General Election 2017: LucidTalk polling plans and Northern Ireland seat predictions
So here we go again! No-one saw it coming – a UK General Election on June 8.
After successfully forecasting the Northern Ireland Assembly election to within 1%, we're now once again getting our plans and resources together for our polling projects during the upcoming Westminster election campaign. The election will be six weeks from Thursday – so taking into account this timescale, LucidTalk (LT) has scheduled three pre-election NI-Wide polls during the campaign period on:
- 27th – 29th April (60 Hours): Results available – May 1
- 15th – 17th May (60 Hours): Results available – May 18
- 1st – 3rd June (60 Hours): Results available – June 4
For each Poll-project we will be targeting our now nearly 8,000-member representative online NI Opinion Panel, and from the resulting several thousand expected responses, a robust 2,580 NI sample will be collated – structured as a balanced-representative sample of NI in terms of all key demographics e.g. gender, community, age, residence area, etc. This will comply with all professional polling and market research standards for accurate representation of opinion.
Plus, as well as this general NI wide polling of LT’s representative online Opinion Panel, LT will also be carrying out ‘deep polling’ in key constituencies by specifically targeting the sub-sectors of the LT Opinion Panel resident and voting in those constituencies (representative sample sizes of at least 500 in each constituency).
These ‘key’ constituencies will be: North Belfast, South Belfast, East Belfast, South Antrim, Fermanagh and South Tyrone, and Upper Bann. With these separate ‘deep polls’ in each of these key ‘marginal constituencies’ we’ll be able to accurately track trends, who’s winning, and make firm predictions/forecasts e.g. probability of Naomi Long winning East Belfast etc.
We will also be using our systems based NI Westminster Election Seat Predictor model (last used very successfully for the GE 2015). This computer based predictor model is constantly being improved and enhanced, and we’re again confident that it will be able to closely predict this upcoming Westminster election in NI.
The model uses the last three major NI election results, and the most recent LT polling, as data input to the model – the model then predicts likelihood in % terms, of seats being won by the various parties, broken down by the 18 Westminster constituencies. We’ve run the model, using the 3 most recent NI elections and LT polls (all 2017 LT polls – i.e. pre the 2017 NI Assembly election), and it generated the results shown in the enclosed table.
This will be our ‘base table’ (used as a starting point for this Westminster election), and we will re-run the seat predictor model again after each of the polls identified above (i.e. using that polls data as the main input to the model) – this will enable tracking of trends up or down for each of the political parties in each of the 18 NI Westminster constituencies as the campaign progresses, and therefore the seat predictor model should also get ‘more accurate’, and up-to-date, as the campaign progresses, as it will be using/analysing the latest poll data.
So what does the model currently show? Interestingly, the model predicts that Sinn Fein are now the narrow favourites to re-capture the Fermanagh and South Tyrone seat from the Ulster Unionists – they are rated at 55% for this seat. This isn’t surprising, taking into account that our recent 2017 pre-NI Assembly election polls, and the election itself, all showed and produced a strong Sinn Fein performance.
Remember the data from our polls and the recent election are the key inputs into our computer-based seat predictor model. However, taking into account the Sinn Fein surge in recent NI Assembly election we were surprised the seat predictor model didn’t predict Sinn Fein for Fermanagh and South Tyrone at 60% plus. However, this rating may change following our polls during the campaign. Perhaps what’s keeping the Sinn Fein rating for this seat at only 55% is that we also have an incumbency weighting built-in to our predictor model – incumbents (in this case the UUP's Tom Elliott) always have an advantage in Westminster first-past-the-post elections.
East Belfast is showing as a DUP hold, but at a probability of 65%. This is again lower than expected for an incumbent MP (Gavin Robinson) and for someone from NI’s largest party. However, the model has probably been impacted by the Alliance Partys strong showing in the recent 2017 NI Assembly election and in recent LT polls.
Alliance are the main challengers for this seat, and will have a chance if party leader Naomi Long is the candidate – a 65% DUP win probability is OK, but it’s by no means certain. Again of course, this win-probability score may change following our upcoming polling. Alasdair McDonnell, who is defending the marginal South Belfast constituency for the SDLP, comes in at 60% probability. The model reflecting that, yes, he’s the current MP, but is vulnerable to a strong Unionist candidate, plus the further seat predictor model forecasts for this seat will obviously be dependent on whether there are election pacts between certain parties (this also applies to other seats).
Not surprisingly, nine of the 18 seats are showing win-probability scores of 100% for the named political party (all of them the incumbent party) e.g. North Antrim for the DUP, West Belfast for Sinn Fein etc. Unless something dramatic happens we don’t see these scores changing for any of these nine seats – although North Down is dependent on whether the sitting MP Sylvia Hermon runs again, but at this stage the view is that she will.
So watch out for the results from the next NI Westminster seat predictor model to be published between May 3 and 5, following our first scheduled pre-election poll.
Bill White, is Managing Director of Belfast based LucidTalk Polling and Market Research. You can follow LucidTalk on Twitter at @LucidTalk.