Categories
politicalbetting.com » Blog Archive » Dear Dr. Fisher. This is what Professor John Curtice descri
politicalbetting.com » Blog Archive » Dear Dr. Fisher. This is what Professor John Curtice describes as “Labour’s crutch” – the 2010 LD>LAB switchers
Dr Stephen Fisher has responded to the first wave of comments on his new GE2015 prediction model which suggests that the Tories have a 58% chance of a majority.
0
This discussion has been closed.
Comments
The fact that polls have completely changed over this period, especially post-1992, is what makes Rod's by-election swingback thing interesting, because by-elections have used the same methodology consistently: Polling voters in seats where the sitting MP has died or been sent to prison, using a large, opt-in sample, and incentivizing them to take part in the survey by giving the winner an actual seat in parliament.
Not too sure Ed Balls underpants and environs are a suitable topic for weekend viewing !!
Further, the Tories didn't want a coalition, appear to resent their partners and appear to take pleasure in thwarting their objectives.
I was, and remain of, the opinion that mathematically and logically Clegg had no little or no choice in May 2010, but he's played a good hand very badly.
What a pathetic and sad existence you do lead. For existence it is. Most certainly not a life.
You cling 24/7 like some pustule of infected matter to the body politic of PB bringing disrepute to the party you support by the smears and lies that trip constantly from the seeping wound of your desperate existence.
You need help but I doubt that you have the nerve to seek it.
Gordon Brown has one last gift to give the Labour party he loves
Ed Miliband is hobbled by the myth of the mess Labour left behind. His predecessor can explode it, but he must say where he went wrong
http://www.theguardian.com/commentisfree/2013/oct/25/gordon-brown-last-gift-to-labour-party-he-loves?commentpage=1
However let's not forget that Ed Miliband was a fully paid up member of the "Cabinet of the Guilty." He wants to airbrush his history as Energy Secretary and his tenure as one of Gordon Brown's willing accomplices.
"More importantly, he could try to nail once and for all the notion that the increased deficit was due to incontinent Labour spending. Patiently, he should explain that the deficit mushroomed because of the great crash, which triggered an instant collapse in output and tax revenues: Labour had to borrow more because suddenly and unavoidably less money was coming in. That was the result of a global economic crisis that was not caused by the decision-making of Gordon Brown. To repeat the under-used Labour line of the time, the recession was made on Wall Street, not Downing Street."
"A flat purchased by Scottish actor Alan Cumming is facing an investigation by an electoral authority after questions were raised about the star’s eligibility to vote."
http://www.scotsman.com/news/politics/top-stories/alan-cumming-s-edinburgh-flat-faces-electoral-quiz-1-3158045
# Predicting an election five-hundred days out is difficult but, based upon historical analysis, it is likely that the Tories will pick-up vote-share as the election approaches,
# If the Tories do not start to see a recovery from two-hundred and forty days out then the model will reduce the probabilities currently assigned to them, and
# The last few weeks of the election campaign are unlikely to upset the then model predictions. *
Nothing new here is there? The bias to swing-back has been discussed here often but Stephen's model uses regression-analysis methods to compensate for outliers and untested datasets (whereas others exclude those that they don't like). The fixation on the "anomoly" of Labour being more likely to form a majority than be largest party underlies the uncertainty of forming predictions this far out!
Modeling using historical data-sets is obviously a flawed method to predict future events but it is a standard means of identifying trends. One has only to look at the forecasts of Brown, Darling and the OBR to identify that errors are often apparent over time. Instead of focusing on the outcome it would be better if the critics of the model tried to understand it's workings. This is why casino-simulations are important: This far out they may produce more outliers but these can be fed into the model to compensate for sampling-errors. If it was an easy science then Stephen's paper would not be contraversal but it is not: Trying to explain it's complexity to "believers" of double- and triple-dip recessions would require the patience of a saint....
* I would expect the model would need to be tweaked to account for "Presidential Campaigns", the internet and Al-Beeb bias.
The phrase "this time its different" has rightly been called the most expensive in the English language but certain key differences seem to at least question his premise. Firstly, there is the polling problem in past Parliaments. Without a sound basis in the data the model surely cannot work.
Secondly, there is the lefty Lib Dem problem. This is similar to a distortion in the polling in that it affects the actual recorded results of elections. It suggests that many natural Labour supporters have been voting Lib Dem in a way that they are no longer willing to do to keep tories out. The consequence is that Labour support is actually understated in the actual election results.
I am not sure this is the unadulterated bad news for the tories that it is portrayed in that the reason that those lefty supporters were voting Lib Dem was indeed to keep the tories out and if they don't then the tories will in many cases get in. In short many of these extra Labour votes will be wasted votes. But the consequence of such a distortion unwinding is to undermine the data basis of Dr Fisher's model.
Thirdly, as Mike has pointed out there is the robustness of tory support. If hardly any tory supporters have switched to Labour (Sunil seems almost it as far as I can tell) there are none to switch back.
I very much hope he is right of course. It is just a little too good to be true for me though.
Pr(Lab majority) = 15% & Pr(Lab largest party) = 12% are not compatible.
How can Lab have a bigger chance of a majority than being the largest party?
Fair cop. These figures are logically incompatible. Clearly I need to revise the method to ensure these are fully consistent, and I will do. For computational reasons the method does not currently estimate a full uncertainty distribution over all possible seat outcomes which is necessary for guaranteed strictly consistent estimates. If you read the paper you'll see that, at present, the probabilities for Con majority, Lab majority and difference between Con and Lab are generated from three different assumed normal distributions from the prediction intervals (for Con seats, Lab seats and the difference respectively). These are not strictly defined to be fully compatible. Much of the time this shouldn't matter but did on in this situation and I should have at least flagged this up and discussed it or solved the problem before publishing. Sorry.
See OGH link
http://electionsetc.blogspot.co.uk/2013/10/initial-responses-to-comments-so-far-on.html
As a tory supporter I would be absolutely delighted if he increased his profile. It might even improve the prospects of Dr Fisher being right!
Secondly what is going to happen to the 2010 LD>CON switchers. In your model they seem to evaporate.
Third point the detail from almost all polls shows that very little of current LAB support comes from CON converts.
To what degree does the model take into account the point frequently made by Mike Smithson, that the loss of half the Lib Dem vote from 2010 has mostly gone to Labour, and there appears to be little sign that it will either go back to the Lib Dems, and very little sign that it will migrate to the Conservatives?
Model doesn't look at all at individual level switching, just overall levels of change. The predictions say nothing about who switches which way, just the net effect. Switching up to now is reflected in current poll levels which influence the forecast but just based on current levels of support, not origins or voting histories.
This very simple effort in many ways was one of the most significant British Rail commercials in that it uncovered Tony Kaye one of the more interesting directors of the last 20 years. He famously used 120 rolls of film.
What is the significance of that? How much is 120 rolls of film? How does it compare with what would normally be used on a 42-second advert? Is it about average? Or is it far more than normal? Or is it far less then normal? And, whichever it is, what is the reason?
In other words - bearing in mind that I have no frame of reference to compare it with - why am I supposed to be impressed by the statement that he "used 120 rolls of film"?
A standard reel of 35mm film runs 11 minutes - so if that's what Roger means the director shot 131 minutes or 7860 seconds of film to produce 42 seconds of commercial......(actually 40 - so the YouTube one is not exactly the one that ended up on air).
Roger may see that as an indication of the director's dedication to his craft.
As a client I would have viewed it as self indulgent waste of the highest order - one of the reasons most commercials are now shot on video.....
The danger of a Brown intervention is internal to the Labour Party. First, it might reignite the Blair/Brown wars; second, it would distract from Labour's current message (if and when Labour develops a current message).
Good morning, everyone.
No tips, but my pre-qualifying piece for India is here:
http://enormo-haddock.blogspot.co.uk/2013/10/india-pre-qualifying.html
He could also state with complete honesty, but little humility, that once the crisis had struck, he acted swiftly and decisively to save the country from the most catastrophic consequences of the failure of the banking system. As evidence, he could point out that not only did we avoid a complete meltdow, but that countries did too by following his example. In this respect, if in no other, his leadership at the G20 Summit was crucial and decisive.
I thought he was a wretched Prime Minister, the worst since WWII. I did not vote Labour, largely because of this. But it could be argued that his handling of the banking crisis outweighed all his other faults when you consider what the consequences would have been had he not acted so decisively in, effectively, nationalising the banks.
The big problem for me is the anomoly of the 1997 election and it's impact in 2001 and the inertia of 2005: They are the oddity as it represents a "historic" outcome for Labour. However I can counter this by equating this by the anomoly of 1983 (with those SDP spannahs) causing 'above-trend' movements causing impacts in 1987 and inertia in 1992. If we seek to negate them and your 2010 candidate we won't have much evidence to base the model on!
I think the big criticism of any feedback that Stephen has received has come in the form "but the polls are not saying that" as if current opinion is a better proxy for a future GE outcome. That is not what Stephen is using his model for: It is more behavioural. It appears to demonstrate a number of traits - based upon historical precedence and outcomes - to define a prediction for electoral outcomes based on current polling, government factors, historical trends and the time left until the next election. And as Carlotta has pointed out the model has feedback mechanisms that will refine these predictions. "We are where we are" as one droll contributor keeps pointing out....
"Two threads ago, Roger wrote.........."
Sorry but I got called away and ended up leaving just a skeleton post which I should have deleted. 120 rolls is 8 hours. Take away the titles and animation and the film is just over 30 secs live action which means roughly 4 secs used for every hour shot. It also means that just to view the material takes the editor a working day which was my original reason for the post because I used the same editor.
Seeing American History X is well worth doing as is reading about the making. He's very talented. Most of those who work in the industry will have a story about him. My political one is when the ruling party in Germany got him to do a PPB for them and he was thrown out of the chancellory for insulting just about everyone not least the chancellor. I worked with his crew just after his aborted PPB. His original production company was known as 'The Wandering Jew'.
http://upload.wikimedia.org/wikipedia/commons/2/25/Tony_Kaye_2011_Shankbone.JPG
"Owning a property in Scotland would not be sufficient to allow an individual to vote. According to a spokesman for Edinburgh’s electoral registration office, Cumming would be required to live in the flat for at least six months of the year."
Cumming thought he could buy a vote for £82,000 - but he can't.
So much of the LibDem vote in recent years has comprised the "neither of the other 2" fraternity and are our experts seriously suggesting they will now vote for one of the other 2.
The LibDem vote outside a dozen or so seats has not been substantial for more than 25 years. The majority of LibDem switchers are likely to be people who moved from Labour to LibDem in the first place just as most of the LibDem to Tory switchers will be people who moved from Tory to LibDem since 1997 (1987 in Scotland).
http://www.aboutmyvote.co.uk/the_independence_referendum/frequentlyasked_questions.aspx
Cumming is a US citizen resident in Manhattan.
It is massively in the Tory and Lib Dem interests to manage the coalition split so that Lib Dem voters leave Labour and come back to them. This hugely increases a) the Lib Dem chances of retaining their seats and b) the Tory chances of winning in the marginals.
We are starting to see more open criticism of eachother between Lib Dems and Tories and this will be the beginning. It will be slightly stage-managed and appealing to their individual bases, but the success of how Nick Clegg extricates the Lib Dems from the Tories is critical.
Brown got 42% in Scotland - you have a point there.
(OT. Carlotta. Oh that he did just use 130 mins of film! it was 120 rolls. A roll of 35mm is usually 4 mins. 10 mins is a long roll or 16mm. He used 120 rolls at 4 mins a roll. (15-25 might have been normal). The cost of about £200 a roll isn't really a deal breaker in the scheme of things but I'm glad I discovered what a parsimonious client you might have been had we worked together!.)
I think the Tories have an enormously difficult task of winning the next GE. Outside of southern England they are far more disliked than they were in 2010. Even if there is no great desire for Labour in 2015, it can't be helpful to the Tories that Cameron's star is on the wane.
If you were Nick Clegg, and you knew you were in charge of the Lib Dem 2015 campaign, at what stage would you leave the coalition? I think the Lib Dems have to leave, and they have to do in such a way that they take credit for certain things and slam the Tories on others. A really difficult task. But the timing is the first critical thing. When would you break?
I have never heard of this "six-month" rule before.
I think the Lib Dems will stay to the bitter end - they'd feel right chumps walking out before a give away budget in 2015!
This cohort of voters is motivated, consistent and IMO likely to vote in large numbers. It is in that respect quite different to the classic Labour AND Tory vote, some of whom for both parties frankly vote from a resigned belief that the party is a lesser evil and who might not bother on the day.
Another interesting Fisher quote (and we should welcome his willinginess to do a model and discuss it lucidly): "For want of a better reason to do differently I have effectively produced the forecast in the same way as if the Conservatives were in a single party government." He argues that people see the Government primarily as a Tory one so the dynamics should work similarly. I'm not sure that's true. The general impression is IMO of a somewhat chaotic coalition muddling along in admittedly difficult circumstances. It's one reason they don't get credit for successess, which are seen as happening despite them rather than because of any clear sense of common direction.
Unless of course he's paying UK tax.....but I'm guessing he spends fewer than 90 days a year in the UK......
In his own words:
Greetings my fellow Americans!
http://www.alancumming.com/blog.php?id=25
I'd go next summer, after the 2014 Brazil World Cup.
The coalition split will be a big news event, the two sides could plan the split. Do it amicably. Call a moratorium on criticism (from both sides) for a full two weeks and extol the fact that a mature coalition was formed between a centre right and centre left party and it broadly worked. They can then take as much credit as they can on jobs, deficit reduction, reform etc with the media all over it. It will at least give both sides a chance to steal the agenda for a fortnight.
I'd also be really honest about it if I were Cameron and Clegg. And explain to the public that electoral realities mean that both sides have to fight GE2015 as enemies and take the fight to Labour. By going into the GE as 'friends' they hand the election to Labour by default.
It may not work, but if they go all the way to April 2015 as a coalition I think it would be disastrous for the Lib Dems, and that would mean carnage for the Tories in the marginals. Perfect scenario for Labour.
This Unionist seems to think eleven days residence is sufficient:
‘People can register to vote up to 11 working days before polling day – it used to be 8 weeks. The referendum will be held on Thursday 18 September 2014, this means that an EU citizen who arrives in Scotland with a place to stay, on Wednesday 3rd September 2014, and who registers to vote that day, will have a say in whether Scotland should split from the UK, at the referendum a mere 2 weeks later.’
http://tinyurl.com/layjcoy
The "pathetic" thing here is some luvvie thinking they can buy a vote for £82,000.
Will Sean Connery be voting?
I have.
If you think that, Fenster, then you simply must take Paddy Power's 9/2 about the coalition breaking up in 2014.
I have.
Cheers Peter, will do. I lose every bet I every put down on politics so here's a new start!
I think this sentence has a bit of the lies, damned lies and statistics feel to it. Whilst it may be true it seems to be a very forced construct given that the most recent increase in the Tory vote between 2005 and 2010 was much greater and amounted to almost 2 million votes - well in excess of the LD-L switchers Mike is referring to.
And as I have said before it also ignores the huge numbers of voters who chose not to vote at the last election. To take Mike's baseline they are down 6% on 1997 - or over 3.5 million. The total number of potential voters who chose not to vote in 2010 was just shy of 16 million. They are a far greater pool of potential support for any party than any switchers. We also don't know how many voters - in light of the ever reducing lack of choice between the three main parties might just decide not to bother next time.
There are simply too many variables in play and looking at one specific Lib Dem faction as being the most meaningful indicator of the result of the next election seems daft.
"He argues that people see the Government primarily as a Tory one so the dynamics should work similarly. I'm not sure that's true. The general impression is IMO of a somewhat chaotic coalition muddling along"
I see it as a 'Tory government' which is one of the reasons the Lib Dems are in such trouble. They're not seen as being in government or being in opposition. Pollsters are reflecting the question 'what's the point of them?'
Nats would have no probs with that eh ?
I have.
Cheers Peter, will do. I lose every bet I every put down on politics so here's a new start!
Well, it might not come in, but I was chewing this over with PfP recently. He reckoned it should be 9/4, I reckoned more like 5/4; either way, there's heaps of value in the bet, and every likelihood you can hedge later.
I got £40 on in one of PP's shops. On reflection, I think I should have gone heavier, and may top up idc.
"The "pathetic" thing here is some luvvie thinking they can buy a vote for £82,000."
Bloody expensive isn't it even for a luvvie!
Well, it might not come in, but I was chewing this over with PfP recently. He reckoned it should be 9/4, I reckoned more like 5/4; either way, there's heaps of value in the bet, and every likelihood you can hedge later.
I got £40 on in one of PP's shops. On reflection, I think I should have gone heavier, and may top up idc.
Cool, I'll go for it. I think they will split. I can't see any other 'event' that'll cause Lib Dem voters to return.
I got £40 on in one of PP's shops. On reflection, I think I should have gone heavier, and may top up idc.
Cool, I'll go for it. I think they will split. I can't see any other 'event' that'll cause Lib Dem voters to return.
If the coalition breaks down it will be because of unilateral action by Tories not LDs. The yellows will stay until election day.
If Tories do kick LDs out they'd lose vote of confidence and there would be a 2014 general election. Ain't going to happen.
That group switched back almost immediately in May 2010 and have barely shifted since. I doubt there is much that the Lib Dems can do to lure them back while they are in bed with the Tories. They might come over all left-wing again at some time in the future but I doubt anyone would buy it in the next 18 months.
Aside from that the other key factor will be the extent that the Tories can get UKIP voters back in line. Problem is they will become like Dr Doolittle's Pushmepullyou, veering to the right to try and attract kippers and then tracking back to try and keep on board the 25% of current Tory voters who (according to YouGov) would not vote for them if they were in a pact with UKIP.
I can see the Tories being the largest party in 2015 but a 58% chance of an overall majority does not look right to me at all
The East Village, Manhattan, the Catskills, Polwarth.....? It's not exactly Merchiston, is it?
"Not when the average price of a 1 bed flat in Edinburgh is £110,000!"
I didn't mean for a flat (which sounds ridiculously cheap for Edinburgh) I meant for a vote!!
Taking in present house price increases, I suspect he’ll have bought a vote for free –or even a small profit.
"Ineos’s disciplinary action against Mr Deans began on July 9, four days after Labour called in the police at the height of the Falkirk row."
http://www.ft.com/cms/s/0/6ed4a1ea-3cb9-11e3-86ef-00144feab7de.html#axzz2ip1rapOG
The Labour Party call in the Police & his employers are supposed to do nothing?
Can't see it apart from at a NUT conference ...
I got £40 on in one of PP's shops. On reflection, I think I should have gone heavier, and may top up idc.
Cool, I'll go for it. I think they will split. I can't see any other 'event' that'll cause Lib Dem voters to return.
There have been people betting that the Coalition would break down and there will be an early GE every year since 2010 ( some of them the same people every year . They have been wrong all the time . The next GE will be as scheduled in 2015 , any other bet has and will be free money for the bookies .
1. Recovery: there was a short term boost timed to hit the election. This was achieved by a spending splurge and by bringing forward capex. Temporary, and utterly unsustainable. The Coalition had a choice between scaring the bejezuz out of (a) the people or (b) the markets. They chose to maintain the confidence of the market. Perhaps right, perhaps wrong - certainly an area for debate (I think they chose right). But not the open and shut case the author appears to believe
2. The deficit mushroomed because tax revenues fell. However, with hindsight it was clear that these revenues were never sustainable. Labour's failure was partly during the boom years - believing that they had achieved "no more boom and bust" and locking in permanently higher spending rather than assuming that revenues would come down in future. Imprudent at best. The other mistake was once revenues had come down assuming that they didn't need to adjust spending to account for their new income levels. This was what perpetuated the deficit.
Massive anti-Tory sentiment because the economy is recovering and unemployment is falling? Then again they are Lib Dems so nothing would surprise me
Every Free School story, every day that Gove is in place, is a big driver of tactical anti-Tory Lib Dems voting Labour, it's not just about the economy.
Although in the Lab/Con Marginals the 2010 LD's give Labour a 21% lead on the economy and jobs.
Along with a 22% lead on education and a huge 44% on the NHS
I would have thought Gove would regard upsetting Lib Dems on education as a badge of honour. No one has done more to ruin the chances of working class kids in this country than the Lib Dems
As well as many films, Bird's TV credits included Spooks, Cracker and, more recently, BBC One's The Village, starring John Simm.
(‘Cracker’ starring Robbie Coltrane was the best thing on TV during the mid 90s imho – RIP)