January 2023 Projections Update

Seat Map

Five polls between December and January, but only three with breakdowns – two B&A and one RedC (the other two were from Ireland Thinks), and despite a lot happening in politics, not a huge amount has changed. Before we get into it, I want to apologise for quality issues on November’s post – it appears I published a draft with several errors and a placeholder section that amusingly said “words go here”. This has been fixed.

Fine Gael’s supposed “budget bounce” – which I explained in the November article was at best speculative nonsense – continues to be a myth. Ireland Thinks did have them at 25%, their best result in over a year, but two weeks later, B&A had them at 19%, their worst result in a B&A poll since May. So on average, nothing much has changed; all of this washes out at a change of less than 1% on national RPA since November. The party has decided to get involved in a dizzying array of scandals, but I’m not convinced they are going to have a massive impact on polling – doesn’t feel like any of them have really provoked a mass wave of public anger, rather just resignation and vague digust. But they are certainly trying!

Similarly, Sinn Féin have continued the minor downward movement in recent months, but we are talking a change of less than 1% on national RPA since November – nothing to get excited about, sure, but, combined with movement for other parties, it does change the calculus in a couple of marginal seats, as we will see below.

It would be remiss not to mention at this point as well that Labour’s national RPA is the lowest I’ve calculated it to be since June 2021, as their struggles outside of Dublin under Bacik continue. Her leadership has certainly resulted in progress in the capital, but polling support there has subsided a couple of points. This article is worth reading; it’s not hard to understand the party’s struggles in light of what they themselves reveal there.

A couple of more detailed thoughts below – including a very long one about a truly dreadful piece of presentation by RedC on the issue of refugees – but if you would prefer to skip straight to the seat changes, click here.

Fianna Fáil polling remains a mystery

We’ve been here before, but I think it’s worth flagging again just how spectacularly disparate the two main polling companies are on Fianna Fáil in their January polls. RedC do have FF consistently lower than B&A do – but the overall gap this month is 10 points – RedC on 15% and B&A on 25%. That might be the biggest gap we have seen in a single month since the last general election.

The gaps regionally are also pretty intense, as can be seen above. This indicates a systemic difference rather than a regional problem. It may be connected to weighting in methodologies or who they are polling – or even the methods of contact, phrasing of questions, breakdown of options etc. But it’s hard to pinpoint a single factor; while pointing to weighting is tempting, the unadjusted figures in B&A still show FF on 25%. I don’t really know what’s going on here, honestly.

Indeed, as we can see here, with the exception of Fine Gael in Connacht-Ulster, all of the biggest gaps are on Fianna Fáil – this is a consistent phenomenon that doesn’t seem to be indicative of a wider polling error, but one localised to FF. The possible exception to that is Connacht-Ulster where the gaps on FF and FG are pretty close – indicating there’s probably something pretty cleanly divided between how supporters of those two parties are being polled.

The excellent Irish Polling Indicator reckons that B&A and RedC, on average, get FF wrong by roughly equal amounts in opposite directions. That might mean that this washes out to a large extent by using polling averages, but it still leaves a large amount of uncertainty about how they will actually do when an election comes around – unless we see a significant convergence in the numbers when we get closer to one. And while RedC were closer in the end in 2020, B&A also stopped polling much earlier than they did (missing the SF surge entirely as a result) and their final poll was a positive outlier for FF even by their standards, so unfortunately that isn’t particularly indicative either.

Interesting results for the Social Democrats

Let’s get it out of the way up top – RedC had the Soc Dems on 11% in Dublin. Before we get too excited and start yelling about a purple wave or whatever, we have been here before, in March ’22, where that number was shown to be an outlier. I would advocate strongly to take the same approach here unless we see this reflect in a sustained manner, and of course it must also be noted that RedC tend to be the most friendly pollster for the SDs.

By the way, if the SDs did manage to get 11% in Dublin, do you know how many additional seats the model would give them in the city, assuming that other parties also get their results from the same poll? Around 4, on top of returning all their currently sitting TDs. That would be, needless to say, absolutely massive for them.

However, while the Dublin figure is attention grabbing, something much more quiet but more sustainable is happening elsewhere. I mentioned very briefly at the end of the last post that the SDs had seen relatively better polling in Munster recently, and that has continued. While extra caution must naturally be exercised with small figures, their RPA has risen to 2.8%, which might not sound like much, but represents a doubling of where they were a few months ago.

Now, the SDs have polled better than this in Munster, and the model hasn’t allocated them a seat in Cork South-West, but with Sinn Féin doing comparatively less well in Munster, it’s enough to indicate a change in who the model favours. I think measuring this is tricky overall because Holly Cairns ran well ahead of the party average in the province, and their support numbers across Munster are low in general, but I still think this is a more concrete positive than one poll in Dublin.

Irresponsible polling from RedC on refugees

So here’s another difference between B&A and RedC. Recently we’ve seen a bunch of far-right agitators – who poll about as well as ebola, but are similarly dangerous if left unchecked – lead a bunch of anti-refugee “protests” (they are not protests, so much as sustained intimidation campaigns), as well as violent actions. RedC have polled on this and, man, it’s not good. Now, the selection of questions is certainly the Sunday Business Post’s fault, but RedC have compounded this by acting in an irresponsible manner. I have covered this kind of thing before and it’s depressing to see it come up over and over again.

RedC lead their analysis with the claim that “anti-immigration sentiment [is] bubbling up to become a major political issue in Ireland“, and the SBP appear to have got the headline they want out of it. And note that this article is written by the fella from RedC, not a pol corr. However, the evidence from this in the poll itself is really weak, so this presentation is really irresponsible. This kind of abuse of polling data – and it is abuse – to create reactionary headlines and false narratives is dangerous at the best of times, and doubly so when it is turned on vulnerable groups.

Firstly, let’s be very clear – there is nothing, absolutely nothing, in this poll or in any other poll that indicates any kind of political support for the parties that are behind these protests. Yes, there are quite a few people in the Oireachtas who one could easily claim hold racist views about immigrants, but guess what? They were there already, long before this round of violence and intimidation. There is no evidence of any political support or electoral movement coalescing around this issue.

Then we get to the actual questions themselves. There are three, and they are all complete rubbish and categorically do not support the claims being made that anti-immigrant sentiment is the hot new thing. Let’s look at why.

Irish people welcome refugees on the whole, it’s just far-right activists opposing this (55% agree, 34% disagree, 11% no opinion)

This doesn’t assess sentiment. This doesn’t tell us anything about how people feel about refugees – it tells us a very basic thing about people’s perception. And what does it mean by just the far right? Does someone who says “no” mean “I also am racist”, or simply “I am aware that Direct Provision exists and is bad”, or any of the many, many things in between? Because there’s a huge gulf of overlapping perceptions that could easily be bucketed as yes or no. This doesn’t indicate anything about how widespread anti-refugee sentiment is, nor does it speak at all to the political salience.

The government is doing a good job at dealing with the Ukrainian refugee crisis (43% agree, 49% disagree, 8% no opinion)

Put aside for a moment that the word “crisis” is doing a lot of heavy lifting in this sentence. We have a very similar problem here as we did with the previous question. This is merely people saying that they think the government is making a mess of dealing with refugees – which honestly has been the case since, forever. How on earth is this a measure of anti-immigrant sentiment? Saying “no” here because refugees are being forced by the state to live in crap conditions isn’t anti-immigrant. There’s an inherent presumption in the presentation that anyone who is unhappy with the state’s treatment of refugees is doing so out of some kind of opposition to the refugees. That is insane.

I would support the government using new powers to install modular homes for Ukrainian refugees, without planning permission, in my area (46% agree, 45% disagree, 9% no opinion)

This is probably the closest that they have to proving their point – if only because they asked the same question about housing waiting lists, and got a more positive response, with a swing of about 17%. So yes, on one very specific proposed solution, with four caveats (“new powers”, “modular homes”, “without planning permission”, “in my area”), people are more about it being applied to housing lists than refugees. Great. But there are four caveats. Do we know which one is the issue? Do we know if other highly specific solutions have similar or different results? This is really isolated, and at the end of the day, they asked Irish people if they want to let someone build a house without planning permission. Come on, man.

But it gets worse. Far more Sinn Féin supporters said no on all questions than Fine Gael or Fianna Fáil supporters. This can hardly be surprising; opposition supporters don’t think the government has handled things well, and thinks people other than the far-right aren’t welcoming to refugees (for example, the government aren’t very welcoming right now). And how is this presented? “Government support[er]s [are] more likely to welcome refugees”. And that is verbatim from the RedC report, despite there not being a single scrap of data that supports this claim.

It’s true that SF are not great on refugees – their 2020 manifesto was weak on the issue and they have failed to get a grip on the current situation. They hardly need this kind of fabrication on top.

In short: this is fucking awful. Stop abusing polling data to reach unsupported conclusions. This feels like pushing to create a dramatic narrative, ignoring that it will put real people at risk. This ought be beneath RedC – asking the questions is bad enough but I understand that’s the nature of the business. This kind of presentation and editorialising is reckless.

(Anyway, seems if they really had to do this it would have been easier to just do the kind of ranked-issue polling we have seen in the past, where, by the way, immigration has historically not been a priority issue for 95%+ of people. But abstruse questions mean that the answers can be interpreted to mean whatever you want, so you can get your dramatic headline about how everyone, but especially Sinn Féin, hates refugees now.)

B&A, on the other hand, didn’t ask any polling questions, merely stating, and I quote: “There have been some protests against refugees, with some elements attempting to hijacking [sic] these to promote their message of anti-refuge [sic] sentiment.“

No concern for spelling, no concern for coherence, no thought for the fact that they are making the nonsensical proposition that anti-refugee protests have been hijacked by, uh, anti-refugee protestors. What on earth are they talking about. Complete nonsense, yet somehow still less nonsense than the SBP and RedC’s attempt at polling this issue seriously.

Seat Changes

Despite there not being a ton of polling movement, we do have a few seats showing changes in likely winners under the model – five, in fact. Two of these are reversions from November in Green Party seats; the rest are new.

Clicking on the linked name for each constituency should jump to the relevant section of this page. Changes on this page indicate changes from November 2022’s projections; changes on constituency pages indicate changes from current composition.

Note: The projections reflect, and always have reflected, most likely outcomes. So if a final seat is more likely for candidate X over Y, the model will show X winning the seat.

This does not mean the scenario where Y wins doesn’t exist, or even is necessarily unlikely (there’s a lot of marginal calls!). It also does not mean that every single “most likely” scenario will come true; statistically that in and of itself is probably not going to happen. This is true from from a simple probability point of view, even if we ignore deficiencies in underlying data. A projected result merely means that the model thinks X winning is the most likely outcome.

• Cork South-West (SD +1, SF -1)
• Dublin Bay South (GP +1, LAB -1)
• Dublin Rathdown (GP +1, SF -1)
• Dublin West (SF +1, PBP -1)
• Laois-Offaly (IND +1, SF -1)

As mentioned in the main article, there have been positive signs for the SDs in Munster, enough so that the model now favours Holly Cairns to hold her seat – which hasn’t been the case for a very long time. Also important in this is Sinn Féin dropping off from their highs in Munster. While they are still well ahead in the province, they are down a bit from their RPA peak of 37.4% to 33.3%. This has previously resulted in a few seats the model projected to go them slipping away, and Cork South-West is the latest among these.

I will once again emphasise that Cairns is likely to outrun her party to at least some degree – I do believe the model underestimates her somewhat. So while nothing is guaranteed, this is in my view a solid sign that the SDs should hold this seat if current polling conditions stay stable.

Dublin Bay South has been featuring a lot recently, and I suspect may continue to do so. Two things factoring into a reversal of November’s change here – the Greens RPA in Dublin has moved up from 7.1% to 7.9%, while Labour’s has dropped from 7% to 6.5%. Small movements are once again enough to tip the model in what continues to look like an extremely tough to call constituency. Ivana Bacik and Eamon Ryan are both big profile politicians – for better or for ill – and I still wouldn’t rule out the possibility that both of them end up holding their seats ahead of FF and a 2nd FG candidate when all is said and done. But for now, the model sees the advantage moving back to Ryan.

Similar to Dublin Bay South, that small RPA bump in Dublin for the Greens since November – from 7.1% to 7.9% – proves to be enough to swing the odds back in their favour for Catherine Martin to hold her seat. I did mention last time that the Sinn Féin seat in Rathdown was probably illusory, so I’m not too surprised about this. This is a really tough constituency for SF – or FF for that matter – with three high-profile incumbents and a somewhat socially-liberal but otherwise highly conservative voter base. Still, see previous comments about this place being weird as all hell politically. It’ll probably end up electing Hugo MacNeill or something.

Sinn Féin up a little in Dublin, People Before Profit/Solidarity down a little and that pushes this seat back into SF’s column. There were always, I think, a ton of variables going into a PBP/S gain here but it’s still not an impossibility.

Not much else to add that hasn’t already been mentioned in recent coverage of this constituency, but worth noting as well that SF and PBP/S aren’t the only possibilities here – the Greens could, with a recovery of a couple of percentage points be competitive here. Also there is a scenario where FG win a second seat – but their historically terrible management of vote splits in Dublin is factored into the model, and thus disfavours them against SF, even if the projected FPV share is roughly equivalent.

Independents have been polling a bit better recently so seeing another one slide into a hold isn’t a great surprise, nor should it be that Carol Nolan is close to the top of that list. The model has had her in and out of the seat a few times since it began, and it’s still fairly close, but she’s looked competitive fairly consistently. As for Sinn Féin dropping the seat – that’s the most likely, but they are not far from the 2nd Fianna Fáil seat. Transfers, which are, as I have mentioned many times, incredibly tricky to project, will be key in a constituency that the model has three seats locked down and two open.

Then again this constituency is a prime candidate to be split before 2025. Of the current TDs, there are three based in Laois (SF, FF, FG) and two in Offaly (FF, IND). How that affects the dynamics of a race if they are split remains to be seen, particularly if both are three-seaters.

Other notes

• Kerry: Not much of a polling note, but incumbent FG TD Brendan Griffin has announced he won’t run again. I’m not sure how this will affect things – measuring incumbency vs replacement performance is quite dodgy ground – but FG are potentially vulnerable here if they can’t find a strong replacement. Kerry may also end up getting split again, so for what it’s worth, Griffin was first elected in the old South Kerry constituency.
• Kildare South: If the Ceann Comhairle opts to retire, the model now reckons Cathal Berry (IND) will keep his seat ahead of Labour and Sinn Féin. I haven’t gamed out what will happen if Fiona McLoughlin-Healy’s retirement from the County Council is also a retirement from contesting national elections, but I suspect her not running will also help Berry, who got a plurality of her transfers in 2020.

Thanks so much for reading! This website is done entirely in my spare time and run without ads, so if you want to donate, please do so via Patreon or Ko-Fi – any support is greatly appreciated and 100% of any donations received are invested into the costs of running and improving the site.