2015 Result:
Conservative: 24400 (43.5%)
Labour: 14606 (26%)
Lib Dem: 10152 (18.1%)
Green: 1332 (2.4%)
UKIP: 5481 (9.8%)
TUSC: 178 (0.3%)
MAJORITY: 9794 (17.4%)

Category: Semi-marginal Conservative seat

Geography: Eastern, Hertfordshire. The whole of the Watford council area and part of the Three Rivers council area.

Main population centres: Watford, Abbots Langley.

Profile: Covers the borough of Watford and some of the commuter villages around it, including Abbots Langley and Langleybury. It is part of the London commuter belt, but is also an economic centre in its own right, housing the headquaters of several major companies such as JD Wetherspoon, Mothercare and Camelot.

Politics: Historically this was a classic Labour vs Conservative marginal, regularly swapping between the parties with the ebb and flow of party popularity. It was won by Labour in the 1997 landslide, with Claire Ward becoming the second youngest MP at the age of 24. Since then the Liberal Democrats have made strong advances in the seat, narrowly taking second place from the Conservatives in 2005 and making this a tight three way marginal by the time of the 2010 election. The Conservative victory here was in the face of a double handicap - not only did they start in third place behind the Liberal Democrats, but their original candidate for the election, Ian Oakley, was forced to stand down in 2008, and was later convicted of criminal damage and harrassing his Lib Dem opponent. Richard Harrington eventually won anyway, with a small majority over the Liberal Democrats in second. The collapse of the Liberal Democrats in 2015 left him with a far safer majority.

Current MP
RICHARD HARRINGTON (Conservative) Born 1957, Leeds. Educated at Leeds Grammar School and Oxford University. Former Managing Director then Chairman of a holiday resort development company. First elected as MP for Watford in 2010. Junior international development minister since 2015.
Past Results
Con: 19291 (35%)
Lab: 14750 (27%)
LDem: 17866 (32%)
BNP: 1217 (2%)
Oth: 2084 (4%)
MAJ: 1425 (3%)
Con: 14634 (30%)
Lab: 16575 (34%)
LDem: 15427 (31%)
GRN: 1466 (3%)
Oth: 1292 (3%)
MAJ: 1148 (2%)
Con: 15437 (33%)
Lab: 20992 (45%)
LDem: 8088 (17%)
GRN: 900 (2%)
Oth: 955 (2%)
MAJ: 5555 (12%)
Con: 19227 (35%)
Lab: 25019 (45%)
LDem: 9272 (17%)
Oth: 234 (0%)
MAJ: 5792 (10%)

2015 Candidates
RICHARD HARRINGTON (Conservative) See above.
MATT TURMAINE (Labour) Head of Client Services for BBC Worldwide. Watford councillor since 2012.
DOROTHY THORNHILL (Liberal Democrat) Born 1955, Tenby. Former teacher. Former Watford councillor. Elected mayor of Watford since 2002.
NICK LINCOLN (UKIP) Born 1969. Businessman.
AIDAN COTTRELL-BOYCE (Green) Born 1987, Liverpool. Educated at Bristol University.
Comments - 586 Responses on “Watford”
  1. What makes Watford different is that only a year ago she got 46% of the first preference votes in the Mayoral election, with Labour second.. This on the same day the Lib Dems got smashed in the Euro elections…

    As Chris says, many Tories evidently like her, and because of that vote (and the Mayoral votes before that) she deserves to be treated like a Lib Dem incumbent, rather better off than the 39% Julian Huppert got in Cambridge (for example). Of course there are some bits of the constituency that are in Three Rivers not Watford, but I am sure “Mayor Dorothy” has good name recognition there as well, and the Lib Dems hold most of the council seats.

    But what makes this so unpredictable is that there is also a Tory incumbent. The last Ashcroft poll way back in November had Thornhill a close second to Harrington, and if there is going to be a Lib Dem gain anywhere it will be here.

  2. Without saying it in these words, many are dismissing talk of the LDs here as fantasy politics, and that therefore this is not a particularly interesting seat to talk about.

    So let me play devil’s advocate here. Let’s assume the LDs are vastly overrated and that this seat will behave “normally”.

    In that scenario Watford would presumably revert to a classic Con/Lab marginal, albeit we have relatively weak evidence about the exact support levels of the parties due to the LDs breaking 30% in 2005 and 2010. Given their local base, let’s say the LDs get roughly 17% of the vote in this “normal” seat in 2015 (1992/1997/2001 levels).

    If this seat is “normal”, then we should also be looking at Con to Lab UNS. Nationally Labour and the Tories are neck-and-neck, which represents a swing of 3-4%. It will be a bit higher in England due to the Lab to Con swing in Scotland being considerable (a misleading phrase in any other context, but appropriate when talking about the effect Scotland will have on UK Lab and Con share). That would suggest an English swing of well over 5%. To counter that we should allow for Harrington’s first-time incumbency boost. If this is a “normal” seat, then by extension the bookies are giving absurdly generous odds for Labour in what statistically should be a “normal” ultra-marginal Con/Lab contest on Friday.

    Let’s also explore the argument that Harrington will double or treble his majority because – “as normal” – the LDs are overhyped and/or overrated by Ashcroft. The LD share in Ashcroft polling is huge by “normal” LD standards, as was Thornhill’s mayoral vote relative to LD performance at council level on the same day. Some feel both sets of figures are grossly inflated with regard to how it will translate into a Westminster ballot box. Understandable. But if LD VI here is grossly inflated, then can we trust the figures for any party in Ashcroft? If not then what evidence do we have that Harrington is actually ahead? Surely we are left with national polling (which suggests the sort of swing that puts Labour well within range here) and bookies’ odds (which prior to Ashcroft polling had Labour and Conservative neck and neck). Even if we can say that before the LDs disperse Ashcroft had the big picture right, the changed VI from ex-LDs is bound to have a noticeable effect on the subsequent balance between the big two.

    Devil’s advocate session over. Apart from my sarcastic use of “normal”, I think I’ve been pretty fair.

    48 hours before we should get the result here – though it will probably be later. Plenty of time for anyone who tips Harrington to increase his majority to explain why, in the face of what looks to be a 5% or so swing to Labour in England. I hope that there was more to those tips than “I want him to win”, “he works hard”, or “1992, 1992, 1992, 1992”.

  3. Chris,

    Yes, Labour need a 4% swing to take Watford, so if it was a “normal” seat, it would be just in the top 50 LAB-CON marginals. The swing from LAB to CON in 2010 was 6%, just a little bit more than UNS. There is no particular evidence that the Lib Dem vote comes from one side or the other. I presume they did their best to squeeze down the Tory vote in 2010 but it obviously did not work!

    This time I see no reason why the Tory vote should increase – incumbency should be more than balanced by the national swing to Labour.

    Comparing the Euro and mayoral elections for Watford borough (which excludes the most Tory parts of the seat), Lib Dems were up 30%, Labour stayed still on 26%, UKIP were down 9%, Tories down 7% and 13% of greens and others did not stand. I am sure that quite a lot of the UKIP votes went to Labour as well as the Tories and direct to LIb Dems, and one would think so did Green votes. So in all probability Thornhill picked up significant votes from both Labour and Tory.

    Given that the Labour vote in Three Rivers in the Euro election was only 13% (almost exactly the same as the Lib Dems) it seems certain that they polled less than 25% in the constituency

    To be honest, I find it hard to imagine a stronger Lib Dem challenger for any seat than a Mayor for that seat elected four times with huge majorities. The Ashcroft poll in November showing her within MoE of the Tories shows she has already overcome the national swing away from the Lib Dems. So without any local knowledge of Watford other than what Chris has kindly provided, I will predict a narrow Lib Dem win here, with Labour probably a little below their 2010 percentage (similar to the Euro vote, see above). If that happens it will be the only swing from Labour to Lib Dem in England, although in reality a swing much more to “MayorDorothy” than to the Lib Dems.

    If I am correct I will be back on this thread on Friday… If not then of course I will be quiet!

  4. You two could give Ronnie Corbett a run for his money in terms of long winded waffle. Have some common sense for heavens sake. If mayoral elections were a good guide to general election voting then the Tories would be just about to sweep London and Hartlepool would have voted for a monkey in 2005. Watford is a nondescript town just to the north of London full of Mr and Mrs Average. Of course the national picture cannot be disregarded and the Lib Dem unpopularity will affect them to certain extent here just like everywhere else.

  5. Big Mouth Strikes Again.

    Thornhill does have a personal vote- I know this from three Labour supporters J have known for 30 years who are switching LAB-LD.

  6. I predict that deeply concerning national issues on who runs the economy etc. will edge it for Conservatives in spite of some recovery for LDs at a national level since the last Watford poll, but still don’t rule out a LD win completely here given the exceptional situation.

    I really do still think a LD win here is more likely than Labour, although again I don’t rule out LD coming third.

    A rare 3-way marginal between these parties that still exists.

  7. Yes, me too.

    CON HOLD but just from Thornhill – although the Racing Post today picked Labour to win here at 4/1 as one of the 8 best bets of the GE.

    The reason they give is that …”Turmaine has been busily raising his profile in the last 6 months.”

  8. H.Hemmelig

    If I thought Mayoral elections were a “good guide” I would have DT on 46%!

    And I happen to think Boris would have won any Lab-CON marginal in London, if he had had the courage to stand in one…

  9. 4/1 is an outstanding price if you take the HH view of how the LDs will do here.

    If the LDs are irrelevance in the race to win, then surely in an urban seat where Labour start 8 points behind an average first time incumbent Conservative, the outcome should be very close?

  10. 4/1 is an outstanding price if you take the HH view of how the LDs will do here.

    If the LDs are irrelevant in the race to win, then surely in an urban seat where Labour start 8 points behind an average first time incumbent Conservative, the outcome should be very close?

  11. I have a feeling LD will shade this by approx 1000 votes.

  12. LD Gain. 1,200 maj.

  13. Full Result:

    Con 24,400 43.5%
    Lab 14,606 26.0%
    LDem 10,152 18.1%
    UKIP 5,481 9.8%
    Green 1,332 2.4%
    Other 178 0.3%

    Majority 9,794.

    I think we need a debrief from Mr Hornet (I tipped Dorothy Thornhill to win based on his postings!).

  14. I did keep saying that the LDs had no chance of winning.

  15. I Predicted CON HOLD – one of my, now famed “CON HOLDS” list posted late April on the LABOUR TARGETS thread.

    Congratulations to Mr Harrington – you have to say that the majority is impressive.

  16. Well done Andy for sticking to your guns like that. Despite the nonsense here Thornhill didn’t do an awful lot, did she?

  17. To be fair, hardly anyone on here predicted a Lib Dem win (I said Harrington would win). As predicted, Labour failed to make big inroads, but where I and just about everyone locally was way off was in thinking the Lib Dems would be more competitive. The combo of national factors and the local stuff previously mentioned (Farm Terrace Allotments etc) did for them big style.

    (N.b. Anthony’s main article on this website yesterday said Watford was a three way marginal, so this result was a big surprise to most interested people).

  18. Well, I will eat big humble pie here! I thought Dorothy Thornhill would be much more competitive, especially with Ashcroft polls last year showing her very much in contention. Shows that mayoral votes really are very different from General Elections..

    In the end though she has suffered the swing that afflicted almost all the Lib Dem MP’s, so hardly surprising in that context

  19. Horrendous LD result – but I didn’t think they would win.

  20. t would be interesting if someone (now who could I be thinking of) could make a constituency by constituency comparison between 1992 and 2015.

    Its made complicated by two sets of boundary changes.

    Conservative losses since 1992
    At least a dozen in London, many on enormous swings eg Croydon North which has a Labour majority of over 20,000.
    Ten in Scotland
    Middle class areas in conurbations – Hallam, Leeds NW, Leeds NE, Edgbaston, Coventry S, Wolverhampton SW, Tynemouth, Bury S, Bristol W, Hove, Brighton Pavilion
    Middle class ‘Greater Scouseland’ – Wirral W, Wirral S, Chester, Sefton C, Southport
    Carshalton, Norfolk N and Westmoreland to strong LibDems

    Conservative gains since 1992
    Various from the LibDems – Bath, Cheltenham, Cornwall N, Truro, Devon N, Yeovil, Brecon, Montgomershire, Berwick
    Middle England industrial areas – Nuneaton, Warwickshire N, Cannock, Sherwood, Morley, Crewe, Pendle, Rossendale, Thurrock
    Three gains in SW Wales

    The Conservatives would also have benefited from boundary reviews creating more rural constituencies.

    There’s a few odd constituencies – Lancaster, Chorley, Bolton NW, Batley, Gedling for example which might have other demographic and candidate factors.

  21. Richard, also need to add Middlesbrough South & E Cleveland (although the C results were ok ish in 2010/15).
    A loss since 1992.

  22. I hate to say it…but wasn’t this supposed to be a surprise Lib Dem gain (as well as all the others)?

  23. A daunting task for Labour: they need to overturn a Tory majority of 9,794 votes here in Watford to win an overall majority in 2020 on uniform swing, and that’s before boundary changes.

  24. Labour did very well to gain second place here considering their overall performance outside the major cities. This gives them a chance here for 2020 and 2025. Very dissapointing result for the Lib Dems here considering Thornhill is the Mayor and is popular locally and the success of the Lib Dems during last years local elections here.

  25. From a starting position of almost 10,000 votes less.

  26. Very strong result for Richard Harrington here – showing that liberal Conservatives can build up robust majorities in what should be marginal seats

    Its better than any of the majorities Labour managed during the Blair years

  27. Even Dorothy Thornhill as Lib Dem candidate amounted to nada,in fact taking them backwards though it’s likely due to the national picture. 8% lower than 2nd place Labour.

  28. Just had a read of some of the predictions on this thread – only just got to this one although I did comment just after.
    Hilarious LD ramping.

    Excellent result for Richard Harrington.

  29. I can’t understand why anyone thought the LDs could win Watford, Maidstone or Oxford West when it was clear they were going to lose about two-thirds of their 2010 support. You can’t buck the national trend to that extent.

  30. Andy I had the same argument with posters on here about that over and over again, and basically the rebuttal to this rather obvious fact was that national polling didn’t apply to the LDs in any seats they held or were targeting.

  31. Yes and the extent of that particular delusion was remarkable – it extended beyond the usual Lib Dem rampers to include others as well.

    What was equally extraordinary was the widespread dismissal of the considerable body of evidence that pointed to the Lib Dems doing very badly.

  32. That’s right – the Lib Dems always claim to hold up better in particular areas. You can’t do that when the fall in popular vote is so severe although it can be achieved in less severe low points like 1979 and 1992.

    I was taken back though how badly they did in some areas where they have survived at local level – Easteigh, Sutton and Cheam, possibly Cheltenham although I had a feeling T & Yate and some others would unravel.

    We now need to hit them even harder – and encourage Greens and UKIP to provide the non two party vote.

  33. Polls showed them to be well ahead in one or 2 of these seats, but clearly when they came to vote some people wanted to vote against a Labour government, and saw the Conservatives as the only worthwhile vote for that purpose.

  34. (Most of) the polls were quite simply wrong.

    But of course a lot of people wanted to believe them. This is understandable, especially in the partisan heat of a GE campaign when the committed (either to a party or to a view) tend to grasp hold of any evidence that points ‘the right way’.

  35. Worth noting that ICM’s poll of Sheffield Hallam (Clegg 42, Coppard 35) was relatively close to the actual result (Clegg 40, Coppard 36). Perhaps their approach of naming candidates is somewhat effective in constituency polling, particularly close to an election where candidates have been declared and have been campaigning.

  36. ‘But of course a lot of people wanted to believe them. This is understandable, especially in the partisan heat of a GE campaign’

    To suggest that people believed the polls because they ‘wanted to’ is a ridiculous statement – there was no compelling reason to disbelieve them, and they were pretty much our only evidence.

    And for people going through some seats saying ‘lol look at the hilarious ramping from the LDs, why were people taken in by this?’ could just as easily go through loads of Labour target seats where that party failed. The fact is that the polls were wrong, people looking at them were not ramping, they were simply using evidence that turned out to be flawed.

  37. Meh when it came to the LDs if there was a poll showing success then that was gospel, otherwise it was left to savvy analysis which emphasised how dug in Lib Dems got at local level, how respected their MPs were, and how even though the polls didn’t really show it, Labour voters would still tactically vote for them on the day.

    As a result we had polls that backed up LD holds, and ‘in the know predictions’ that we might have shock holds in places like St Ives or Mid Dorset and landslide gains in Maidstone or OxWAb

  38. You’re exaggerating, most of you. I didn’t want LDs to win here, but thought they had a chance – I certainly didn’t think this of OxWAb or Maidstone – though nobody predicted ‘landslides’ you naughty boy.

    In the event, Thornhill’s 18% down from 32% was a good result compared to most seats where they had c.30% in 2010 – which is an awful inditement of how they did nationally overall.

  39. Robberbutton, please stop being deliberately dense.

    St Ives is actually a good example – you deride people predicting a Lib Dem hold there, but it saw a decent (by the standards of the night) performance, and remains the number 8 target seat – so would have been seat 16, a number very few people expected the party to fall below.

    The Lib Dems did not actually do significantly worse than expected – a little worse, yes, but not hugely. The main difference was that the Tories did much better, so while we expected the LDs to hold seats as the Tory vote fell, allowing incumbents to hold on with a lower or only slightly higher share than their challenger in 2010, the increase in the Tory vote meant the Lib Dems fell like ninepins.

  40. The way the Ashcroft polls were treated was quite instructive I think. Q1 was disregarded, Q2 given all the attention. Why? This was an untested methodology.

    The answer is of course that Q2 chimed perfectly with all the mythical Lib Dem arguments about incumbency, local strength etc etc. This is just confirmation bias. On the same basis, the Comres poll of SW seats which showed the Lib Dems doing very badly was dismissed outright.

    As for the idea that the polls were the only evidence available – what rot that is. Just looking at the itineraries of the party big guns (especially the Tories) gave you a big hint as to where the real front line was – why bother going to North Warks, Yeovil etc.? And there were other local pointers too – but we should never underestimate the capacity of people to ignore news they don’t like.

  41. ComRes on today’s BBC Daily Politics said differential turnout/’lazy Labour’ amongst the young, poor etc was more a factor as to why the polls were wrong than ‘shy Tories.’

    Keller for YouGov said yesterday they’d examining 7 factors; but, in typical fashion said he/they got it right 6 times – just not this time or 1992.

  42. Frankly I would not take any of these pollsters’ arguments too seriously, at this stage at least.

    Getting things this badly wrong is potentially devastating for their business and they will be throwing as much chaff up as possible while hoping, eventually, to find some convincing explanation…

  43. ‘Just looking at the itineraries of the party big guns (especially the Tories) gave you a big hint as to where the real front line was – why bother going to North Warks, Yeovil’

    A little confirmation bias of your own there – Clegg visited Maidstone!

  44. Well if we are charitable we can put that down to him just trying to keep his beleaguered side’s spirits up.

    Less charitably, perhaps he was dim enough to believe his own side’s ramping as so many Lib Dems on here were.

    The key point is the Tory itineraries were based on sound intelligence. We all know that now – some of us had a hint of it before.

  45. “Robberbutton, please stop being deliberately dense”

    Well Iain, I’ll admit that I mustn’t have the same razor sharp intellect as yourself, then I might have been able to have the clued in knowledge to describe NE Fife as a safe LD seat, or confidently predict 10 Liberal Democrat seats in South West England.

    I might even be able to understand how “The Lib Dems did not actually do significantly worse than expected”

  46. Indeed. I seem to recall a particular poster confidently stating how much better informed he and others on this site were than the general public. And yet…

  47. I have already held my hands up on my way-out predictions on this one but I do think the “holier than thou” attitude of a couple of people who decided to disbelieve the Ashcroft polls is really ridiculous.

    It was not the Lib Dems who particularly believed the Ashcroft poll CVI – it was first and foremost that well know Lib Dem Lord Ashcroft (who only ever gave CVI in his headline data), followed by pretty much every so-called expert commentator.

    Remember the Ashcroft polls were spectacularly wrong in many LAB-CON marginals as well, where both SVI and CVI were wrong. Virtually every forecaster predicted between 20 and 30 Lib Dem seats, and in that world Thornhill would have come much closer.

    On a site which is supposed to be about polling perhaps it would be more appropriate to try and work out why Ashcroft’s huge investment in constituency polling went so spectacularly wrong, rather than continuing to congratulate each other on the victory of gut-feeling over actual polling evidence almost a month after the event!

  48. Andrew, I’m really gutted too.
    It really hurts.
    To be honest, in 2010, if Sal Brinton had run a local dentist and welded everyones teeth together I thought she would still have been elected.

    Likewise in 2015, I felt the Lib Dems would have had to fall down the stairs and smash all the furniture not to get this one.

    The Lib Dems will be polling 40-45% by 2020 and just able to get a majority with stupendous momentum.

  49. Not that she did – I said if hypotheticaly she had, I thought Sal would still have won.

  50. Slightly unexpected to see the UKIP and LD candidates giving each other a quick hug and kiss after the result was announced:


Leave a Reply

NB: Before commenting please make sure you are familiar with the Comments Policy. UKPollingReport is a site for non-partisan discussion of polls.

You are not currently logged into UKPollingReport. Registration is not compulsory, but is strongly encouraged. Either login here, or register here (commenters who have previously registered on the Constituency Guide section of the site *should* be able to use their existing login)