Anyway, my lengthy thoughts about Stoke:
One of my pet peeves about journalism is the ability of the majority of journalists to take a critical look at statistics and convey to readers what a report or study shows, and what it doesn’t show and whether this matches up with the claims of those who have created the report. We can all point to how the Mail, but actually virtually every reputable media outlet is as bad. They will get a press release and pretty much repeat it without ever trying to scratch below the surface of it. This tends to be as a result of tight deadlines and not enough journalists rather than it being a deliberate thing, but it is a massive failing of our current media.
I really think the same is true of trying to understand the numbers around elections. The media is currently obsessed with how UKIP might win because it is such a leave area. Yet since the referendum there is absolutely nothing to suggest there might be a UKIP breakthrough here. Nothing apart from the hype. UKIP thrive on hype. Good or bad. They are dancing the UKIP dance.
All the local election data has suggested that people are turning away from UKIP. The national polls are showing people turning away from UKIP. Most importantly of all, now the country has given the establishment a kicking, how much more of an incentive is there to vote UKIP?
The media is doing a good job of making out that people from Stoke are thick and racist. That’s not the Stoke I know. Yes it can be like going back in time when you go there, with attitudes to match, but it’s still a proud city. That should not be under estimated. A vote for UKIP will reinforce the stereotype that Stokkies are thick. It is precisely why many may make a point that they are not. If there is a credible candidate to do that with.
Then there are the numbers. Everything is saying that Labour’s core vote has collapsed since the EU Referendum. This isn’t strictly true. It happened in 2015. What masked it was a massive collapse in the Liberal Democrat vote.
Take a look at the results for the Stoke Central seat over the last twenty years.
The turnout in the nineties was higher; in 1992 it was 68.1%, in 1997 it was 62.8%. Then came Tony Blair and turnout dropped. By 2001 it dropped to 47.4% and has remained around that mark since then. Turnout was 48.4 in 2005, 53.2% in 2010 and 49.9% in 2015. The difference between the number of Labour voters in 1997 and 2001 was roughly 9500 voters. Last year’s EU Referendum led to an increase in voter turnout. We don’t know what it was for Stoke Central alone, but if you calculate the number of extra voters who came out for the EU ref for the whole city and assume that its evenly split, then the number of extra voters turns out to be about 9300 voters. This is, of course, why Leave.EU and UKIP are so excited. There is a lot to work with here.
The battle is to get people to the polling stations who have not been engaged in politics for 20 years. People who may already feel that they have stuck one to the establishment. People who in 2015 didn’t think it worthwhile to vote UKIP…. Point being why would you make a protest vote now, but not then? (To get them interested, you have to hype it.)
Not to forget that Labour were rejected at local level previously: the BNP were a disaster and the local council is currently made up of a bunch of independents, several Tories and one kipper (its effectively lead by the Tories who blame the Independents for the unpopular stuff). The point is that UKIP were unable to get their act together at this level and capitalise and there are plenty of disaffected Labour voters, who won’t see UKIP as a good idea for protest vote as they are too aware that it doesn’t necessarily change anything. Where is the motivation for a protest vote in Stoke. They’ve effectively had a series of them already. (Unless you hype it).
The referendum perhaps has provided some data for Leave.Eu for who to target. But there is a legal problem here, and I hope one that perhaps people pick up on. It depends on how Leave.EU collected database. If Leave.Eu did so, saying it was only in relation to the referendum, if they then share information with UKIP they could be in breech of data protection. They can only (legally) share general trends about where to find people and target the in campaigning. Whether its legal or not may be mute point to people who are so, how shall we put this, ‘rather economical with the truth’. The bottom line is, that UKIP need to know exactly who people are and get them to the polling station on 23rd Feb. Hype will help, but by-elections are unsexy beasts and that's where knowing the address of your people helps.
Let’s go back to the data of past elections again.
Whilst Labour’s number of votes and vote share has wavered, there has never been any doubt about the result. Stoke Central is Labour. 2015 was the closest anyone got with Labour on 12,000 votes to UKIPs 7000. There is no point in voting in Stoke as no one else has a chance. Or so the logical train of thought goes. This is what UKIP are counting on – that people care enough to seize their only chance to get rid of Labour and to see UKIP as the chance to do that. Do people really care enough though?
Leave.EU have suggested that their prediction are that there will be a 33% turnout and they will storm it. This is of course a way of trying to get some momentum behind people to encourage them to get to the polls.
Given the last General Election was 49.9% this would be a decent turnout relatively speaking. By-elections tend much lower than General Elections. I’m still not convinced it will be that high, though it might be if people are engaged in the election.
33% turnout represents roughly 20,000 votes cast. To win, you are probably looking at needing roughly 5000 – 6000 votes. Remember UKIP got 7000 votes in 2015 to Labour’s 12000. This is where it starts to look more difficult for them to achieve. At this point, you should start to ask questions. UKIP need several things to happen to a greater or lesser degree: high turnout from their 2015 voters, defections from Labour, a large number of disengaged voters to change their habits and turnout out, a depression of the Labour vote and defection from Labour to the Lib Dems. Oh and to hope that their 2015 voters don’t switch to the Conservatives at the same time. It is not impossible, but the perfect storm, is perhaps harder than the press is suggesting.
Back to the data again.
Lets look at the Conservatives and Liberal Democrats.
Conservative performance looks like this: 1992 – 12477 (27.9%), 1997 – 6738 (16.7%) drop in turnout happens here, 2001 – 5325 (18.8%), 2005 – 4823 (17.3%), 2010 – 6833 (21%), 2015 – 7008 (22.7%).
Liberal Democrat performance looks like this: 1992 – 6073 (13.6%), 1997 – 4809 (11.9%) drop in turnout happens here, 2001 – 4148 (14.7%), 2005 – 4986 (17.9%), 2010 – 7039 (21.7%), 2015 – 1296 (4.2%).
There are several significant points here.
- These people turnout out to vote despite there never being a cat in hell’s chance of either the Conservatives or LDs beating Labour here. You can conclude that these people are highly likely to turnout out for all elections. They are reliable voters that can be counted on. Especially if there is a sniff of a chance that their party might have even a slim chance of winning.
- When turnout out dropped the number of votes for each party stayed similar. There is a consistent core vote.
- They are politically engaged and therefore it is also more likely that their parties will know who these voters are, and where they live with their party data, particularly the Lib Dems, because of the way they operate. Labour never had a reason to collect data about where their supporters were in the past so both parties are likely to be ahead of Labour, and there is a fair chance that their data might well be as good if not better than Leave.Eu’s / UKIPs
- If the turnout is indeed in the region of 33% and the winner needs 5000 – 6000 votes, it is potentially achievable for both parties. Remember: These are by and large voters who will vote no matter what.
- There is one anomaly here and it’s the one that perhaps matters most to the outcome of the by-election. It’s not clear who this will favour, but it’s the bit of data no one is looking at closely enough.
Normally, the sitting government don’t fair well in by-elections. But this isn’t normal times and the party to kick in Stoke Central isn’t the government first, its Labour. The Conservatives might well lose votes to the LDs and through disaffection, but they could potentially pick up votes from both Labour and UKIP too, though perhaps these are not as reliable in turning out.
The big question mark is the one hanging over the 2010 Lib Dem voters. Between 2010 and 2015, 5743 abandoned the Lib Dems. Remember, these are people who are much more likely to vote, rather them just making a one off vote in 2010. No one was ever going to beat Labour in Stoke Central and everyone voting anything different in 2010 knew this even regardless of whatever point they wanted to make.
To illustrate the point, the number of people voting in Stoke Central has stayed pretty similar since 2001. It is not an unreasonable assumption to say it’s the same people who vote in every GE in Stoke Central (or more precisely the same demographics). This make the 5743 voters who abandoned the LD in 2015 every bit as important, if not more so, than the 9500 extra voters at the EU Referendum. What are they going to do now?
Some may vote Labour purely to keep out UKIP, but Labour have fielded a candidate so bad that they might just think to hell with it, and vote with conscious rather than tactics at this point. Also, beware the Shy Lib Dem in Stoke. It is one of the few places it’s a possibility.
Turnout was also slightly up in 2010 when Labour got kicked out of government. This benefitted the Conservatives and LDs most, which is why it is important. Some people voted LD in 2010 as a protest and then voted UKIP in 2015, but that’s generally considered a relatively small percentage nationally. Potentially it was a bigger effect in Stoke Central, but I’m not sure I buy that totally. (Plus, its one thing to make a protest vote, when the one you are voting for is never going to win, and another, when they have a good chance of winning and you might actually get stuck with them for the next 3 years). Most of these 2010 LD voters either went to the Conservatives or Labour in 2015. Yet the Conservatives didn’t gain much here in 2015. Which leaves, the majority of them going to Labour in Stoke Central in 2015. This is exactly where Labour have been faring worst in by-elections both national and council since the referendum. Both YouGov and ComRes have said that significant numbers of Labour voters are defecting to the LDs. Defecting or returning? As I said initially, Labour’s vote collapsed in 2015 but was masked by disgruntled LDs. I don’t believe there is a collapse of the centre in this country. Just a centre that has been disowned by political parties. The centre are currently homeless. And ANGRY about it.
There is more than one perfect storm scenario. In a perfect storm, UKIP could win. I’m not sure that’s the most likely scenario though. They still have a lot to do to make it happen. Labour still have the upper hand, no matter what is being said. In a perfect storm, the LDs could also win. That perfect storm needs UKIP voters to struggle in finding the polling station and Labour to hide under the duvet whimpering. The LD perfect storm could find either Labour and UKIP in third place or below. Or indeed both. A high postal vote, might work well for the LDs normally, but early voters might vote Labour to keep UKIP out, and any controversy over the Lab candidate might have come too late. Don’t forget the Conservatives either. They like the LDs haven’t got anything to lose here and can quietly campaign whilst Labour and UKIP outdo each other in their public gaffs. I don’t think the Conservatives will quite have the steam to pull off a shock, but I think their vote is solid, stable and reliable and that might in itself surprise people.
Three final thoughts:
Yesterday I was watching one of the other Labour Stoke MPs do an interview. She was talking about what voters were saying to her. She said they were fed up and wanted a change. This does not bode well for Labour. The exit polls in the US elections had people saying that, even though the polling suggested that Hillary would win. That doesn't mean that votes will go to UKIP though. More that they won't necessarily go to Labour and Labour voters are definitely depressed.
Labour are also desperate to make it a two horse race - like UKIP, and have down played how much the LD and Conservatives have been campaigning. I'm not sure how true this is. I have heard stuff which conflicts with this position. The fact Labour are doing this, perhaps tells you they don't think they can rely on their core support alone. Its the classic faulty LD bar chart leaflets tactic.
The other thing I noticed is that on the betting significantly more bets have been put on UKIP rather than Labour, even though Labour are still favourite. This is the same pattern as Brexit and Trump. I wouldn't get too panicky about this yet though. This is a by-election. Its ALL about the turnout.