Copyright sluggerotoole
For three of the four Executive parties the misery continues – with the UUP the only one escaping the consequence of belonging to an Executive which gives its members no power to deliver but only to obstruct. The structures were designed to force the parties to pull together instead they are hanging apart. The last LT poll before the restoration of the Executive gave its current four members a 83%. This is the story since: Sinn Féin is down 6% since that October ’23 poll, with the DUP and Alliance down 5% each. Only the UUP is smiling, up from 4%. In the same time period support for the parties outside the Executive (and in the case of Aontú outside the Assembly) has rocketed from a combined 14% to 32% today. The normal rules of politics (or the rules of normal politics) would say that sooner or later something has to give. But…… Let us now turn to the individual party scores. As you can see the party polling charts have been changed this time. As we get closer to the Assembly and Council elections in eighteen months’ time, I thought it would be interesting to take a longer time perspective than just the latest 12 months. LucidTalk normally conduct a poll every quarter, with an extra one sometimes coming just before an election. I have excluded their polls for the Westminster election, where voting intentions differ from Assembly elections due to tactical voting. Each party chart now shows the average for LucidTalk polls over the 12 months before the last Assembly election, and for the following two 12 month periods. Then we see the result of the August poll, and finally today’s figure. In between are the actual vote shares achieved at the Assembly and Council/Local Government (LG) elections. Bear in mind that Council elections are not always directly comparable with Assembly elections since some people chose to vote differently for personal reasons or due to the presence of a popular Independent, or are forced to vote differently if their preferred party has no candidate in their area. After the party poll charts you will find constituency level detail of what this could mean for the DUP and Sinn Féin, and this time I have added the Alliance party. Can I remind everybody that each individual figure is subject to a possible margin of error of up to 2.3%. And because numbers have been rounded to the nearest 1% not to invest too much reliance on a movement of only 1% unless it appears to be part of a trend. Seats won 2022: 27 My central projection based on this poll would be 25 or 26 seats. Apart from West Belfast and Fermanagh South Tyrone all Sinn Féin seats have healthy padding which should keep them safe even if the margin of error meant their current support level was overstated by the maximum 2.3%. Council seats are a different story. 25% would probably see them losing around a couple of dozen councillors. Details for the vulnerability/safety level of individual Assembly seats for SF, DUP and Alliance are given after the party scores. Seats won 2022: 17 The central projection on the May swing is 14 to 16. The very worst case is probably 13 and the very best 18 if the margin of error is out to the maximum either way. Somewhere around 10 council seats could be lost. Again, see below for individual seat vulnerabilities. The party may breath a sigh of relief that the decline has stopped, and hope they have reached rock bottom. Seats won 2022: 25 Seats held now: 26 Notional gains from boundary and candidate changes: 2 If DUP suffered no vote loss it could expect to win 28 seats at the next election. My projection based on this poll would be at the 13 to 18 seats. I don’t think that the penny has dropped about just how bad these figures are for the DUP. There seems to be the assumption that there will be a big swing back to the DUP in time for the election, as in 2022. There are three reasons to avoid such complacency. Firstly, the swing back to the DUP in 2022 is probably mainly down to the DUP’s appeal to unionists to unite behind them to prevent Sinn Féin becoming First Minister. The fact that the DUP failed to achieve that objective may well undermine a repeat of that appeal in 2027. And if the TUV maintain their polling share up to the election that will further undermine the credibility of that messaging. The DUP may well be obliged to campaign on the proposition that it is vital to keep the Assembly and Executive functioning. Probably a harder sell. Secondly, the DUP vote in 2022 gave them a bonus of six seats over and above their vote share, while the TUV brought home six fewer than their proportion of the vote. That advantage disappears on the current polling figures. Indeed, looking at the individual constituency figures it is possible that the bonus could be for the TUV, in part or in whole. Thirdly, the DUP will be defending 9 constituencies where they hold two seats plus another notional second seat in North Antrim. Two seats are always more difficult hold than one, since a perfect balance between two candidates is almost impossible to achieve. The one thing that the DUP has going for it, and it could well be worth a few seats, is that that it has a stronger, deeper and more widespread party organisation than the TUV. But even here there is a danger. The polling figures hold out the prospect of major losses from the DUP to the TUV in the Council elections, which will be held on the same day as the Assembly. Although it would come too late to assist the TUV in the 2027 Assembly election, that would degrade the DUP machine in the future and offer the TUV the opportunity to put down deeper roots. Seats won 2022: 9 Notional loss from candidate change: 1 (North Antrim without Robin Swann’s large personal vote.) If the UUP made no vote gains it could expect to win 8 seats It is too soon to say whether this poll’s 12% is a one-off statistical blip or the start of an improvement. My central projection would have been 8 seats without this change, but I would now call it 8 or 9. Seats won 2022: 1 So far, the TUV is holding its recent polling gains. My projection from this poll is 8 to 15 seats. Again, you can see the possibilities in individual constituencies below. If it can manage to run sufficient candidates in the right places (which it failed to do in 2023) it could gather a major haul in the Council elections. Seats won 2022: 8 My central projection would be 8 or 9 from this poll. The party can absorb a lot of extra votes in seats it won with low first preference quotas last time before it begins to add significantly to its current toll. The very best case would 11 if the margin of error has understated its support by the maximum 2.3%, and the worst probably 7 if it has overstated it by 2.3%. The party might make a handful of Council gains, but this could be limited by the propensity of some SDLP voters to back Independents in Council elections. Seats won 2022: 1 Notional loss from boundary change: 1 If PBP made no vote gains it could expect to win 0 seats Central projection 1. The poll suggests that it has a high chance of taking the fourth Sinn Féin seat in West Belfast, although the SDLP could also be in the running. Seats won 2022: 0 My projection from this poll is 0 or 1. They will be delighted to be back in contention. Seats won 2022: 0 Central projection 0. We can’t know for sure until LucidTalk publish their detailed tables in few days’ time, but I would guess that the reduction in this element is due to a movement of unionists, who are normally the biggest proportion of these poll figures, into the bigger parties. It is highly unlikely that the unionist total share has risen by 2% points. As ever, this month’s totals await the LucidTalk tables. VULNERABILITY/SAFETY OF DUP, SINN FÉIN AND ALLIANCE SEATS The tables below show the swing in support from the last election which would be required to put a seat in danger. For a brief explanation of how swing is calculated please see the bottom of this article. Swing is normally used in the analysis of First Past the Post elections, but the peculiar nature of Northern Irish elections makes it useful also for its Single Transferable Vote system of proportional representation. NI elections function as a blessed trinity – three elections in one. A unionist election in which Unionist parties compete with other unionists, with only a limited interaction outside the designation. A nationalist election with the same characteristics. The vast majority of unionist and nationalist voters stick within their own designations for the purposes of transferring – many even being willing to see their least favourite party gain a seat rather than transfer to a ‘lesser evil’. The concept of swing therefore works well between the two biggest parties in each of those designations – especially when, as now, the support for smaller parties within those designations is quite stable. The third election is more messy. It is both a competition between Other parties and with a significant overlap into the two other designations. Although a small majority of Other voters prefer to transfer within the designation they are markedly more willing to transfer outside their designation than either nationalists or unionists. However they are overwhelmingly the preferred option for those nationalists and unionists who do transfer outside their designation. When we want to look at what swing it would take for the SDLP, SF, the TUV or UUP to take an Alliance seat we cannot look at the two parties in isolation. For example, the swing from Alliance to the SDLP since the last Assembly election was 1.4% in the LucidTalk’s May poll, but if we used this figure we would get a highly misleading result. That is because in all the constituencies where the SDLP would be the principal nationalist challenger to Alliance they are reliant on transfers from Sinn Féin. But the poll shows the Sinn Féin vote down, which would mean fewer transfers to the SDLP. Instead, we need to look at the swing from Alliance to the total of nationalist parties. In May that came in at 0.1%. Balancing between candidates of the same party, and transfer patterns also play a role, which is why you see more than just red and green on the charts below. Also bear in mind that the swing is never identical in all constituencies. There are always significant deviations around the average. As we can see the combined total for these two parties is highly stable. They rise and fall in direct relationship with each other. We cannot calculate the precise swing for this poll until we see the LucidTalk tables. The chart therefore shows the swing as it was in May. The October figure will be about 0.5% points lower. You can visualise that the swing somewhere nearer to 4%. This chart shows the degree of vulnerability of either the second, or the only seat won by the DUP in each constituency at the last election to the TUV. Also included is the seat in North Down won by the Independent Alex Easton which he gifted to the DUP upon his election to Westminster. There are two other constituencies where I have deemed there to be an extra notional DUP seat. West Belfast, where recent boundary changes would have almost certainly given a seat to the DUP instead of People Before Profit had they been in force in 2022, and North Antim where the DUP would have taken a second seat had it not been for the extraordinarily high personal vote for the UUP’s Robin Swann who will not be standing there again since his election as MP for South Antrim. The boundary changes in West Belfast were expected to increase the unionist share of the vote, and this proved to be the case at last year’s Westminster election. Appling that increase to the 2022 result and reducing all other candidates proportionately whilst maintaining transfer patterns produced a notional DUP win at the expense of People Before Profit. As well as being vulnerable to a swing to the SDLP, the fourth SF seat is also vulnerable to a swing to PBP. This poll suggests that the swing from Sinn Féin to PBP is about 0.5% higher than it was in May. You should visualise the current swing somewhere around 2%. This poll suggests that the swing from Alliance to Green is around 1% higher than it was in May. You should visualise the current swing somewhere nearer to 2%. Because Alliance is down 1% in this poll and the nationalist total looks to have fallen equally, the current swing is probably close to the same as May. There are interesting anomalies which we can see above. The Alliance second seat in Strangford is its most marginal. Even the smallest swing to nationalist would see it fall – but probably not to the SDLP. A swing of much less than 2% to nationalists would almost certainly produce a TUV gain. (The TUV was the runner up, and in total unionist parties won 4.05 quotas but only three seats.) The swing would have to be almost 2% for the SDLP to be the likely winner. A similar situation would apply in North Antrim (where the UUP would not have won a seat last time without Robin Swann’s personal vote.) Again, the current swing is probably much the same as in May. Once again, we are considering the notional North Antrim result. The current swing should be visualised as around 1.5%. The current swing should be visualised as around 1.5%. AN EXPLANATION OF SWING Swing is not the increase or decrease in one party’s vote, it is a measure of the relative change in support between two parties. Three examples. If Party A gains 3% while Party B drops 3%, they have moved 6% points apart. The effect is the same as if 3% of voters had moved directly from Party B to Party A. The swing is 3%. Party F gains 4% while Party G drops 2%. They have also moved 6% points apart. The swing is 3%. Party X gains 8% while Party Y also gains 2%. Although they have both gained, they have still moved 6% points apart. So once again the swing is 3%.
