I’ve been catching up on sleep after the election, but this is just to add a brief, post-election round up of how the polls performed. In 2015 and 2017 the equivalent posts were all about how the polls had got it wrong, and what might have caused it (even in 2010, when the polls got the gap between Labour and the Conservatives pretty much spot on, there were questions about the overstatement of the Liberal Democrats). It’s therefore rather a relief to be able to write up an election when the polls were pretty much correct.

The majority of the final polls had all the main parties within two points, with Ipsos MORI and Opinium almost spot on – well done both of them. The only companies that really missed the mark were ICM and ComRes, who understated the Tories and overstated Labour, meaning they had Conservative leads of only 6 and 5 points in their final polls.

My perception during the campaign was that much of the difference between polling companies showing small Conservative leads and those companies showing bigger leads was down to how and if they were accounting for false recall when weighting using past vote – I suspect this may well explain the spread in the final polls. Those companies that came closest were those who either do not weight by past vote (MORI & NCPolitics), adjusted for it (Kantar), or used data collected in 2017 (Opinium & YouGov). ComRes and ICM were, as far as I know, both just weighting recalled 2017 past vote to actual 2017 vote shares, something that would risk overstating Labour support if people disproportionately failed to recall voting Labour in 2017.

The YouGov MRP performed less well than in 2017. The final vote shares it produced were all within 2 points of the actual shares, but the seat predictions showed a smaller Tory majority than happened in reality. Ben Lauderdale who designed the model has already posted his thoughts on what happened here. Part of it is simply a function of vote share (a small difference in vote share makes a big difference to seat numbers), part of it was an overstatement of Brexit party support in the key Conservative target seats. Whether that was having too many Brexit supporters in the sample, or Brexit party supporters swinging back to the Tories in the last 48 hours will be clearer once we’ve got some recontact data.

Finally, the 2019 election saw a resurgence of individual constituency polling, primarily from Survation and Deltapoll. Constituency polling is difficult (and I understand has become even more so since the advent of GDPR, as it has reduced the availability of purchasable database of mobile phone numbers from specific areas), and with small sample sizes of 400 or 500 it will inevitably be imprecise. Overall, it performed well this time though – particularly given that many of the constituency polls were conducted in seats you would expect to be hard to poll: unusual seats, or places with independents or high profile defectors standing. David Gauke’s support was understated, for example, and in Putney constituency polling overstated Lib Dem support at the expense of Labour. However, in many places it performed well, particularly the Chelsea & Fulham, Wimbledon, Finchley and Esher & Walton polls.

And with that, I’m off for a nice Christmas break. Have a good Christmas and happy new year.

2,835 Responses to “General election polling – post mortem”

1 55 56 57
  1. This is a significant week regarding indyref2 announcements by the SG.

    I commented before that it had been notable that the UK Government had not offered anything other than a flat no to the SG which did not seem a very shrewd political response.

    Yesterday the SG made a very detailed proposal for a devolution of immigration policy and powers ( something as a possibility during the EU referendum campaign by Gove) suggested . The UK Government rejected it within 90 minutes. This morning the former Tory MP for Stirling, Stephen Kerr, said the proposal warranted serious discussion. Business organisations in Scotland, which had been involved in preparing the policy supported it. Will the UK Government take a more shrewd approach?

  2. SDA: Strange that HS2’s greatest support comes from the north and its worst support from the south.

    So a canny solution might be to build the two northern legs plus a new trans-pennine link and leave the Chilterns to fester in their prettiness. Maybe a few of the less controversial bits of the southern leg could be continued where money has already been spent, but I’m not sure how easily that could be patched into existing provision.

  3. “Maybe a few of the less controversial bits of the southern leg could be continued where money has already been spent, but I’m not sure how easily that could be patched into existing provision.”

    My understanding is that the existing main lines from London to the midlands and north of England are being used at capacity.

  4. No change on Twitter then (I do wish folk wouldn’t link to it, as it increase brain cell decay).

    “How can the SNP poll 5%? That is larger than the voting population in Scotland.”

    A quick search, and yes, he’s a person outside of the polity of which talks, and he’s pro-Brexit.

    My general rule of thumb is:

    Scotland is approx 8-9%. I favour 8.7%, and thus we can calc 100 divided by 8.7 equal 11.5.

    The range of 5% goes from 4.50 to 5.49, and equates as 51.8% to 63.1%.

    However, if the sample size is 150 in the cross-break, the value will be subject to +/- 8% points (43.8% to 71.1%).

    And that, people, is why it is so important to not only have reasonable poll samples, but also to go down to at least one decimal point. I’m at two where possible, as it’s not difficult.

  5. @ JAMES E

    Thanks for the article; it looks like our fisherfolk are going to have at least two problems: the EU may refuse tariff-free access to their fish markets, and/or may restrict the activities of the City of London if we try to ‘take back control’ of our fishing waters.

    Not looking good for them, esp considering that fishing accounts for 0.04% of our economy, and financial services is 7.1%.

  6. Patching Londonish bits of HS2 into an existing network is rather difficult. The existing WCML ex Euston heads north westish via Hampstead, Wembley, Harrow, Watford. The HS2 route heads almost due west at first using old Great Western land, in the main. I guess this is a leftover from when it was due to have a spur to Heathrow.

    It then heads out only slightly north of west ’til after Denham when it heads up more north towards Aylesbury. So, by the time HS2 reaches the edge of Greater London it’s about 8 miles away from the WCML and running beside the Chiltern Line.

    Don’t forget that as a high speed line bends have to be kept to the minimum and are very, very shallow, otherwise speeding trains fall off the rails.

    If you wanted to run a line the 8 miles involved is either prime Green Belt or prime suburbia and you’d have to start the planning all over again. You could meld the HS2 London bit into the Chiltern Line, but it would be fairly pointless without substantial upgrading to provide line capacity for the additional trains involved, That would involve planning and similar disruption to what’s already required for HS2. You could meld it earlier and plug into the Great Western mainline before reach the edge of Greater London, but why run a high speed stub from Euston to Old Oak, just a couple of miles. Again, taking it further along the GWR mainline would mean replanning and disruption as extra lines are added westward. Brunel’s railway is generally straight and flat at the eastern end but you’ve got bridges over the Thames etc and tunnels that would need doubling further west.

    I think the decision will be made for political reasons, not necessarily for the good of the country, which is not to say that it won’t be the right decision, but it will be made based on the wrong reasons.


    Can’t remember the reference, but I read somewhere that making travel connections quicker between an economically weaker area, e.g. Brum, and a stronger one, e.g. London, usually benefits the stronger area.

    On this theory HS2 will make the north/south divide worse, not better; it will mean that people can live in Brum and commute to London more easily.

  8. I can’t see people wanting to commute from Central Birmingham to London. It would be a very costly journey, as well as having to pay city centre prices for property. There are huge numbers of other places within an hour’s train travel of London which would be cheaper from the point of view of both transport and housing.

  9. @JAMES E

    Yes, you may well be right – I only mentioned it as a general theory on the difficulties of regenerating ‘left behind’ regions.

    Interesting article, about 10 days ago, by Larry Elliott (The Guardian) on Labour’s ‘Red Wall’ seats lost to the Tories. He points out that LAB made various attempts to regenerate these, without success – it’s not at all obvious, how to do this; and, of course, austerity made things much worse.

  10. Progress Scotland, established to research and promote the case for independence and led by Angus Robertson, commissioned polling in Scotland on which Parliament should have the final say on holding indyref2.
    Excluding don’t knows, 61% said the Scottish Parliament and 31% the UK Parliament:


  11. @tobyebert

    Tariff free access is not necessarily the key issue for the fishing industry especially for the inshore sector. Customs and regulatory issues are just if not more important as explained in this flow chart issed by the UK Government:


  12. Tobye

    As you say, regeneration isn’t easy, if it was we’d have nowhere that actually needed it.

    Mind, leaving the EU might have made it even more difficult. I recall that back in the ’90’s Prince Charles took an interest in Ripon and his Institute or Trust did a lot of work trying to help the city’s development. The city got substantial funds from the EU to carry out a number of regeneration projects. I can’t say whether any of them actually improved matters, but I guess it would’ve been in a worse state without them.

  13. SDA

    @”Strange that HS2’s greatest support comes from the north and its worst support from the south.”

    Not sure about that-local connectivity rather than a few minutes off the trip to London seems to be the Northern priority/


  14. “I can’t see people wanting to commute from Central Birmingham to London. It would be a very costly journey,”
    @James E January 28th, 2020 at 1:24 pm

    Yes, but you don’t need to keep going down to London every day. I’ve worked for companies in the past where just the sales guy goes down for the day, and the agreed work is actually done remotely. With more and more work moving on line, you need fewer days face-to-face. Indeed it was not just sales — when starting on a new project one face-fo-face meeting can be really valuable, and productive work can then continue remotely. In my industry, IT, this method of working is extremely common.

    Hopefully (but this is a forlorn hope) the ticket prices will not be stupid. The last time I travelled to London[1] I had to get an open return, because you don’t know when the meeting will finish, it was something ridiculous like £250 return from Manchester. I could drive and stay over in a hotel for that; or fly.

    That extra hour HS2 promises either means another meeting, or a chance to go to the pub to consolidate a relationship. In other words it offers more opportunities.

    [1] Admittedly a few years back now, these days I never leave the house as I work for a US team.

  15. Colin

    I think that some people are focussing entirely on the services which HS2 itself would provide rather than on the knock-on benefits from freeing up capacity used by the current fastest services. See this, for example:


  16. On the subject of regeneration I wonder if Cardiff demonstrates a truth that money follows, geographically, the political power. Cardiff’s population in 2001 (two years after the start of the assembly was 292,150, the current estimate of population is estimated in excess of 477,000 63% larger. Swansea the other large (relatively) city in South Wales in 2001 population: 167,440 and the 2020 estimate is in excess of 246,000 a 46% increase. The comparative growth of population is just one metric which shows over the last twenty years the economy and vibrancy of Cardiff has shot ahead of Swansea.
    In 1999 I campaigned for the assembly to come to Swansea for that very reason: if you want to distribute the wealth between the regions and nations of the UK you have to start, meaningfully, distributing the political power.

  17. @jamese

    It seems to me that one of the undermining troubles which the case for HS2 has suffered is the changing justification: it started with the emphasis on the economic benefits of reducing journey times, then moved on to freeing up capacity with consequent benefits for other services, and over the weekend the argument emerged that it couldn’t be cancelled anyway because it would be too expensive to do so and it would risk a number of large construction and civil engineering firms going under!

    This report from the New Economics Foundation raised questions about the spatial distribution of benefits ( London would be the major beneficiary) and the social distribution (top earners use the railways the most – even in Europe – and will benefit the most):


    At the moment HS2 is consuming over a third of total Government subsidy to the UK railways and presumably this would increase even more given the estimated huge and inevitably uncapped increase in project costs.

    There are claims about the benefits to the UK as a whole which presumably are needed to justify its designation as a UK project and so avoid Barnett consequential for the devolved governments as there seems to be scant evidence other than reducing joinery times to London by the 2040s (and bear in mind NEF report for possible consequences). Any benefits to Scotland will I understand require major upgrades to the ECML and WCML (which are not included in the the HS2 budget) and the immediate effect is that HS2 trains would have to run more slowly on the WCML at least than current trains.

    At the moment HS2 looks to me like a very expensive nineteenth century transport solution to a complex set of twenty-first century economic, environmental and social issues with the major benefits accruing to England and within England to the major conurbations which are already faring best.

  18. I’m glad we’re back up and running. I was worried that Anthony might have lost patience and just pulled the plug without warning.

    In case something of that sort does happen one day, what do folks think about setting up a back-up site? I’m a bit out of date but perhaps the modern equivalent of a Yahoo group or something?

  19. Not sure a backup site would achieve much that doesn’t already exist. By and large, we can count on AW maintaining a relatively non-partisan perspective of polling, while all those visiting tend to ‘interpret’ polls as they see fit.

    Who gets to host the other place? And what if said host has a particular stance?

    It would just get messy imho. I’d sooner offer AW some tech support than pick a new place.

  20. Horton

    I think you may have fallen into the trap of believing what the newspapers tell you. HS2 was always predicated on both speed of journey and freeing capacity on other lines. Sadly the papers chose to just focus on speed when it was first promoted, presumably they didn’t think the public would understand the capacity issue, or it wasn’t sexy enough to sell copy.

    The other myth is that tickets will be expensive. They can’t be. The only way to get a return on such a system is to have it as full as it can possibly be. You cannot charge high prices and fill trains, unless there is limited capacity. If HS2 isn’t built then such pricing will start on the existing Classic lines to reduce demand. In a way it already has started – ergo your £250 ticket.

    Compatible trains will run at current line speed on existing track to/from Glasgow, Edinburgh, Newcastle, Liverpool etc and once on HS2 they will increase speed to 250mph, the same as the Captive trains that run on HS2 only. What happens to high speed lines further north than the currently planned tracks is dependent on decisions by the Scottish Government.

    I occasionally used to use the ECML from York to Kings X and know how busy they used to get 15 years ago. I now occasionally use the WCML from Crewe to Euston and they’re just as busy. Taking all that combined express traffic down one route on modern purpose built high speed track, picking up ex Midland mainline traffic at Notts/Derby will free up capacity on all three mainlines for additional commuter and freight traffic, easing pressure on the road system too.

  21. @ Stageek

    The issue was that rumours are floating around that Anthony is unhappy with the “cesspit” here and had thoughts at least of pulling the plug on the comments.

    I did sort of miss it as part of my daily getting up routine (email/Guardian/ebay/torrent site for live recordings/footie forum/UKPR) and if not in a rush it was a good place to end up. I would miss it a lot during elections where you can pick up info much easier than on twitter where you need to be following so many accounts.

    So it’s not so much a case of needing an alternative as having a Plan B if AW did close the comments.

    A few of us bookmarked discord for election night if this went down:


    It wouldn’t be a proper replacement as basically live chat but it would be a starting point as a way of pointing to a forum that someone could set up if the plug was pulled. Only saw a couple of people on it while this site was down so it would be a nice place to bookmark for if Anthony did pack it in.

  22. @Shevii

    Appreciate it, but as I said at the time, Discord is a 3rd party app, and some people don’t use it. It was supposed to be for VoIP, or chat; not as a message board.

  23. Discord is at least somewhere to check in and decide what to do next if we do ever lose access to this site (which I very much hope we don’t – we appreciate you, AW!)

  24. @sda

    I’m not doubting that there are journey time and capacity benefits, the questions are:

    – Are they worth £80bn, £100bn or more?
    – Where and who benefits from such huge public expenditure?
    – What benefits would other transport investment bring?

    There doesn’t seem to be much doubt that the cost: benefit ratio of HS2 has and is deteriorating sharply as costs rise seemingly out of control nor that the major benefits will accrue to London and major conurbations in England and the relatively affluent.

    Incidentally, I read that HS2 trains will run more slowly than current pendolino rolling stock on the WCML.

  25. On HS2 I hope that messagers here are not forgetting the large conservation losses it will cause due to removal of ancient woodlands. That alone makes me against it, let alone that the travelling benefits will go mostly to parts of Southern England less in need than much of Northern England.

    And this promise of mass tree planting to compensate for losing the ancient woodlands is sheer meaningless spin.

    But not everybody is taken in by the Tory and tabloids` celebration of tree-planting programmes. Our local paper has the Aberdeen City rangers organising working parties to pull up young spruce saplings, on the grounds that they threaten red squirrels since will eventually oust Scots pines.

  26. since THEY will eventually oust Scots pines.


  27. Horton

    I couldn’t say if it would be worth £100bn, once you get into £bn’s it becomes sort of funny money.

    The trouble is, if you don’t build it, what would the costs and issues be without it? It’s the old chestnut that our existing lines are based on Victorian engineering. Swapping that out to update the current mainlines to get anywhere near the same capacity will take decades, cause as much disruption and probably cost the same. Doing nothing just means rationing by price and a gradually deteriorating service plus more and more road traffic.

    As to who/what benefits, well:
    MP’s travelling to and from their constituencies accessible from the line and it’s onward extensions.
    The rail industry
    Civil engineering
    Research and development
    Improved national skills base
    Those from London and hinterland travelling to:
    Birmingham and hinterland
    Crewe and hinterland
    Manchester Airport
    Manchester and hinterland
    Liverpool and hinterland
    Anywhere north of Warrington on the WCML and their hinterland
    Derby/Nottingham and hinterland
    Sheffield and hinterland
    Leeds and hinterland
    Anywhere north of York on the ECML and their hinterland
    Anywhere south of York on the ECML and their hinterland
    Anywhere south of Derby on the MML and their hinterland
    Anywhere south of Birmingham on the WCML and their hinterland
    And vice versa
    Air quality all the way along the route.

    All of those places will either be served directly or get more frequent Classic trains.

    As to Compatible trains being slower than current ones on the WCML, I haven’t heard that anywhere else and whilst the tilting Pendolino’s might have a cornering speed advantage they can’t match the Compatible’s straight line speed. I guess it might be possible that over the bendy bits of the WCML the Compatibles might have to limit their speed but they more than make up for it once on high speed track. These days it’s not the power of the engines that limit speed but the physics of keeping a train safely on the tracks around bends. If it’s an issue, then a Pendolino Glasgow to Warrington would be a couple of minutes faster than a Compatible, but the Compatible would be about an hour quicker into Euston.

  28. I must say I tend to favour HS2. A “shovel ready” project that will bring supply chain benefits throughout the country.

    I’d also favour simultaneously building the Northern Powerhouse HS line at the same time.

  29. There’s a good explanation in the article below about why it makes sense to have a separate network for High Speed trains.

    “Britain’s railways were largely built in the Victorian era, for a different kind of travel. Today, the same lines carry a mix of express intercity trains – the kind which HS2 will take – and stopping local and commuter services, the kind people use to get to work, or pop to a neighbouring town.
    This mix is a very inefficient way to run a railway, for a reason that is quite obvious if you think about it: trains cannot overtake each other on the same set of tracks. They would bang into the back of one another if they tried. Not good. To get around this, local stopping trains need a large gap behind them in the timetable, so the express trains behind them do not catch up. That reduces the number of trains you can have per hour on a line, dramatically reducing its capacity for every type of service – local and express.The engineering thinking behind HS2 is to take those express services off the older mainlines, leaving them for stopping local and commuter services. When trains are all travelling at roughly the same speed on a line, you can fit a lot more in, because the gaps needed between them are smaller.
    HS2 will take express trains off the West Coast Main Line that links London with Birmingham and the cities of the northwest; the Midland Main Line that links London with the East Midlands and Sheffield; and to an extent the East Coast Main Line that goes up to Leeds and Newcastle. That frees up capacity across a huge swathe of the country for local services, and it’s the whole rationale behind the project. That the government hasn’t been explaining this ad nauseum is inexcusable.”


  30. Not an expert of railways, but to my mind the problem with the capacity argument to justify HS2 is that you are choosing a hugely expensive option to boost capacity.

    I remain highly doubtful that the best economic and social developments will be secured from spending this amount of money on HS2, as opposed to spending a similar sum on conventional rail systems. Adding new conventional lines dedicated to slow freight would free up capacity on existing passenger lines, and presumably using @SDA’s logic re train speeds mean additional fast passenger capacity boosts on current lines as you move more slow freight off these lines.

    The key question is really what is it that we want to move around that is going to make the biggest impact outside London. My guess is that we shouldn’t be thinking so much about moving people, but more about moving information and goods, which means broadband and freight rail.

  31. JiB`s country can only mean England. I can see negligible benefits from HS2 coming to Scotland, Wales and NI, but its great cost will mean that rail improvements in those polities will be much curtailed.

  32. @ Alec

    Freight trains are already mostly routed onto lesser used lines rather than the most intensively used routes. There would therefore not be much to be gained by opting for a new ‘low speed’ freight network rather than a high speed passenger network.

  33. @Davwel

    For North Wales there are considerable benefits from HS2 and NPHS in terms of connecting with English Regions. There’s admittedly less benefit for South Wales.

    Surely there must be advantages for Scotland ( at least Glasgow and Edinburgh?).

  34. Shevii
    “A few of us bookmarked discord for election night if this went down:


    Thanks for that. Exactly what I had in mind in case there is some sort of disaster and there is no longer anywhere to argue about trains and electric cars etc. :)

1 55 56 57