September 2022 Projections Update

Seat Map

Four polls since Irish politics took its traditional break at the start of August – two from Ireland Thinks, including, unusually, one in August itself, one from RedC and one from B&A. Only the latter two have regional breakdowns, as usual. All of the polls are relatively in line on the national figures, barring B&A, as usual, having Fianna Fáil way ahead of where others have them, but more on this below.

Sinn Féin are still far ahead, but still will need a significant boost to get to the point where forming a government without Fianna Fáil seems likely. Even if they did corral all of the parties on the left (or nominally on the left, depending on your interpretation), they would be 5 seats short of a majority, and I’m not sure they would be able to make that up from Independents.

Conversely, the government projected total is also five seats short – and while in theory on an ideological basis they should have an easier time rounding up five right-wing Independents, I’m not clear how easy a sell it would be to get them to hitch their wagon.

Either way, the numbers, as they have for a while now, indicate that coalition forming will be extremely mess, and the output may not be particularly stable – unless FF and SF put aside their differences, though even doing that would risk alienating a good chunk of people.

Some other things to discuss this month, including some methodology tweaks, but if you would prefer to skip straight to the seat changes, click here.

Fianna Fáil unaffected by a month of scandal

Something that may be a little surprising, given that FF spent most of August embroiled in a scandal over property registrations, is that this hasn’t been a bad polling period for them at all. They are up in both B&A and RedC compared to the previous polls (up three points in both), and while these fluctuations are normal, it is still telling that they didn’t seem to lose out. This is despite RedC putting them at a mere 7% in Connacht-Ulster, which seems a bit off to me.

Maybe everyone who cares about this kind of thing has already made up their mind, or maybe this kind of petty corruption has brought back memories of the good old days of Fianna Fáil among their supporters.

Fine Gael are worried

On the other hand, things look a bit rougher for Fine Gael, who continue to slowly decline. 18% is a new low for them in RedC, and B&A having them at 20%, four points behind Fianna Fáil is not good either – even accounting for B&A’s tendency to overstate FF. Now, that said, they are still ahead of their partners, but with their turn leading government rapidly approaching, this is not a good situation for them to be in. In September 2020, FG were nearly 20 points clear of FF on an RPA calculation. Right now, that gap is less than two points.

While FF have recovered a little from their post-election lows, FG have suffered much more, losing over 35% of their support over the last two years. While we occasionally hear rumours of panic in FG HQ in response to a particular poll – as we did with the B&A poll in September – the problems run a lot deeper than one-off instances. Yes, they are not too far below what they got in the last GE, but in the aftermath of government formation, there was huge poll movement towards them and it looked like they could unify the right and centre behind them. That opportunity now seems to be gone.

Good signs for Greens in Dublin

Here’s a quiet one though – after a few months of really disastrous polling, something seemed to change for the Greens this month in Dublin. Both RedC and B&A had them high – 10% and 11%. Both pollsters agreeing on this makes me think this is less likely to be an outlier; these are the Green’s highest polling results in Dublin since February. While this hasn’t resulted in seat gains in the model yet, another similar month could change this – which is interesting, because this comes right after the party seemed on the brink of losing that final seat under modelling conditions.

It’s quite hard to work out what to attribute this to – nothing really seemed to change with the party in Dublin over the last little while, and Eamon Ryan managed to once again set a new record-low approval rating this month, at -36%. But attempting to work backwards to look for policy as driving polling is a bit of a fool’s errand, so let’s see if this lasts or is just a blip.

Some methodology fixes

A few notes on some tweaks to the methodology that I’ve put in this month:

• Made adjustments to Independents to reflect whether or not they beat provincial trends in 2020. Limited in scope as it can only be calculated for people who ran in both 2016 and 2020, and probably will need improvement as time goes on, if I think it worthwhile and it doesn’t start producing strange results. This did move a bunch of Indos a little bit higher, but only one into a seat at this point.
• Adjusted Social Democrat numbers to account for some regional polling errors in 2020 – this has minimal impact, except making Holly Cairns more competitive in Cork South-West
• Adjusted Donegal to try to account for Joe McHugh (FG) retiring – this is a work in progress but is something that will need to be accounted when we face more retirements heading into the next GE

Seat Changes

Three changes again this month as trends broadly continue, however there are, as discussed above, some underlying things that happened in September that haven’t manifested in a seat change yet.

Clicking on the linked name for each constituency should jump to the relevant section of this page. Changes on this page indicate changes from July 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.

• Dublin Bay South (FF +1 FG -1)
• Galway East (IND +1, FF -1)
• Tipperary (SF +1, FG -1)

This probably shouldn’t be surprising given what we’ve seen here so far this year, and with Fine Gael continuing to slide in Dublin this should now favour Fianna Fáil. That said, there are realistically four candidates – FF, GP, Lab and a second FG one – competing for the final two seats and the probabilities right now are still really close. This is one of the most brutal constituencies in the country, with four big hitting incumbents, and at least one seat that Fine Gael will absolutely win – so on current polling, there will be a high-profile casualty, whatever happens.

There’s still unquantifiable factors that modelling can’t really account for – for example, how much support for Bacik (LAB) holds over from the by-election, how much the general unpopularity of Ryan (GP) will be a factor, and the fact that Fine Gael don’t even seem to have one, let alone two, inspiring candidates in this constituency. Indeed, just how much Ryan and Bacik are going to be competing for the same votes is a huge consideration and could well result in one or the other doing far better than modelling can predict on the day of an election.

So really, one seat for Fine Gael, one seat for Sinn Féin, and then honestly, who knows. It’s also worth noting that with the FG seat melting away here, there are only two constituencies left where the model thinks Fine Gael will win multiple seats – Rathdown, and Dún Laoghaire.

Making the changes this month to the Independents didn’t change many seats – though it did make some more or less likely – but this is one of the ones where it did. I think Seán Canney (IND) was definitely being underestimated by prior modelling, so a movement in his direction here makes sense. Canney has topped the poll here in the last two outings, albeit because other parties ran multiple candidates.

Sinn Féin were less than 500 votes away from a seat here in 2020, so it seems inevitable on current numbers that they will take one, with Fianna Fáil’s Anne Rabbitte now looking the most likely to lose out. That said it still looks tight even after the adjustments, so wouldn’t take anything for granted – Fianna Fáil and Fine Gael both ran two candidates each in 2020. Repeating that in the next GE would be ill-advised, particularly for FF, and running as a solo candidate next time may benefit Rabbitte.

This one is interesting – on current numbers we have a very tight situation for the last two seats between Mattie McGrath (IND), Fine Gael and a second Sinn Féin candidate. The model has never been particularly fond of McGrath, but the adjustments made this month do shore him up a little bit, giving him and a second SF candidate a very small edge over FG.

That said, even last time out, McGrath got less FPV than the combined Fianna Fáil or Fine Gael totals, and general polling trends indicate that SF have eaten a bunch of Independent votes – though again, how much any of those will be relevant to somewhere as idiosyncratic as Tipperary remains to be seen. Regardless, FG are in really rough shape across Munster, and this may be a harder nut to crack for SF than regional polling indicates.

Finally, worth noting that there’s a lot up in the air here still if Alan Kelly (LAB) opts not to run again – given how deeply geographically divided the constituency is, it’s hard to see that benefitting McGrath. However, by the same token, the absence of Séamus Healy (IND) is likely to work in McGrath’s favour; despite the vast political gulf between the two, McGrath got a substantial plurality of Healy’s transfers in 2020 – 42%, discounting non-transferrable votes. SF also took 33% of those transfers, which to me further indicates that Fine Gael will be up against it.

Other notes

• Cork South-Central: Fine Gael are doing really badly in Munster, and how weak they look here is reflective of that. I think the model may be underestimating them, as Simon Coveney is a strong candidate, but the fact that their seat here is even showing up as under threat indicates the depth of their problem. If Coveney gets promoted to Europe, this could get spicy.
• Dublin Central: Very tight here between the Greens, Soc Dems and Fianna Fáil still. I wouldn’t call this as anyone’s at the moment, the gap between FF and the SDs is basically zero, and if the Greens have another month of polling like they did in September, they could pass them both.
• Dublin Fingal: Similar to the above, we have the Greens looking a lot better after the last round of polls and much closer to holding the seat than before. The really interesting thing here is that if they do come back up, Sinn Féin’s projected second seat is, right now, the most likely casualty.
• Dublin Rathdown: Another one where the Greens are again close to showing a hold again. There are also scenarios starting to develop where Fine Gael don’t hold two seats here, it wouldn’t take too much more movement to throw this all into chaos – and to add to that, this is one of those places where a Labour candidate really could sneak something as a anti-but-not-too-anti-government option.
• Kildare South: Just a quick note, as Fiona McLoughlin-Healy (IND) has left her council seat. Modelling currently assumes she will run for the Dáil, but if she opts to quit politics entirely (which, based on her reasons for resigning, doesn’t seem impossible), that potentially puts a significant chunk of votes up for grabs. Her transfers went all over the place last time, but fellow Independent Cathal Berry was the biggest beneficiary. One to keep an eye on.

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.