Scottish Swingometer
Posted on January 29th, 2007 by Anthony Wells
I’ve had a couple of comments in the past about the lack of a nice online swingometer for the Scottish elections. For those after one Weber Shandwick have launched a rather beautiful guide to the Scottish elections, including a Swingometer, at Scotland Votes.
Filed under: Scotland, Uncategorized
…and yet still nothing about the Welsh Assembly elections being held on the same day. I can understand that the Scottish Parliament, having more powers and being the focus of the devolution debate, is the subject of more attention than Wales, but for there to be no polls and no news at all? This sounds unusual. Anthony, anything you can offer Wales watchers will be welcomed.
Thanks for the site reference, It seems pretty useful. I’ve just got the tory regional list rankings from the site, which means one’s picture of the forhcoming elections is almost complete now. it’s a better site than the alba publishing site on the scottish elections in many respects.
Dave – alas not, as far as I’m aware no one has commissioned any voting intention polls for Wales recently (though I’d be delighted to hear differently).
For a bit of Welsh perspective, check out:
http://independent-wales.blogspot.com/
Support for Welsh independence is growing strongly and there will be a number of shocks in May….
I think a less biased view on what is happening in Wales would be of rather more use .
Wikipedia reports (http://en.wikipedia.org/wiki/Welsh_Assembly_Election_2007#Opinion_Polls) that the Institute of Welsh Affairs has conducted a poll, which has Labour down to 25 seats (although I can’t find any reference to it on their website).
Thoughts, Anthony?
New poll in the Scotsman today has
FPTP (local)
SNP= 33%, Labour= 31%, LibDem= 17%, Con= 13%, SSP= 3% ( even though they aren’t standing FPTP) Others= 3%.
Regional (List)
SNP= 33%, Labour=27%, LibDem= 17%, Con= 14%, Green= 5%, no figures for others.
On these figures John Curtice is predicting,
SNP= 44, Labour= 41, LibDem= 23, Tory= 17, and Green= 1, with 3 to others.
However, put that in the predictor and you get.
SNP= 45, Labour= 42, LibDem= 23, Tory= 18, Green =1, which I suspect means that the predictor has trouble with the likes of Denis Canavan ( though he won’t be standing) and Margo McDonald.
Peter.
re Peter’s comment. The predictor does take into account Denis Canavan, however there are difficulties, 1 being Jean Turner who announced she was standing down and then said she may still stand. The major problem is in the figures published – ther predictor allows you to input results for others (which would cover SSCUP for example) and Independents, which would cover Margo. The published figures don’t seem to split these two and that is where the difference comes in. That’s the one area where some human intervention helps to analyse the results!
It’s actually harder to calculate which seats go where than a lot of people assume. Proportional Reprsentation doesn’t just mean allocating seats where they should go by working out percentages of the vote. Furthermore, the PR element is a top up, so not all regions are that proportional anyway. Whatever the result in May, Labour will have more seats in West, Central and Glasgow than entitled to under strict proportionality and the same with the LDs in highlands.
Excell spreadsheets are quite useful for calculating PR entitlements under the d’hondt rule actually, but you have to take away the seats a party has already won on FPTP, and that will mean that parties which should have been awarded the last seats in the region will miss out if others overscore on FPTP.
Joe,
I think you mean underscore. If you get four FPTP seats it’s divided by five, but if you get five it’s divided by six, so you can just fail to get a FPTP and make it up on the list, but if you get the FPTP you may lose the list seat.
Again even if you romp a FPTP seat getting more votes than everyone else put together it still only counts as one. Equally the three island seats all put together have scaresly more voters than a glasgow seat.
All in all Open list STV would be a fairer and better system.
As far as I can see the predictor works out the FPTP seats first and allocates them by party and region, and then applies the top up soit works quite well. Where it fails as all national polls do is that it doesn’t account for regional variations.
At some point in the next two months I am hoping for a 2,000+ survey that will give us some kindof regional breakdown. I
t’s just anecdotal but I have heard that although both Nicol Stephen and tavish scott willl be easily safe, the controversy over their ( and Raffen’s) expenses has tarnished the Libdems Goody Goody image, in places like Edinburgh.
Having said that I am in Rossshire, so I’d be interested to know if it is a factor with people in the central belt.
Peter.
Peter,
I do mean overscore. What I mean is in highlands, on a strict d’hondt principle out of 16 seats (9+7) the LDs would score 4 seats. They get five. That means that the the party that would get the sixteenth seat under a strictly proportional basis (I think it was the SSP last time) doesn’t get that seat. In Central, Labour’s over performance in FPTP deprived the SNP of a seat they would have had if it was just worked out on d’hondt. This is partly why Labour losing FPTP seats to the SNP benefits everybody- it gets parties seats they should have had anyway under strict proportionality even if they don’t improve their vote.
Ultimately, it’s why Labour can- and I think will win more seats than the SNP even if the two parties are close in actual votes cast.
One interseting effect of the current prediction is the way the map changes to see Labour reduced to a block of predominantly West Central Scotland seats many of which are dominated by economic decline.
That may be a sign that old allegiences that have lasted far longer in Scotland are now breaking down. It could also be used by cameron as a sign of a failed party that can now only hold on where it has created dependency, and failed economically.
Peter.
Peter,
what are your gut feelings on tactical voting? Intuitively, the SNP will look to gain Galloway & Upper Nithsdale from the Tories, yet their vote collapsed there in 2005 – are they likely to win it or could the Labour/LibDem vote be squeezed to let the Tories in? Similarly, could they perhaps win constituency seats like Perth or Angus as Labour voters tactically back the non-entity Tories to deprive the SNP of a seat? Likewise in Pentlands, Ayr, Dumfries – will SNP voters put a Tory in the constituency to cost Labour seats?
Finally, with top-up seats, will it make any difference to the overall standings?
Paul D,
Part gut feelin, part the yougov poll breakdown.
All the indications are that SNP voters will be the most loyal in this election, with LibDems the most likely to switch. I think that if SNP voters switch it will be tactically against either labour, LibDem or tory where the SNP can’t win, to either LibDem or Labour.
I just can’t see them voting Tory even to get Jack the lad the boot.
Of more importance to the overall outcome will be if labourvoters switch to the SNP in greater numbers than a uniform swing to get rid of LibDems, or LibDems to get rid of Labour.
I think you need to draw a difference between members and activists on one hand and the general public on the other.
There is general disgruntlement with the execuitive and a time for change mood, and I think labour will suffer from both ageneral swing and tactical voting. I also think the SNP will be the main beneficary.
What I can’t put my finger on is whether the current LibDem tactic of “we’re not like those nasty parties”, will work as the have been part of the execuitive for eight years.
Somhow I just don’t think they can get away with saying, ” We LibDems are responsible for all the goood things, and all the bad bits are labours fault”.
They claim credit for “Free Care”, but they were part of the execuitive opposed it up until the last hour on the last day of debate.
If I was the SNP campaign manage I would run a Party Political Broadcast of all the senior LibDems in Parliiament making speeches against it, contrasting what they are claiming credit for with what they argued for.
Peter.
Latest iCM poll in the scotsman, has;
Constituency; SNP- 34%, Labour- 29%, LibDems-16%, Tories-16%.
Regional List; SNP- 32%, Labour- 28%, LibDems-17%, Tories- 15%.
I only read iton-line so I haven’t seen the full tables, so I can’t give figures for the SSP or Greens. The Swing calculator ( I used 0% SSP, 3% Ind, 2% other on FPTP, and 1% SSP, 4% Green, 2% Ind, and 1% others on list) gives.
42 Labour (-8), 45 SNP (+18), 23, LibDem (+5), and 17 Tory (-1), It also shows a wipe out for the SSP and the greens losing six of their seven seats, with only 1 independant and no others.
That would give an SNP/LibDem alliance a majority of seven, and make a Labour/Libdem alliance, a majority of one. But then as the speaker can’t vote, it might need to be a tory or green, or even and here’s a thought Margo McDonald, who looks likely to be the only Independent to survive.
The New Scottish Voice may not have registeredyet and the pensioners party either.
This is all bad news for the greens who are on a knife edge again, a 1% gain on their 2003 result could give them three more seats to ten, a one percent drop on 2003 could see them all but wiped out. It also makes a rainbow coalition look farless likely.
It also highlights again the problems with the constituency list split caused by the way Labour and the Libdems designed the system in the constitutional convention to suit themselves.
The libDems are level with the Tories FPTP and two points ahead on the List, butwhere as the tories get 17 seats (13%), the Libdems get 23 (18%). Having said that in any system just as with the green for smaller parties one or two percent can make a big difference to the seats won, and it’s still miles better than Westminster.
It also means that if Labourclose the gap by just one or two percent they could still be the largest party ( I think that is what Jack McConnell is hoping for), or if they slip one or two, they could get under 40 seats putting another coalition out of reach.
Equally if the LibDem second vote is as soft as it appears from Yougovs last poll the greens could still pick up enough to get a half dozen seats, and who knows the tories might even come above the LibDems, but get fewer seats.
Peter.