UK General Election - May 7th 2015 (The Official Campaign Thread)

[jaichind]

I think 538.com prediction is merely the forecast from http://www.electionforecast.co.uk/

[jaichind]

Nigel Farage claimed an early advantage in the televised general election leaders' debate, with 24% of viewers polled at the half-way point rating him the best performer, ahead of Ed Miliband on 21% and David Cameron on 19%.   Pollsters ComRes questioned 615 viewers for ITV News half-way through the two-hour show and found Farage was rated best performer followed by Miliband and Cameron,  with the Scottish National Party's Nicola Sturgeon on 18%, Liberal Democrat Nick Clegg on 10%, Green Natalie Bennett on 7% and Ms Wood on 2%.

[jaichind]

You Gov Debate Result: Sturgeon wins

 Cameron: 18% Miliband: 15% Clegg: 10% Farage: 20% Bennett: 5% Sturgeon: 28% Wood: 4% 1117 GB adults

[jaichind]

Comres poll

PERFORMED BEST
21% Cam
21% Mili
9% Clegg
21% Farage
5% Bennett
20% Sturgeon
2% Wood

[jaichind]

ICM

Miliband: 25%
Cameron: 24%
Farage: 19%
Sturgeon: 17%
Clegg: 9%
Bennett: 3%
Wood: 2%

[jaichind]

Looks like the polls really show a 4 way tie between Cameron, Miliband, Sturgeon, and Farage.  In that sense this should be seen as a victory for Sturgeon and Farage.

[jaichind]

Conservative parliamentary candidate for Hull West and Hessle, Mike Whitehead, leaves the party to join UKIP, leader Nigel Farage says in Tweet.

[jaichind]

Here's my forecast using the average of the polls in the first week



Conservatives 273
Labour 302
Liberal Democrats 38
Scottish Nationalists 15
Plaid Cymru 2
Greens 1
Speaker 1

I will be forecasting Northern Ireland once nominations close today

You really should change your methodology as far as Scotland is concerned. Most prediction models have the SNP getting three times as many MPs. If they only win 15 seats it would be one of the worst polling failures in history.

I agree.  You should use something like http://www.electoralcalculus.co.uk/homepage.html which allows you to put in votes share predictions by region which will give a more realistic seats distribution than just a all UK wide polling.

[jaichind]

Thinking about this from a game theory point of view, I think the CON should do the following given the situation in Scotland:  Publicly ask LAB for a CON-LAB pact that which ever party emerges with the largest number of seats, the other party would support the winning party to form a government and would abstain in the vote of confidence. 

If LAB agrees and signs this agreement, then the CON can appeal to UKIP supporters saying "Vote CON to ensure a CON government since UKIP MPs are not necessary to prop up a CON government that will give you a referendum on EU."  While LAB can appeal to SNP supporters in Scotland saying "Vote LAB to make LAB the biggest party and block the CON from forming a government."  It seems to me that the first appeal would be more effective than the second since in Scottish voter eyes, this deal is and attempt to marginalize Scotland and for dignity sake they will more or less still go with SNP.

So say LAB realizes this and rejects this deal.  Then CON can appeal to the centrist voters saying "by rejecting this deal LAB shows it true colors of a secret deal with the far left SNP to form a government while the CON is trying to do everything possible to keep fringe parties like SNP and UKIP from power"  The CON can also appeal to the right, especially UKIP supporters, saying "see, Labor wants hoist the separatist SNP on the UK.  People of England ! Unite behind CON to stop this !" 

[jaichind]

The long-expected Tory surge doesn't seem to happen.

Their only (unlikely) hope for gaining a majority is another verion of 1992's "Shy Tory factor".

I'm not sure why the bookies still have them as favorites to win the most seats - would have thought it to be 50/50 at best. Any explanations for this?

I am surprised at this as well.  Perhaps what the bookies are counting on is LAB getting low single digit seats in Scotland but getting a fair share of the vote in Scotland and losing to SNP by narrow margins across the board. 

[jaichind]

I noticed over the last couple of weeks that the odds on Ladbrokes in terms of seat over/under

http://sports.ladbrokes.com/en-gb/Politics-c110000037#

always seems to match (a seat or two)

http://www.electionforecast.co.uk/

Did the two make a deal where ladbrokes will just use http://www.electionforecast.co.uk/ number ?

[jaichind]

Just watching Election Tonight on BBC Parliament.

The presenter talked about how reliable the opinion polls have been over the last 5 general elections. The one common theme in all of them is that the Conservative vote has been underestimated in all of them (and in 1992 by a large amount).

Could be the shy Tory voter effect or a late swing over the last couple of days before polling day as undecided voters finally make up their mind.

So keeping this in mind expect the Tory vote to be a couple of points higher on election day than the final opinion polls end up showing.

Yes, but my impression was that in 2010 the polls generally got the CON vote share correct.  They overestimated LD and underestimated LAB.

[jaichind]

Just watching Election Tonight on BBC Parliament.

The presenter talked about how reliable the opinion polls have been over the last 5 general elections. The one common theme in all of them is that the Conservative vote has been underestimated in all of them (and in 1992 by a large amount).

Could be the shy Tory voter effect or a late swing over the last couple of days before polling day as undecided voters finally make up their mind.

So keeping this in mind expect the Tory vote to be a couple of points higher on election day than the final opinion polls end up showing.

1992 was the last time the Tories were underestimated to a meaningful extent by the final polls, although Labour has been overestimated in all since and including then (with the exception of 2010, when they were underestimated).

Well, if you look at it from the point of LAB-CON gap, in 2005 2001 1997 it was the clear that the LAB lead over CON were exaggerated, mostly because the LAB vote was overestimated even as the polls got the CON vote share mostly correct.  So perhaps we should speak about the tendency of polls since 1992 to overestimate the LAB-CON difference in favor of LAB, except in 2010 of course where it was the other way around.   

[jaichind]

Is there a chance that Sinn Féin will take their seats if it make a difference in being about to bring down a shaky coalition minority government, especially a CON one?

[jaichind]

So I guess the battle will be which "bloc" will end up bigger: "CON+LD" vs "LAB+SNP+PC+Green".  In that sense even in fairly pessimistic scenarios the center-left bloc will have the upper hand even as CON will more likely than not be the largest party.   Only thing CON hope for now between now and the election is the shy Tory effect and some UKIP voters going to CON tactically. 

[jaichind]

A substantial number of English voters would rather see the United Kingdom break up than accept a Scottish nationalist role in the Westminster government, according to a new poll.  The Survation survey for the Mail on Sunday found that more than two-fifths (44%) said that they would rather see Scotland given its independence than have a minority Labour government rule with help from the Scottish National Party, against 31% who said they would prefer the UK to stay together.  And more than one-third of English voters (34%) said that an election result which led to Ed Miliband running the UK with the backing of SNP leader Nicola Sturgeon would not be "legitimate" - just a fraction fewer than the 35% who said that it would be legitimate.
 

[jaichind]

http://www.politics.co.uk/news/2015/04/26/miliband-rules-out-confidence-and-supply-deal-with-snp

Miliband rules out confidence and supply deal with SNP

[jaichind]

Any sense out there on what is going to take place in Bradford West ?  Betfair seems to have Galloway as the favorite but all the election projection sites seems to put Galloway in the single digits in terms of vote share with LAB winning by a landslide.  I cannot seem to find any polls on this.   

[jaichind]

Any sense out there on what is going to take place in Bradford West ?  Betfair seems to have Galloway as the favorite but all the election projection sites seems to put Galloway in the single digits in terms of vote share with LAB winning by a landslide.  I cannot seem to find any polls on this.   

Putting Galloway in single digits is a somewhat odd way of asking 'who knows?'. He's probably going to hold most of his core demographic (which is a good chunk of the Bradford West constituency, of course) which may or may not be enough to hold on.

Yeah.  See  http://www.electoralcalculus.co.uk/conlist_a_b.html#BradfordWest

Which puts LAB chances of wining at 94% with OTH which I have to assume is RESPECT a 1.3% !!!   

Of course

http://www.electionforecast.co.uk/tables/predicted_vote_by_seat.html projects Other at 3% vote share

And

http://www.ukelect.co.uk/20150425ForecastUK/MainPartyPercs.html  has Others at 5.42%

Kinds of make it hard for me to take these sites seriously.  I get it of course, since the way they derive these results are based on a statistical model based on polls and 2010 results.  My response to that is why cannot they also take by-election results into account as part of that model. 

[jaichind]

He does have a contingency plan if he does lose ... He's going to become the new Lutfur Rahman.

The very fact he would say something like that tells me he thinks he will lose.

[jaichind]

One thing that is interesting that took place last couple of weeks is the surge of GBP by over 5%.  This is partly about the decline of USD, but a key part of it is the believe among currency traders that a rate rise is coming sooner than later.  I noticed looking at intraday GBP charts that when a poll that shows a significant lead comes out GBP does spike up.  The main takeaway for me is

1) FX traders tend to feel that CON victory is more likely than what the polls or CW believe which is good news for CON
2) Even if LAB wins or forms the government that would not really alter the economic landscape that much and a rate rise will come anyway which does knock a hole in the CON narrative that this election is about preventing a disastrous Left-Far Left LAB-SNP government.

[jaichind]

1) FX traders tend to feel that CON victory is more likely than what the polls or CW believe which is good news for CON

Why? These people have no better information than the rest of us.

They have more $$$ on the line which means they will take the time to figure out what would take place.  You can argue that political betting markets play the same role but the liquidity of FX markets are a lot greater so the mounts of $$$ involved are lot greater which in turn means the level of research that much higher.  Of course this still could mean they are way wrong.

[jaichind]

http://washpost.bloomberg.com/Story?docId=1376-NNGJOM6S972O01-0JUKG9906S9KG5CPCR5EV4V91S

U.K.’s Wealthy Non-Domiciles Are Packing Their Bags, Forte Says

[jaichind]

You can interpret that quote how ever you like...

Then afterwards, Ed calls Nicola and asks: "Do you think they bought it?"

"Yeah, you were awesome. You almost convinced me."

Then they laugh and laugh and laugh.

I tend to think that if there is going to be a SNP landslide in Scotland then a very different behind-the-scenes deal might be in the works, that of a CON-SNP government.  If SNP wins 55-59 seats with a majority of the vote share in Scotland and the CON beats LAB by a solid 15+ seats then a CON-SNP deal might emerge.  The basis of such an alliance would be both a Scotland Independence referendum in Scotland as well as a EU referendum in UK.  CON cannot get LD to agree to a EU referendum and SNP cannot get LAB to agreed to a Scotland Independence referendum.  SNP would in fact benefit from holding both referendums on the same say and would urge Scotland to vote for independence since there was a risk that UK would leave EU.  CON would benefit from a rump UK where it could dominate England.   

[jaichind]

https://uk.isidewith.com/political-quiz

CON   94
UKIP  78
DUP   61
LD     27
LAB   12

Interesting.   Despite this if I get to vote I would vote UKIP if tactical voting were not to take place.