Previously, I looked at what might be expected from next month’s Assembly election if the current polling, showing a fall in support for the DUP with a corresponding rise in support for the Alliance Party and the UUP, is indicative of public support. The forecast model anticipated that the UUP could expect to receive 14 seats under such a scenario, which would be a respectable result given that they won 16 out of 108 seats in the previous mandate. However, this does not reflect the vulnerability that the UUP face in a range of races should the anticipated rise in support fail to materialise.
To demonstrate this, the following table shows the results of the constituency forecast model if the share of the popular vote is the same this year as it was in 2016. As well as the five main parties, in this scenario the Greens and People Before Profit would only retain seats in North Down and West Belfast respectively, with Jim Allister (easily) and Claire Sugden (narrowly) expected to also retain their seats. Grey denotes that the party is a non-incumbent.
Of the 37 seats the DUP are defending (the DUP are only running two candidates in North Down this time), they are either favoured to win or in a virtual tie in 35 of them, meaning that the DUP could go from having 35.2% of the seats in the Assembly to 38.9% without any rise in support. To contrast, the UUP are only favoured in 10 of the 16 seats they are defending in this election, meaning their share of the Assembly seats could fall from 14.8% to 11.1%, even with their popular vote share remaining the same.
Sinn Féin would be expected to lose five of their 29 seats in this scenario, and the SDLP are expected to lose three seats, two of them almost certainly.
Despite being the largest party in the Assembly, the DUP look to be the party affected the least by the move to five member constituencies out of the four largest parties, with the UUP potentially the party most affected. A modest increase in support for the UUP at the ballot box at the DUP’s expense is unlikely to have a significant change in the composition of the Assembly. The UUP have to run to stand still, due to the structural advantages held by the DUP in the new five seat constituencies.
The risks faced by both the UUP and SDLP are apparent when you look at the forecast model that uses current polling rather than 2016 vote shares as a model input. The model for West Belfast has been updated to correct an incorrect model input for the Sinn Féin vote, who are now narrowly favoured to win the seat People Before Profit are targeting in West Belfast. The Greens are still expected to gain the DUP’s second seat in South Belfast.
Under this scenario, the UUP are favourites in 14 seats. But it should be noted that, whilst favoured, none of the UUP’s target seats from 10-14 are easy. To win 14 seats the UUP have to win a seat in North Antrim and South Down, take a seat from the DUP in Fermanagh & South Tyrone, a second seat in Lagan Valley and finally a seat in East Belfast.
In terms of seats that could be considered “safe”, i.e. have a 90% change or higher, the UUP only really have six, the same as the SDLP. Arguably the Alliance Party have a safer seventh seat (Lagan Valley) than either the SDLP (a second in Foyle) or the UUP (a seat in West Tyrone). The level of uncertainty for both the SDLP and the UUP is high; both could conceivably end up with more MLAs even in absolute terms than they won in 2016, but both could also face the prospect of being virtually wiped out.
On the other hand, the DUP appear to have a stronger wall of safe seats to protect themselves against a drop in support. Whilst they could end up fewer than 30 MLAs and therefore be unable to sign a Petition of Concern without the support of other parties, targets 26-28 (South Down, Newry & Armagh, and Foyle) would require very dramatic swings away from the DUP were they to lose any of them.
Despite the structural advantages that the DUP possess in five member constituencies, the UUP have a real chance to make up ground on their Unionist rivals and pave the way for a more serious challenge in a future electoral cycle. However, they need to be careful, as they are far from secure in a number of seats that they are defending.
To put it another way; you come at the king, you best not miss.
A qualified accountant and data analyst, interested in politics, economics and data. Twitter: @peterdonaghy