UK General Election Prediction Thread

[Oryxslayer]

So, lets engage in a little though experiment that I ran last night. Lets take the 2017 election results (a 'normal' party breakdown) and the 2019 elections results (a 'brexit' focused breakdown) by consistency into a spreadsheet. On one  side we will have the Tories, UKIP, and Brexit - this will be called the Leave block. On the  other side we will have Labour, Lib-Dems, SNP, Greens, PC, and CHUK - this will be referred to as the  opposition. Add up each of their vote shares in each constituency for each election. Ignore NI because there are no 2019 numbers for that and it's elections are weird. Lets be sure to apply some weight to the opposition numbers, say a multiplier of .85 on their result in each seat to account for the naturally inefficient vote splitting between SNP/Lab/Remain forces. Now, lets weight each result (2017 and 2019) by 50% and add them together to see what the seat breakdown is.

We get a total Leave Block seat count of around 345-350. This thought experiment does not predict seat outcomes, but I think it is a good indicator of a overall total. Boris has been chasing the metaphorical dragon of the leave-labour seats and even if he gets just a bit it will be more than enough to counter losses in Scotland, London, the London bedroom communities, and various urban regions throughout the country.

Quote from the other thread.

[Oryxslayer]

Almost nothing about this election is predictable at this stage? We have very little idea of what it will look like in any sense; other than that it won't look like the last one.

Not even sure we can count on that. Soul crushing though it may be to contemplate, a near identical seat split (315/260/35/15/other) is very much possible.

Something similar to the current allotment is certainly possible, but the geographic distribution (which I suspect is what he was referring to) would be far different. For example, I would be shocked even at this stage if the Tories don't lose St. Albans and Watford, cracking their stranglehold of Herfordshire, adding on the previously cracked Surrey.

[Oryxslayer]

An overlooked factor is a possible 3 seat pickup by SDLP in NI, in
- Foyle (very likely)
- Down South (likely, if 3 point swing from Sinn Féin)
- Belfast South (possible, if 2 point swing swing from DUP)

Numbers here.

Contingent of Sinn Féin & DUP losing ground after their high water mark of 2017, these 3 seats would be very pro-Remain and necessitate a higher Tory threshhold.

Don't forget Belfast North. Sinn Féin has gained ground in that seat every election for 2 decades.

Also, seeing Nigel Dodds lose his seat would be Greek drama levels of irony.

Northern Ireland is going to be a proper mess this year...but when is it not? Here's an analysis from a northern Irishman from another forum:

Anyway, as regards Northern Ireland one thing I'm quite interested in is how the overall vote goes. In 2017 the DUP and Sinn Fein surged well out in front - Sinn Fein finished on 238,915 (+62,683) and the DUP on 292,316 (+108,056). The SDLP mostly stood still on 95,419 (-4,390), Alliance likewise on 64,553 (+2,997), while the UUP's dropped like a stone to 83,280 (-31,655).

Both the DUP and Sinn Fein ran excellent campaigns that time around. Sinn Fein were able to take advantage of the recent collapse of the Assembly, the DUP's blocking of an Irish Language Act, the simplicity of the "Equality, Respect, Integrity" message, anger about Brexit and the sea change that was their phenomenal performance in the Assembly Election three months prior to ratchet up nationalist turnout in a way that hadn't been seen in some years. Sinn Fein's success was also the DUP's greatest campaigning tool. Sinn Fein's success wasn't just a political rival doing well, it was an existential threat. In March 2017 Sinn Fein reduced the gap between them and the DUP from the ten seats and and forty thousand votes it had been a year before to one seat and one thousand first preferences. It gave confidence to nationalism, but also sparked a rally round the flag effect in unionism that the DUP took advantage of to ruinous effect. While Sinn Fein's voters had previously stayed at home on election day the DUP's were swept up from across the board - some were from new turnout, some from minor unionist parties (between them in 2015 UKIP and the TUV got some thirty five thousand votes - in 2017 the former stood no candidates, the latter only one, who got three thousand), and a solid chunk from the UUP, who were themselves coming off a wretched Assembly election that had resulted in their leader stepping down.

Now things are a bit less rosy for them. Northern Ireland has been without a government for over one thousand days due primarily to the fact that these two parties can't get around a table and agree something between the two of them. For the DUP there are a litany of other problems - dodgy dealings like Wee Ian's Sri Lankan getaways, the drip drip of RHI (and Sam McBride's comprehensive retelling of it in Burned, which has sold like hotcakes), the fact that most of the money they crowed about from their deal with the Tories hasn't actually appeared, the Brexit related bind in which they find themselves, and the fact that Sinn Fein themselves aren't so scary. The momentum they had in 2017 has faded a little. Gerry Adams and his Radical Republican Agenda are gone, replaced by Mary Lou and Michelle who are both just a bit crap really. They're also coming off a dispiriting electoral performance and poor polling down south, and their abstentionist policy is coming under renewed pressure after two years where nationalism has abdicated representation of Northern Ireland in London to the DUP, and when seven Sinn Fein MPs could have made more of a difference than any nationalists in the House of Commons in the last century. That puts them in a bind, but it also makes them a less effective bogeyman for the DUP to use to rile up their own base.

The last couple of elections we've had locally have indicated (with the usual provisos about how it's a mug's game extrapolating local/European results to a general election) that there isn't quite so much momentum behind either party as there was in 2017. In the locals in May the DUP did see their first preference vote increase overall (144,928 to 163,615 first preferences) but that was fuelled almost entirely by the collapse of smaller parties (the PUP, TUV and UKIP got twenty councillors and fifty thousand votes in 2014, that fell to nine and twenty three thousand this time around). Sinn Fein meanwhile saw a slight dip in their first preferences overall (six thousand or so, give or take), which'll be a concern given they'd have been hoping for a decent increase if any of that 2017 momentum was sticking around. Both parties also saw their vote drop in European elections in June - the DUP's by six thousand to 124,991, Sinn Fein's by 33,000 to 126,951 - both due to falling turnout and the astonishing Alliance result (more on that later).

As for the other parties the SDLP have stood up reasonably well. In 2017 as noted above they stood still, and actually saw their vote go up in two of the three seats they held (Foyle and Belfast South - South Down was the odd one out but Ritchie was still only a couple of hundred votes off her 2015 total). There have been slight drops in their vote (in the Euros it went from 81,594 to 78,589, the locals 85,237 to 81,419). I'd expect that to stay broadly the same this time around, perhaps a slight uptick with a good campaign.

The UUP will be hoping to undo the damage of the past few years with Aiken's new strategy. The European elections were a particular blow, where they lost a seat they've held since the European Parliament's inception to Alliance and their vote collapsed from 83,438 to 53,052. They held up reasonably well on a local level (101,385 to 95,320), but that was oveshadowed by the loss of 13 seats. Their new confrontational policy sees them actually properly coming out against the DUP for the first time in a while, a sharp contrast to 2017 (where they stood aside for the DUP in Belfast North, Belfast West, North Down and Foyle). Actually standing in all eighteen seats will at least give them a few more votes, but it remains to be seen whether their brave stand will be able to undo any of the damage of the last few elections, or indeed whether it'll make things worse (the DUP are already trying to attacking them for helping to surrender "Edward Carson's Seat" in Belfast North to Sinn Fein).

Alliance are the proper wild card here. The last two elections have seen stratospheric increases in the parties vote. The locals in May saw it jump from 41,769 first preferences to 78,052, while the Euros a month later saw it leap from
44,432 to a phenomenal 105,928. If any of the parties could be said to have the momentum with it at the moment it's Alliance, and most of that is focused upon Belfast East and South. However Alliance's vote increases in March weren't restricted to those two seats, and if the Yellow Surge is replicated in December then Alliance could be the spanner in the works for a few races where they've not really figured before - not necessarily to win obviously, but increasing Alliance votes could throw off predictions in a few places (Belfast North and South Antrim are the two ones to immediately come to mind).

The smaller parties aren't going to feature much. UKIP is dead, the TUV seem to have given up on Westminster contests going off last time (though Jim Allister may make another tilt at Ian Og's seat), the PUP aren't a factor, PBP will be second in West Belfast but not likely to be anywhere elsewhere (though a decent PBP run in Foyle could have an effect on the Sinn Fein-SDLP contest), and while the Greens have had a strong local election that wasn't matched by much of a change in their European result, and even in the seats where they do have a strong presence (Belfast East, Belfast South, North Down) it isn't enough of one to do much more than keep a few deposits.

[Oryxslayer]

Oryxslayer, could you tell me where you found that interesting quote?

Glad I'm not alone in thinking NI could play a big role in 2019, like it did in 2017.
Isn't it curious that such a small, educated, wind-swept corner of the world, with virtually no natural resources or immigration, would elicit such dark passions and hatred?

SDLP winning in South Down makes would change the margin of power in Westminster by only 1, while SDLP flipping Belfast South would change it by 2.
It's only a mess in the sense that political parties reflect more than transactional choices in NI, they become part of our identity.

I'm considering registering to vote in the North, as I have family in Foyle. Not sure it's worth the (considerable) effort of claiming residence.

General election thread on the alternate history forum.

He's also provided a subsequent analysis on some of the more minors who are standing which can be found here.

If you cannot see the links, then it's because it's on one of the boards blocked to those without an account, I forget which ones have that lock.

Also in a unrelated note, the DUP are standing down in Fermanagh and S Tyrone for the UUP candidate.

[Oryxslayer]

You look foolish to so openly mix predictions and desires.

Were six weeks out from the  poll and the campaigns have not even started. If you are making a prediction now, some chunk of any prediction will be based on hope and desire.

[Oryxslayer]

You look foolish to so openly mix predictions and desires.

Were six weeks out from the  poll and the campaigns have not even started. If you are making a prediction now, some chunk of any prediction will be based on hope and desire.

The thread is a prediction thread.

I think  the Political Betting Blog reference gives a reasonable basis for the hope.

I think Farage  could blow things up.  It is duty of his party to cqontrol him.

I'm not arguing with you, I'm just pointing out that any prediction this far out has very little data regarding the situation in 6 weeks. So there will always be a bit of hope. For comparisons sake, the Spanish election is in one week, so polls are fairly accurate.

Like for example, farage today announced he won't stand for a seat. That throws a wrench in a prediction based purely on yesterday's data, no matter what that prediction is.

[Oryxslayer]

snip

I'll throw my two cents  onto the  pile because London, despite her vote splits, seems to be rather easy to call seats for this area in particular. I agree on mostly everything, with one exception and one necessary mention. I general it 'feels' like both with polling subsamples and with the London breakdown that the Tory vote is going to hold up better in the leave'y bits but their vote in what I will call 'the slice' drops thanks to Brexit. There  is a reason why all the remainer tories and new LDs are fleeing to there, its the region that best suits their electorate.

From here on out I will refer to a GIS snip from a 'brexit-weighted regional swing model' I ran last week. The colors shown do not reflect my opinions.



Most of the tossups that you mentioned are tossups here with flips like Enfield Southgate on the map being rather marginal. So lets just ignore that part and go to may main point: 'the slice.' When looking at london one needs to be inquisitive where exactly is the LD surge coming from and most likely to crest. The two obvious places to eliminate are all the visible minority labour seats and the brexity tory seats. We should therefore  be looking at the  white parts of London - which have historically correlated well with Tory voting. It should come at no surprise that 'the slice' is very white. These places are all leafy-wealthy-semi-suburbs/neighborhoods  that contrast oh so starkly with the poverty and high rents of Labours side of london. Then you have the city itself which is a financial district on top of the high wealth of Westminster and the rest of 'the slice.' While I hate the population distortions, the map below nicely highlights how the wealthy parts of London stand out with the rest of the city.



Oh, and these areas all went hard for Remain. It seems clear that 'the slice' is where the LD surge is most likely to manifest - even if the LDs end up falling back to 10% in the polls. Now, I don't agree with everything my swing model says goes orange, I made the  model just out of curiosity after all. However I think the following seats are all within the potential splash zone as Tory voters become LDs at a near 1:1 ratio: Battersea, Cities of London and Westminster, Chelsea and Fulham, Kensington, Putney, Richmond Park, and Wimbledon. Bermondsey and Old Southwark, Hammersmith, Sutton and Cheam, and Vauxhall all have low potential to join the list, but the LDs 'all-or-nothing' style of resource investment probably leave them by the wayside. Theres a reason who Chuka moved out of the minority constituencies and into the rich remainer strip. Its also why Esher and Walton can't be seriously written off either, since it's just the same region only extending beyond the greater city limits. Honestly, these constituencies remind me a lot of the old GOP slices into West Houston and North Dallas: leafy, wealthy, semi-suburbs/neighborhoods who now are all represented by democrats.

However, the LDs are also eating into labour in London, so lets get to the necessary mention: Barnet. The income and racial map of London shows that the Enfield/Barnet region is mostly where I feel these votes will be coming from - along with 2017 labour votes within 'the slice.' You had every Barnet seat as a tossup, and I agree, because this is Corbyn's great thorn in his side. The region like every other in London should be becoming more  accustomed to not voting for Tories: it voted for Remain, and has a sizable labour presence initially. The problem of course is the Jewish vote. Corbyn's labour's unpopularity is most acute with this group, who has a sizable pull in Barnet. There is a reason why Labour slid backwards here in the last local elections. A community that is steadfast in the 'never-corbyn' camp and not satisfied with going back to blue is an opening for the LDs. Considering there is a good number of wealthy whites in the region initially who will be flipping for the LDs regardless of the Jewish vote, one can see why the LDs have put the three seats on their target list. Of course we won't know how this region feels unless there s a polls specifically of the council district, and swing models won't pick up the LDs pouring resources into the region. So tossups all around are appropriate.

[Oryxslayer]

snip

No longer going to make a  big comment because once you leave london it gets harder for a yank to analyze seat-by-seat results. So I make one comment: there are a lot of places in the SE and SW where the LD floor is going to be high, but not high enough to pass the Tories. If the LDs have a moment, or surge (which is a big if) look for it in Surrey and neighboring seats to the north, Devon/Cornwall, and Hertfordshire. May could very well end up with a Portillo moment on her hands if this very unlikely situation occurs. Similarly, if the Tories retain their big lead or have a Leave surge, keep an eye on North Wales, the Non-Leeds York region in general, the Tees Valley seats, and maybe even a Sunderland seat, which would be fun come election night. Labour lacks a surge zone unless they take the lead: their hope is to just push back and hold what was won last time.

[Oryxslayer]

I'm not sure why you are sleeping on Colchester. That model I referenced has the seat going Orange marginally (so tossup) and a separate model a ran two weeks ago to test the oppositions baseline has the seat flipping to someone not-tory. The Lib Dems retain strength locally, held the seat before the coalition, and the seat is only 51-49 for leave, so the oranges still have a healthy base to pull from for their potential plurality. Its also one  of the  many seats in the greater SW/London region that saw a near 1:1 Lib-Dem to labour transfer last cycle since they were the viable remainer ticket. Maybe there is something candidate-wise that just pure statistics cannot pick up.

Also the Cambridge seats are going just be...weird. We had a constituency poll (obvious MOE implied) a while back that showed Cambirdge itself was a marginal Labour lead, and today we have one showing SE Cambridgeshire is a 10% Tory lead over the Libs. I wouldn't therefore be surprised if the  more favorable S Cambridgeshire is a close race between team orange and team blue.

[Oryxslayer]

Right now, with a good idea of who is standing where I think it's likely to be 8 DUP, 5 SINN, 1 SDLP, 1 Alliance, and 3 . SDLP are putting all their eggs into the Foyle and Belfast South baskets. Foyle is a probable pickup since their leader is going there. The DUP get Hermon's seat for the reasons alluded to, but its probably to be an Alliance/DUP fight not UUP/DUP. Good number of potential voters for the Alliance near Belfast now that Hermons gone. Long's back in Westminster since she's in a 1v1 header with the DUP and Alliances fortunes are on the  upswing.

Now, both the DUP and Sinn are down from their 2017 high because NI has been govt-less for years since they cannot compromise. Sinn have  leadership troubles and the DUP have issues related to their support from Westminster. The Alliance are the  only NI party with any sort of 'momentum' right now, increasing support in both the locals and the EU. However, the Alliance lacks targets and is more likely to play a disrupter role. With this in mind, here are the tossups:

North Belfast: Pure DUP/Sinn fight, the DUP have been advantageous with their pacts here.
South Belfast: Alliance surge is a problem here for the traditional contest between SDLP and DUP, and it's a real three way fight.
Fem & S Tyrone: Only chance of UUP breakthrough, its a seat where any Sinn turnout drop or vote split gives away the seat.


FTR, very little  of these are  my takes. I have just as much knowledge of NI politics as you, so I follow an actual residents on-the-ground and up-to date takes. They can be found here, here, and here - though you may need to create an account on this forum to view them, I cannot remember. If so I will repost here, there is a general timeline to them though.

[Oryxslayer]

Don't believe the Alliance hype in NI, as with their mainland cousins the European elections overstate the real strength of their position. They are almost definitely *not* going to win Belfast S or indeed get close.

(Belfast E and North Down are where they should be concentrating their efforts tbh)

I'm not going to defend my post and instead suggest you actually read the justifications from an actual NI'er. I've actually reposted the first one and his justification for the Alliance before - it can be found on page 2. It is the first take  in the timline though, now 10 days old and candidate nominations have made some redundant. Essentially, the EU elections are better data points in NI then they are anywhere else because sectarian voting doesn't stop for Brussels. He's also not comparing vote percentages, which can move with turnout and do make Euro elections useless 90% of the time, but total voters which are not really changing in NI.

[Oryxslayer]

Doesn't 10 Scotland seats for the Tories seems pretty optimistic for them? I was under the impression that Boris is incredibly unpopular in Scotland, close to what Thatcher was.

There are multiple opinions on this. First, if the tories get tied to brexit hard and have a poor Scottish campaign then they are going down to four seats that should be held barring retirement. The other option is if being for/against IndyRef2 becomes the sole focus of the Scottish campaigns, then Tory incumbents are naturally going to hold up better than swing models say. The unionist vote in Scotland has demonstrated far greater ease with voting tactically, and an incumbent is an easy target to rally around. In both scenarios unionist tactical voting will also benefit other parties.

[Oryxslayer]

Is there anyone who isn’t an overly optimistic lefty that doesn’t think this is a Tory win?

Things can change in a month...but if the election were held today BoJo waltz's back into Number 10.What everyone knows though is that these numbers likely won't be final, a month is a long time. This election uniquely has a lot of potential energy trapped, that can push the election in a thousand and one different directions when it erupts. Which direction is what everyone's scared and arguing about.

[Oryxslayer]

Also uncertain how much is Lib-Dem collapse and more just a Lib-Dem realization that their going to be a targeted party once again, so their voters in untargeted constituencies are just heading to where they would always go of the Libs are uncompetitive.

[Oryxslayer]

Why is this only ever true of supposed "heartland Labour seats" being lost and never the opposite?

Just asking.

Labour isn't ahead. If they had the kind of candidate that appealed  to Tory Remainers like BoJo appeals to Labour Brexiteers than there would be serious chatter about seats like Maidenhead, Tunbridge Wells, SW Hertfordshire, Harrogate and Knares, Hitchin & Harpenden, and the rest. But we  don't have that luxurious situation. As it is Labour are struggling to gain Chingford and Woodford Green, a seat that I have to go out on a limb on when I say they gain it via tactical voting and incumbent rejection.

[Oryxslayer]

Here’s my prediction at 10 days out. I’ll make an update next week, but here’s my current expectation/model/projection, or whatever you want to call it. I keep going back and forth on quite a few seats so there will be at least marginal changes by next week.

Conservatives: 345 Seats, up 27 since 2017, working majority of 23

Labour: 209 Seats, down 53 since 2017

SNP: 41 Seats, up 6

Lib-Dems: 30 Seats, up 18

Plaid Cymru: 5 Seats, up 1

Greens: 1 Seat, no change

DUP: 9 Seats, down 1

Sinn Fein: 6 Seats, down 1

SDLP: 2 Seats, up 2

Alliance: 1 Seat, up 1

Speaker: 1 Seat


Seat Map


Seat Change Map since 2017

The general theme is Conservatives making big strides in the north, and losing a bit in the south, but that bit is far less than the multitude of seats that could have been lost if the Lib-Dems caught fire. Whenever I make an ‘out there’ prediction, I’m going to explain my logic. I’m going to break my seats down into groups, and if you disagree with anything I say here, just pipe up. I’m rather far away and can only consume news, so any extra data points are appreciated.

[Oryxslayer]

Labour’s Silver lining

Every election has some seats that buck the general trend and flip towards the party on the backfoot. Unless it’s a tsunami where every seat moves towards one party to some degree, certain regions will still keep up their past trends. In this case, it is the greater South. In these seats it is Labour getting the benefit of tactical voting from educated Remainers, rather than the Lib—Dems. They are all marginal Labour Gains or Tory holds, so Labour Holding or Gaining any of these seats would be shocking. However, that’s what can happen in a region full of seats where Labour gained 10% in 2017. If labour starts to get better results in the next week then seats will be added to this group and removed from the next. Labour Gains or Holds about 33% of these ten:

Crawley

Hastings and Rye

Ipswich

Milton Keynes N

Milton Keynes S

Reading W

Southampton Itchen

Stevenage

Stroud

Watford



The Northern Collapse

This has been the tale of the election: how many seats can BoJo get from Labour in the North to offset potential losses in the South. Well, the losses seem small compared to where they once were, but the North hasn’t seemed to be willing to give Labour another go in many seats. Some losses seem like automatic flips like Barrow & Furness or Darlington, whereas others are more marginal like Bolsover and Bradford South. All told I am seeing 44 Tory flips north of Birmingham and south the Scottish Border (this includes Peterborough & Norfolk North), and the only other gainer in the region is the obvious Lib-Dem flip in Sheffield Hallam. Tory strength is helped by Boris Johnson’s approvals being the best in the Midlands, at least according to multiple regional subsamples. South of this divider there are almost no Tory gains, just the odd Leaver seat held with a marginal majority in 2017 going blue.


Northern Seat Map


Northern Seat Change Map since 2017

Of these 45, there is one seat I have designated as a ‘Shock Seat’ which the media will zoom in on once the result comes in. Currently on the map this is a Tory gain in Blyth Valley. However, it could be any of these others: Hartlepool (Brexit is essentially Tory), NW Durham, Sunderland Central, and a handful of others outside of the immediate NW. If a Sunderland Seat comes in as a marginal or a flip…well the TV stations are going to have something to discuss for a few hours.

Lib-Dem Calculus

The Lib-Dems are starting from a low point and should only be able to go up from their 13 elected MPs, the question is merely by how much. Focused campaigns with strategic voting are hard to capture by polls, and so I made some concessions in their favor when compared to regular models. The thing is, I am seeing the Lib-Dems near their higher end of my expectations. What evidence there is from West London suggests Lib-Dem strength even though the national picture is bleak. More on West London later, but essentially this ‘ladder’ is built by assuming the Lib-Dems are going to get tactical voting boosts in demographically similar seats, albeit not as large of course.

Lib-Dem start: 3 ‘Safer’ seats outside of London or Scotland (Oxford W & Abington, Bath, Westmoreland & Lonsdale)

Easy Flips: If the Lib-Dems don’t gain all these seats, ignore the rest of the ladder. St. Albans, Sheffield Hallam, Winchester, and Cheltenham should all be orange unless the night is bad for the Lib-Dems. This brings the non-London/Scottish total to 7.

Well or Not: If this is a high-end projection, then give the Lib Dems S Cambridgeshire and Guildford, bringing our total to 9. If not, add these two to the next category.

The Pack: Depending on how good you are expecting the Lib-Dems to do you can say any amount of these seats can be holds or gains. However, they are all harder to justify and I have been shrinking the seats accessible in this category daily. When the seats are all tossups, you cut them in two, at least that’s my theory. Right now, it’s ten/twelve, but that can go lower. These ten can be held or gained:

Altrincham and Sale W

Brecon & Radnorshire

Colchester

Eastbourne

Lewes

North Norfolk

SE Cambridgeshire

St. Ives

Totnes

Wokingham

Finally, there is room for one surprise. Everything from that constituency poll yesterday to Raab’s bad reception on the ground suggests that Esher & Walton is more than vulnerable. Raab is well known to be out of step with his district, and when there is a scalp on the line, the opposition get’s frenzied. In my eyes, this is going to be a marginal flip to the Lib-Dems (brining the Non-London/Scottish total to 15), but it will be overshadowed by another high-profile defeat right nearby…

Weird Seats

This is the last group that needs to be covered outside of the London/Scottish/Northern Irish Seats. There are a handful of seats that are just weird, and when your prediction makes errors here one shouldn’t be disappointed. They are:

Beaconsfield: There is no Lib-Dem candidate here, they have endorsed Grieve. Grieve is prominent enough to survive potentially as an independent. His views match his seat’s Remainer lean, and there is 0% chance of any other party winning the seat so tactical voting may be in full swing – especially since it’s essentially a Tory primary. A poll came  out yesterday suggesting that Grieve still couldn’t get a majority…but independents are hard to poll.

Birkenhead: Frank Field was actually smart and has a party to stand for rather than run with the Indie label. Similar to Beaconsfield there is 0% chance of anyone else besides him and Labour winning the the seat, so it becomes a Labour primary. Downside is that Beaconsfield is only marginally for leave, which likely isn’t enough to save field. Labour should easily win, but once again, Independents are hard to poll.

Burnley: This is as close to a 4-way marginal that one may find south of Edinburgh, which makes it especially weird. I say the Tories edge it out since the prominent Lib-Dems and Brexit candidates seem better positioned to eat more from Labour here, but your guess is as good as mine.

East Devon: Claire Wright is running again on her independent ticket, and maybe she has a chance, maybe she doesn’t. On one hand, she is a perennial candidate, and Swire isn’t here anymore for her to run against. On the other hand, her repeated runs have led to a dedicated base forming, and an open seat is easier to flip than an incumbent seat. Also, independents are hard to poll, so who knows.

Southport: Historically a three-way marginal, and those are always hard to call. However, the Lib-Dems seem to have fallen behind, which leads to the question of where their votes will go. The Tory argument is that this seat is in the North where the Blues are gaining, and the Lib-Dems have more in common with the Conservatives than with the low-income workers. The Labour argument is that Southport is an urban seat with a large Remain base who will reject Boris. Who knows.

Ynys Mon: First off, this seat is rather parochial and weird with its preferences. Secondly, it’s a three way race between the Conservatives, Labour, and Plaid. It’s hard to poll Plaid since they get so few votes nationwide, and their voters may always or may never vote tactically depending on the seat. Now personally I think this seat is not going to stay with Labour, it’s either the Tories or Plaid who are getting a pickup, but it’s a three way. Even though I am saying Plaid edges it out, who knows how things may fall.

The London Pushback

Despite the large number of seats flipping away from the Tories, London is a rather simple place to model. It’s one of the only places in the country where politics is similar to US-style coalitions, so we have quite a lot of inflexible seats. Visible Minorities, age gaps, income gaps, homeownership versus income gaps, these all lead to some traditionally stark lines inside the city. This time, these demographics have colluded to present the United Kingdom with two options. They are: Do you believe in the strength of Remainer tactical voting in West London, or do you believe that it won’t be enough?  

There is a lot of on-the-ground data and local polls coming in right now that suggests the Lib-Dem vote is both concentrated and tactical in mind. The leafy, wealthy, white-collar West London seats seem like natural places for the Lib-Dems to gain this cycle. On paper, they seem too Remain for BoJo’s new Conservative party, but also too ‘Tory’ demographically for Corbyn’s radical Labour. We have also got constituency polls that suggest a Lib-Dem surge in this ‘slice’ of the London Pie (obvious disclaimer on the accuracy of constituency polls), with nearly every Tory seat on this side of the city being a 3-way. The more interesting data point though from every constituency poll is that over 50% of the voting population is open to tactical voting depending on who the two realistic candidates are. The same demographic variables that make the region a Troy-Remain stronghold also make it politically attuned to national and local variables. The people here want their votes power to be maximized.

 The data that suggested many seats in the South could flip faded for most of the shire seats, but it never really left inside the city. Instead, it seems to have intensified. Esher & Walton is similar to these west London seats demographically, but it was always closer to the tail end of the Lib-Dem target list. So, if a poll is suggesting E&W is close while we head into the final week, it suggests that the Lib-Dems are on track for more gains to the immediate north. Tactical voting is always missed by even the best models, so don’t be dissuaded by YouGov calling the seats as Conservative Holds. If you think that Tactical voting is going to be occurring to even a reasonable degree in Remain-heavy London, then the Conservative Blue vanishes from all but the outlying seats.





Then there is the other view. This view does not believe in the strength of tactical voting to flip every seat. This view does not deny tactical voting will occur, we already know that, it just won’t go on to the degree that is needed for sweeps. At which case, everything gets far messier. Almost every seat enters a ‘weird’ category and needs to be analyzed individually. Also going against the idea of a sweep is the YouGov poll; remember this is a politically attuned area. Even though tactical voting is missed by the model, the tactical voters will know what they saw, and tons of leaflets may not change that. As it is, Chipping Barnet, Enfield Southgate, and Hendon are in said weird category, thanks to Jewish vote things. Will they go more for the conservatives, hesitant bunkmates that was ‘forced’ upon Barnet in 2017, or will they go Lib-Dem? And then will other conservatives of Labour follow them, and how many voters get stolen in the three way? Or will the Lib-Dems just end up wiping away the tory margin to give the seats to Labour…

Analyzing the seats under the ‘limited’ view requires going individual seat by seat. Richmond Park is more an extension of the already Orange outer seats, the Lib Dems nearly held the seat last time, and the former MP I running. It should flip easily. London & Westminster is less of an easy flip, but Chukka running there should bring over enough Labour voters to make the seat a marginal flip, especially since it’s open and lacking any MP to run on his own non-Boris brand. Putney and Wimbledon are both seats with marginal Tory majorities, but those margins came from Labour being the best Remain horse around in 2017. With the Lib-Dems making serious pushes here, who knows how many Labour voters or even Tory voters will flip tactically. Putney is open whereas Wimbledon still has Stephan Hammond, so there is an argument there that more voters are accessible to team orange or red in Putney. Both will be three-ways in this ‘limited’ scenario since the Labour base was already high before 2017. Kensington is a near perfect three-way so you would just toss the coin up. Battersea could be competitive, but the tactical voters will likely go Labour here. Finchley & Golders Green will have one of the largest swings anywhere in the country, but the limited view puts it in the uncertain pile with the rest of the Barnet seats. Finally, Chelsea & Fulham looks the most likely to reelect their Tory candidate, despite the Star power of the Lib-Dems and the prominence of Hands. It just looks like there are too many lockstep cons here, whereas the other seats were all marginal after 2017.

You, know London is still a Labour city so let’s stop talking all about the Liberal-Democrats. Unless the Tory lead is large enough that every seat moves right in some capacity, the city seems most poised to push back against BoJo. They know the guy after all, and can see through his personas. Like in 2017, Labour will likely pile on more votes here thanks to Remain momentum. Some of this comes from tactical voting, but more just comes from two more years of juiced turnout in in-migration adding more voters. Healthy Remain majorities also make seats more inflexible. This is why YouGov rates nearly every Labour seat in London as safe. All these factors combined are why I think IDS goes down. Remain London is going to move in one way, the Leave north is going to move in another. Labour’s got a good candidate on the ground there, IDS is hated these days in London, and his seat is the natural target for London Labour activists. The Labour will be one of the nights stories if it occurs, and overrule the potential for Raab to become said story. All these factors that I have listed are what’s going on in the Remain majority of London. The minority of Leave seats of the edges of the city are going to experience a similar effect to what’s going on in the rest of the country. Cruddas won’t be able to survive in such a Leave constituency that still had a healthy UKIP vote in 2017. YouGov has the seat as Lean Tory, and I do not disagree.

Scotland

The SNP is rebounding from 2017, the question on everyone’s mind is how much. Scottish politics is different both demographically and factionally from the rest of the UK, and Scotland has twice delivered swings that were unthinkable in other parts of the UK.



Gong back to 2017, the SNP was at a comparative low point. It’s hard to think of things like this, but the SNP has control of the younger demographics north of the border. Labour’s base is older, and more likely to on pensions with memories of Thatcher. The SNP therefore is the principle victim if turnout drops. This is now more of an issue for the SNP since their new base is the urbanized strip between Edinburgh and Glasgow, the rural highland fishing towns have been abandoned to the unionists. The SNP had two urban problems in 2017: Corbyn energized the  youth and flipped some from SNP to Labour, and turnout dropped because of SNP disenchantment. This saw the unionists make a handful of gains in the populated strip.

Looking ahead a week, these problems seem to have reversed. SNP turnout is up, and Labour is down. This is likely not enough to save most Non-SNP MPs between the cities. Only the Lib-Dems are up among the unionist parties, and their vote is contained to a handful of constituencies. Even unionist tactical voting cannot save small majorities of Scottish Labour, something that I will cover in a moment. These are the areas where a turnout bump is going to manifest, and it will flip most of the regions seats Yellow.

Outside of the urban zones though the story is different. The Tories are not down by much, and they are going to hold up in the places most amenable to their unionist message. These are the rural or smalltown highland seats. There is some limited tactical voting that occurs between the unionist parties, and the rural highlands/borders seem most committed to their unionism. The SNP realignment from this area has also done them no favors, and under my prediction nets no Tory losses in the Border or highlands zones. Note how the YouGov model, which doesn’t weight for Tactical voting still has most of the rural seats remaining blue, it’s the denser tory Seats like Stirling and Renfrewshire East that are going to fall to the SNP spike. Finally, I say NE Fife flips to Lib-Dems, but that more of an uncertain call. It’s the next seat on my Lib-Dem list to flip to another party, and I think the unionist vote is going to be more tactical here after the close 2017 result. Remember, the Lib-Dems are up in Scotland on 2017, their campaign strategy tends to be concentrated, and there are only 8-ish seats where that boost can be felt – not all of them realistic Lib-Dem opportunities.

Northern Ireland

I don’t think this sort of projection requires much explanation; it is in a way the average Northern Irish breakdown at this point. The UUP and the SDLP have pursued differing strategies for a revival, with different outcomes. The UUP are seriously contesting everywhere, to maintain their region-wide voter base, but this wide campaign won’t produce any seats. The SLDP are trading long-term voter loyalty in several seats for opportunity in a few specific seats. This seems to be resulting in Westminster gains, but it may not be the best strategy for long term stability in Stormont. Coupled with the expected DUP/Sinn apathy and the limited Alliance surge, we get a few targeted seat flips. These seats are the competitive ones, but this does not mean a guaranteed close race, just a seat that requires investment:



Belfast E: Pure DUP versus Alliance head to head fight. If you believe Alliance still has some momentum, then gains are going to manifest in Greater Eastern Belfast more than anywhere else, flipping this seat.

Belfast N: Nigel Dodds versus Lord Mayor John Finucane was set up to be a grudge match from the very start. I don’t think it’s outlandish to think Sinn’s odds went backwards a bit here after the public urination embarrassment, so I have it as DUP hold.  

Belfast S: Three-way battle between the SDLP, the DUP, and Alliance. From most perspectives it seems as if the SDLP are going to regain this target.

Fermanagh & S Tyrone: UUP versus Sinn in perhaps the UUP’s best seat around. The downside for them though is that the Unionist vote seems to be already consolidated around the UUP, who lacks momentum, whereas Sinn could still pull in Nationalist voters. Currently a Sinn hold.

Foyle: SDLP’s traditional stronghold is a place nobody except the Nationalists should have a chance at winning. The decision to pour in resources and run the  parties leader seems to have paid off, though the resulting gain will likely be marginal.

N Down: Hermon’s retirement opened up a seat for the DUP to potentially gain. There is no nationalist vote here to speak of, but there is a wealthy Remain Unionist vote. Maybe the Alliance can get Hermon’s old base near Belfast to flip Yellow, but right now it looks like the DUP will make this their gain of the night.

S Antrim: Solidly Unionist seat that was a UUP versus DUP battle in the past. The UUP are running their old MP Kinahan, but the slow UUP decline does them no favors. The seat least likely to be a flip on this list, but it will still be marginal.

S Down: One of the few Northern Irish polls released suggested a serious SDLP revival from the pit of 2017. If so, then South Down is back on the target list, even though it wasn’t a SDLP hit at the start of the campaign. It’s more likely to stay with the Sinners though considering the SDLP campaign strategy.

Overall, the 9 – 6 – 2 – 1 breakdown seems like the average and most likely of outcomes at this stage.

[Oryxslayer]

ORYXSLAYER MY $64,000 question for you is:  How do the Lib Dems ever catch fire?

Without proportional representation I do not truly see how the Lib Dems ever catch fire.

FPP systems are set up to discourage third parties.

Also party loyalty runs very deep.  I was a never Trumper until I considered the prospect of Hillary as President   My wife had preceded me and pulled me along.  

Ask Nick Clegg

Okay, in all seriousness, there are two ways for a minor party to surge under FPTP, or another restrictive apportionment system. I'm no fool, I respect Duverger and know not to trifle with statistics. However, the past is littered with examples of when parties, in this case of the Liberal to Left side, supplant each other. This cycle there was the potential for two different types of surges to occur, they just never happened.

The first hypothetical scenario is the case of the 'faster horse.' Historical examples  include Spain 2015/16, Canada 2011, Ontario 2017, Greece after the collapse, and France  2017. In this scenario the traditional party has royally stepped in it and can't wipe away the mess. A bad leader and a tainted brand have sent some 'scouts' out to other parties, but the majority still stay with Old Reliable. The new party meanwhile needs to have a leader with at least some charisma. The right has to be off limits, either  because of historical antagonism or a similarly awful brand/mess. Then a big event occurs, say a debate, notable rally, speech, scandal, endorsement, or whatever. At this point, all the somewhat politically attuned voters hop on the the surging party(ies) leaving only those dyed in the wool. Essentially, the new party just won the inter-left primary. If the system is somewhat proportional like in Spain or France's first round, here is where it ends. In FPTP systems though the lockstep voters look around, see what is happening, tell themselves they will always support the left, and en masse switch to the insurgent.

This scenario does not always result in the death of the old party. Some, like the Canadian examples, show that simply giving the old party time to sort out it's issues will heal all the wounds and the lockstep base returns. In other cases, the well to so tainted that nothing can ever be recovered. There was the potential for such a scenario this election: Corbyn is hated by the general electorate, Swinson was unknown, and Labour has a brand that is inflexibly trying to stitch together a chasm. What approvers corbyn has are the dyed in the wool voters, they would flip if forced to choose between Orange or Blue. the problem is, Swinson lacks charisma, and turned away the 'scouts' with her time in the coalition. Labour, for all it's troubles, played to the base this campaign, which means there is never any chance of desertion because Old Reliable needs trip to the repair shop. It's notable that Labour actually feared this scenario playing out, likely scared by Swinson's presidential campaign, which is why they put more  investment in retaining a hold on their Remain base rather than holding down Leave Northerners. This strategy was changed, likely coincidentally, after the YouGov poll dropped. In some way, the UK already experienced this hypothetical when the Tories 'endorsed' the Brexit party for the EU elections.

The Second scenario is the case of 'divide and conquer.' The right in general loves this tactic more than the left, and is happier to allow someone to play second fiddle if it means that the pair will down out all opposition. It also is easier to set up these kind of arrangements in proportional (think Scandinavia) systems, since such arrangements will not waste votes. In FPTP systems votes will naturally be lost in any sort of understanding, unless the alliance becomes formal and on paper. These arrangements emerge when one party is having trouble reaching out to another (often large, but not limited based on size) demographic or ideological group. So another party rises to fill the new gap that was either abandoned by the old party because of policy failings or a bad brand. In the end, both plaster over each others weaknesses and play to their strengths, since they will both be appealing to groups that never would have considered their partner in crime.

This scenario could have  occurred if Labour focused heavily on reasserting itself against a Leave opposition. Thinking about it, Labour really only has London as a concentrated base of remain seats, albeit London is large and vote heavy. If Labour focused more on BoJo, the Lib-Dems could easily step up and fill their natural role as "we're not Labour" in the Tory-Remain shires. This would clearly distinguish the two parties, but their stark contrast in targeted demographics and seats would work to each others benefit and result in combined gains. This hypothetical never emerged because London is inflexible, and Corbyn still played hard to remain rather then narrowing in on BoJo. London, in many ways, is the place where Corbyn is actually liked. Swinson also did herself no favors and staked out a position even more Remain than many Remain Tories. To some degree this hypothetical is still going on, but the seats it is occurring in are  limited a collection on the Lib-Dem's target list.

So these are two scenarios that work under the confines of Duverger and party loyalty, but still allow for a Lib-Dem surge. But said surge needed to manifest early. For example, early on I noted that Maidenhead had the numbers to flip if either situation occured, but it never came.

[Oryxslayer]

Its now being speculated that the SNP may drop a few more seats to the Tories.

(though personally I will believe that when I actually see it)

The SNP are going to go up in votes since 2017, that is as close to guaranteed as one can get. The problem is that these  votes are not efficiently distributed. This is an oversimplification, but half of the gains appear to be from 2015 voters who dropped off in 2017, and half are 2017 Labour voters. Both of these groups are concentrated in the SNP's new base of support: the urbanized strip between Glasgow and Edinburgh. This points to them getting 5 seats automatically: four from Labour and Stirling, and then potentially Renfrewshire East and Coatbridge as well. Those four Labour gains would have been five if they didn't throw away Kirkcaldy. So, lets say they are at +6.

Then we look at the other side  of the column. The Lib-Dems are  also up in Scotland on 2017, their vote is highly concentrated in about 8 or 9 seats, only and only a handful of those are realistic opportunities. So North East Fife could fall from it's effective tie last time - especially since the  unionists won the vote here and have a clear party to rally behind. Perth and North Perthshire was also extremely tight last time, and it's the last SNP in the highlands, an area which has become increasingly hostile to the SNP since 2014. So maybe the Tories pick that one up this time. The we have Central Ayrshire and Argyll & Butte, two more rural-ish seats that the SNP could face headwinds against a more tactical unionist vote. Finally, the Tories are supposedly getting positive signs from Edinburgh SW and Lanark & Hamilton East, two more suburban seats,though they are  both inside the SNP's urban corridor. If the Tories are seeing promising signs in these seats, then perhaps Renfrewshire East isn't in as much danger as it appears. So, lets say the SNP is down 2, with the  potential for more from this group.

This leaves them on net 4, with 39 seats. Not the most impressive if polarization is as strong as rumored. But that the thing, it's rumors so we shall see.

[Oryxslayer]

Okay, so time for me to update my predication. You can find last weeks general assessment found on the last page. In short, I said Tories 345, Labour 209. I thought that I might need to make changes, but it appears as if polls and models have moved into alignment with my projection last time, rather than the other way around. Conservatives doing good in the North, Labour+Libs holding up in the South and benefiting from tactical voting/targeted campaigns. Therefore this is my locked in prediction with some caveats:

Conservatives: 342 Seats, up 24 since 2017, working majority of 20

Labour: 220 Seats, down 42 since 2017

SNP: 38 Seats, up 3

Lib-Dems: 25 Seats, up 13

Plaid Cymru: 5 Seats, up 1

Greens: 1 Seat, no change

DUP: 8 Seats, down 2

Sinn Fein: 6 Seats, down 1

SDLP: 2 Seats, up 2

Alliance: 2 Seat, up 2

Speaker: 1 Seat



My regional breakdowns here are in response to my breakdowns from last time.

The North





Labour continues to be down by a lot in the Northern, Brexit-voting seats. There are less losses then the 45 north of Birmingham/Coventry last time, this time it's only 40. However, Remain/Leave polarization keeps BoJo on track for his majority.

The South



YouGov and other models show Labour and the Lib-Dem's picking up steam, which is what I always expected. Some gains have been moved about, but the trend is the same. Watford is a key pick for this theme: Lab+Lib can easily outvote the Tories if they vote tactically to stop Brexit, but if they don't then the Cons sneak through. Raab goes down.

Lib-Dems



I mentioned last time that the Lib-Dems are always undershot by models thanks to their campaign strategy. Aggressive targeting is missed by even the best projections because of how the vote is concentrated in a handful of seats and disperses in the rest. Well, at least in the South, the places this vote  is concentrated in appear to have come to an agreeable position in thee light. Every southern Non-London Gain is a pure Lib-Dem vs Tory grudge match for the top slot.

Only Brecon & Radnorshire deserves mentioning because that seat is more of a 'most likely' stand in. One of Ceredigion, Totnes, B & R, Eastbourne, Cheadle, Altrichim and Sale W, and Beaconsfield will not elect a Tory (or PC for Ceredigion), and B & R seems the most likely of the pack to stay with the Lib-Dems.

London



London has significant changes from last time around, even though it may not appear as such. The big thing was that 10 days ago I believed in the 'max Lib-Dem' tactical surge in West London. Data suggested back then that such a event was a realistic possibility. Recent data shows that this never materialized, and so we move down to the second potential outcome described then: confusion. Five seats (Twin Cities, Kensington, Putney, Wimbledon, Finchely & Golders Green) now are all unknowns. The majority of voters in all 5 do not want a Tory Brexit, but that majority is divided between Lib and Lab. In some seats, the Lib-Dems are the clear challengers, in others Labour. This is the area most likely to vote tactically outside  of Scotland, so maybe things will turn around in a single day. Here's how I justified each seat:

Putney & Kensington: Labour are now the clear Tory alternative. Polls show Labour close and enough Lib-Dems prefer them to the Tories. Models suggest both seats should flip. So, Tories only hold if they are able to walk down the middle despite tactical remain voting.

Twin Cities & Wimbledon: Lib-Dems are the clear alternative, but data is less favorable than in the Labour seats. Both should be flipping under normal models, but it is unclear if said models can apply. Wimbledon has better Lib-Dem fundamentals, but Stephan Hammond is the type of guy who can push back against t he unfavorable headwinds. Twin Cities has no incumbent and has a better baseline, but the Labour vote appears inflexible enough to let the tories through. I say one of the two flip, and Wimbledon was the easier seat to call.

Finchley & Golders Green: Every model can't weight for Jews because of their small numbers and flexible  membership, so it's best to ignore every models expectations for what happens in Barnet and go with the fundamentals. the fundamentals suggest the Lib-Dems are the only ones running a Jewish candidate, and certain Jewish communities may flip overwhelmingly to them, something polls will miss.

Besides this, I'm sticking with my IDS and Dagenham flip predictions.

Wales



So I'm taking the  middle path between YouGov's MRP and the welsh barometer. Voter concentration and flexibility suggests though that if Labour are significantly below what the MRP has them at then most of the north moves as a block away from Labour. That's essentially what happens here. Also, Ynys Mons is weird like usual but I called a PC gain a week ago and I'm sticking with it, even though the PC vote is hard to pick up and model.

Scotland



First off, no model can give you an accurate picture of what's going on in Scotland. Too many parties, too much tactical voting, too little historical data. I would be far more confident in giving you my prediction of 38/39 SNP, 13/12 CON, 5 LIB, and 3 LAB, and just leaving it at that. Since this a map thread though, here we go.

With such a mixed picture, you have to find something and cling to it. For this projection, I'm clinging to polarization. The 3 to 4 percent the SNP pick up are concentrated in the urbanized corridor between the cities, and they continue to move backwards in the Highlands. The one seat I am most uncertain about is Lanark and Hamilton East, there is a lot of conflicting information going on there. Lib-Dems are also up, so they gain Fife North East. Labour are down, but tactical voting savings theme in Coatbridge, and the SNP threw away their potential gain in Kirkcaldy.

Northern Ireland



The numbers on the tin have not changed since last week. I still think 9 DUP seats, 6 Sinners, 2 SDLP pickups, and 1 Alliance is the most likely outcome. However, I now think a caveat needs to be added to this normal prediction. In addition to the 2 SDLP pickups and 1 Alliance gain in Belfast East, I am saying a fourth flip will occur. I'm not sure if this flip will be a Sinn Fein gain in Belfast North, a UUP gain in F & S Tyrone, or the Alliance gaining both their tossups. However, the second Alliance gain is the  most likely of those three, so I'm putting it on the  map. Also, low chance of the Alliance getting Belfast South instead of the SDLP.