This coming Friday has the general election in Ireland – for those who are interested, here are the latest polls.

Date Fianna Fail Fine Gael Labour Sinn Fein
Ipsos MRBI/Irish Times 21/02/11 16 37 19 11
RedC/S. Business Post 20/02/11 16 39 17 12
Millward Brown/S. Independent 20/02/11 16 37 20 12
OI/Daily Star 17/02/11 17 39 18 10
Millward Brown/Independent 16/02/11 12 38 23 10
RedC/S. Business Post 13/02/11 15 38 20 10
RedC/S. Business Post 06/02/11 17 35 22 13
Ipsos MRBI/Irish Times 03/02/11 15 33 24 12
RedC/Paddy Power 02/02/11 18 37 19 12
Millward Brown/Independent 02/02/11 16 30 24 13

I’ve no particular insight to offer into polling methodology in Ireland – Millward Brown and Ipsos MRBI are traditional face-to-face polling using quota sampling, RedC is phone polling very much along ICM lines, with past vote weighting and suchlike (I have a recollection that it was set up by a former ICM employee). That said, there isn’t a huge difference between the pollsters anyway (though RedC appear to be showing Labour slightly lower).

Across the board Fianna Fail have collapsed to well under half of their general election vote, Fine Gael are just below 40%, Labour have doubled their support since the election. I haven’t included them in the table, but the Greens are in low single figures, but Independents/Others are up in the mid-teens (I haven’t tracked down any recent polls on the company websites that have broken that “Independent/Other” down into it’s component parts).

With STV I don’t think there is a widely accepted equivalent of a swingometer to translate shares of the vote into seats. Certainly it will depend to some extent how votes transfer between the parties in individual constituencies.

There is a nice table and graphing of voting intention polls here, and some commentary from my Irish equivalent (imitation is the sincerest form of flattery!) at IrishPollingReport.

60 Responses to “The Irish General Election”

1 2
  1. @thegreenbenches:

    The only taboo in ROI transferism is FF transfers to FG or vice versa..

    It’s now much less of a taboo than it would once have been and it doesn’t really apply at all at more personal contests such as local elections.

    There is the smaller issue of FG transfers to SF but ask a Shinner and they’d probably invite the FG to keep their transfers…

    Again, not all Fine Gael supporters at the grassroots are as virulently anti-SF as their leadership would be.

  2. First time poster, so be gentle :)

    As an Irish recent ex-pat (and also disenfranchised), I’ve been watching this election with a mixture of interest, trepidation and horror. I’ve not been too impressed with the accuracy of Irish opinion polls in the past – look back to the Euro and local elections of 2009 for example, where the last-minute opinion polls overestimated Labour and FG, while underestimating FF and others. To be honest, there could well be a shy FF effect here.

    I’ve conducted a very unscientific poll myself – using the odds from a well-known Irish bookmaker (whose name is two brands of Irish whiskey…), and came up with the following seat predictions.

    FG 70, FF 28 + 1 (Ceann Comhairle), Lab 38, SF 13, Greens 0, Others 16.

    For comparison, the results of the 2007 GE were

    FG 51, FF 77 + 1, Lab 20, SF 4, Greens 6, Others 7 (including PDs)

  3. Paddy,

    Your first post was a good one, and I agree with most of it. On the second, you are seeing the glass as half full if I may say so…

    I look forward to measuring the shares from the transfers and judging the extent to which you are correct.. Obviously, a departure from civil war politics would be beneficial but in my view we aint as far along the road as you think we are…

    I hope you post after Friday so we can discuss it further.. :)

  4. OldNat: “You admire (it would seem) the corrupt techniques of SF in Belfast.”

    Corrupt? Not at all, it’s simply playing the system to their advantage and no different from what political parties do in GB. Parties in GB work out who’s likely to vote for them and where and then direct resources accordingly from hopeless seats to marginals.

    You’re probably right though that on this specific occasion all the balancing was unnecessary. I ran it through a spreadsheet, giving Gerry 8000 more and taking 2000 votes off each of the others and on the penultimate count the last SF candidate was 8 votes ahead of the DUP with 270 SF votes still to transfer, so theoretically they could have achieved the same result just by having everyone vote SF 1,2,3,4,5 but that’s a risky strategy and much better is to simply balance candidates as much as possible. Many people don’t vote down the ballot as they’re often too stupid to realise that later preferences never count against higher preferences.

    The southern SF seem to be a bit behind their northern counterparts on this one. The talk is that they’re still busy putting energy into canvassing in Louth, where Adams is certain to be elected, rather than focusing energies on places like Meath West, Cork East or the awfully named Dublin Mid West. Obviously the propaganda boost of having Adams top the poll there is more important for them than having an extra TD or 3.

  5. I think that shows that vote balancing is less important that often thought. It certainly can improve a party’s chances of taking an additional seat when things are tight, but tactical voting is much harder to carry off under STV.

    Much more important is to get out and canvass in seats where the party stands a chance of taking a seat (or two, or three), but that isn’t guaranteed. Just getting supporters of other parties to prefer your candidate to some other party can make the difference in the race for the final seat.

    Some of the comments on the previous page about the “tremendous premium” of getting, for example, 20% of the votes in a five-member seat are well off the mark. If a party comes a little below 20%, they’re almost guaranteed to get enough transfers to win a seat. Even if they didn’t take any transfers, they’d probably win a seat after some other ballots are exhausted. So long as you get enough first preference votes not to be knocked out very early on, you’re still in with a chance, if you’re everyone’s second or third choice.

  6. There is some poor understanding of STV going on here and I think I know the reason for it…

    try this guys..

    Sinn Fein had FIVE candidates.

    In terms of securing transfers from other parties

    a) they are unlikely to “tick em all” hell 33% refuse to tick anyone…

    b) persuading them which of the FIVE to transfer to is the challenge…..

    This key point is being lost in your calculations…

    I dont know if anyone of you have ever seen a West Belfast polling booth… listening to you I doubt it…. It is a an operation of sorts… teams of scorers and canvassers surround the entrance targetting which of the FIVE to transfer too…

  7. If anyone checks the link I provided they should add together DUP plus 66% of UUP transfers to see what they get..

    on this occasion it was theoretical that Diane could have beat a shinner…

  8. Guys to aid your understanding of what I am talking about, i think this article helps show the complexities

  9. To answer your question, yes, I have voted in a Belfast polling booth not that that makes any difference. You’ve also voted in one, yet describe South Belfast above as a 5 person seat, when it elects 6 so I’m not sure that voting there gives either of us any more accurate knowledge.

    The article you link to above also doesn’t tell the full story. The point here is that balancing *is* important and in situations where a party’s support is borderline, it can make the difference between getting an extra seat and losing out on one. However there is also a critical mass beyond which a party is guaranteed to win a certain number of seats provided that they ensure that all their voters transfer to all their candidates.

    In the specific instance of the 2007 election this is exactly the type of situation where SF had the critical mass needed to achieve 5 seats simply by transferring internally (the order in which people transfer is absolutely irrelevant provided they transfer across all 5.)

    Put it this way in 2003 SF had 4.55 quotas, that’s well short of 5 seats. The 4 Unionist candidates between them had 4738 votes with the quota being 4694. Thus Unionists had a quota on that occasion and thus if Unionists transferred sufficiently it was irrelevant whether SF balanced or not.

    In 2007 SF had 4.9 quotas but on that occasion there were also two other parties People Before Profit and Republican Sinn Fein standing who took votes away from them. In subsequent counts SF got 0.16 quotas from those candidates, so SF in 2007 effectively had 5.06 quotas, half a quota more than the previous election in other words. Unionists had fallen to 0.87 quotas from a quota at the previous election so they were short and all SF had to do was avoid transfer leakage to win their seat. Diane could only have beaten a shinner as you claim if SF candidates hadn’t transferred to each other.

    You asked who SDLP votes transfer to if anyone. Well in 2003, of 1402 SDLP votes transferred, 1216 went on to SF, that’s quite a large amount which in this case would have only widened SF’s advantage.

1 2