May 5th.
In the real world, parties need to win something of the order of 11 million votes, and nobody can say how that'll happen.
PS. I do support PR, by the way.
http://www.hovelabour.org/
After years of thinking that there was nothing wrong with FPTP, I'm now having my doubts. A recent "flyer" pushed through our increasingly disfunctional letter box bearing the grinning delights of Dr Evan "Parliamentary Bridge Player" Harris reminded me that he commands an 18% lead (ahead of the Tories) amongst the voters of dear Oxford West and Abingdon. So my vote will make precisely no difference. Thus my enthusiasm to become an "engaged and active" election participant was remarkably short-lived. Indeed, my best chance of exercising some democratic rights would probably be to fork out 500 sterling and put myself up for election. Not sure what the manifesto would look like - probably maximum working week, transferrable tax allowances for partners, legalisation of most recreational drugs, and re-introduction of the stocks - what better way to punish local ne'er-do-wells than hurling rotten eggs at them.
Juniper - If you're planning to stand I'll have a piece of that action - let me know and I'll send a sum of GBP. But you may be too late for the registration deadline this time round? I'm not sure.
Post a Comment
<< Home