Based on these numbers, another Fine Gael-Labour coalition would fall far short
Published 03/08/2015 | 02:30
Yesterday's 'Sunday Independent'-Millward Brown opinion poll offered different messages to the various parties and groupings.
Having experienced improving fortunes in opinion polls in the first half of 2015, Fine Gael lost ground in this one, reflecting a trend that was also evident in the recent Red C opinion poll.
There were, by contrast, notable gains evident for the 'Independents and Others' grouping.
My constituency-level analysis model attempts to tease out how opinion poll figures might translate into Dáil seats.
If the trends were replicated in a general election, party seat numbers should be as follows: Fianna Fáil 41, Fine Gael 44, Sinn Féin 32, Labour 4, Independents and Others 37.
Based on this estimate, a potential Fine Gael-Labour coalition (48 seats) would be a very long way short of the number of seats required to form a government (79 seats) in the next Dáil.
A potential Fianna Fáil-Sinn Féin alliance (combined seat level of 73 seats) would be much closer, however, falling just six seats short of the 79-seat target. On these numbers, Fianna Fáil and Sinn Féin could take power with the support of a small grouping of six or more independent TDs (maybe the Independent Alliance deputies) or a small party, such as the Social Democrats.
A Fine Gael and Sinn Féin pairing would, on these estimates, come even closer to the 79-seat target (combined seat level of 76 seats), but such an alliance looks highly unlikely in the present political climate.
Fine Gael-Fianna Fáil would amount to the only alliance capable of mustering enough seats to form a two-party coalition government without needing the support of other Dáil deputies (it would have a combined 85 seats).
The potential for robust two-party coalitions has declined over the past two years due to a series of strong opinion poll figures for the 'Independents and Others'.
My model recognises that although the Irish electoral system is a proportional one, the number of seats won by parties or groupings does not measure up exactly to their first-preference vote shares, mainly because geography has a significant impact here.
First-preference votes need to be filtered through the system of Dáil constituencies (and the different numbers of seats apportioned to these).
The model is largely based on the recognition that the voting patterns associated with each party (in other words, their geography of support) tend to be consistent across most elections.
For instance, Sinn Féin tends to poll strongly in the border constituencies and in working-class urban areas, but less well in middle-class urban areas and in some rural constituencies.
The geography of support for the 'Independents and Others' grouping is less predictable than for the parties.
The model estimates what first-preference votes for the different parties and groupings would be in the different constituencies, assuming similar (proportional) changes in party vote shares in all constituencies to those being suggested by a particular opinion poll.
Where tally figures are available, the model also takes account of the boundary changes brought in by the 2012 Constituency Commission report, as well as taking account of politicians leaving or joining different parties.
This, of course, is a rough model and it cannot take appropriate account of the fact that changing support levels between elections tend to vary geographically.
A swing to, or from, a particular party may be stronger in some constituencies for a variety of reasons - such as the candidates selected, changes to local party organisations and the impact of contentious local factors.
Variations in support levels can be impacted upon as much by who is selected by other parties, as much as by who runs for the party itself.
The constituency support estimates associated with this model cannot be viewed as accurate constituency predictions, but the model is not about generating 100pc-accurate constituency predictions in any case.
The ultimate aim is to get an overall, national-level estimate of seat numbers and this is based on the proviso that an over-prediction in one constituency may be offset by an under-prediction in another constituency.
The model estimates what party seat levels would be if the national support trends evident in a specific opinion poll were replicated in an actual general election.
For a variety of reasons (such as high levels of undecided voters), the actual result of an election may vary from the figures suggested by an opinion poll, even if the polls are carried out relatively close to general election day.
The likelihood of such variation is not something that can be factored into this model.
Vote transfer patterns cannot be accounted for in the constituency support estimate figures, but I do try to control for these in my set of Dáil seat allocations.