The 11 new district councils – projecting the 2011 votes

For us anoraks, there is both frustration and challenge at the thought of the Northern Ireland council elections next month, taking place for 11 new local government districts, on completely redrawn electoral boundaries, with the full details of the last census not yet out in sufficient detail that the enthusiast can calculate changes down to townland or city block level.

However, some of us are trying. In particular, a contributor to Bangordub’s blog identified only as “Faha”, and a contributor to the Vote UK Forum known only as “Irish Observer”, have both made a go of looking at the new districts, electoral area by electoral area, to work out what the elections of 2011 might have looked like on the new boundaries and hence what we might set as a baseline for next month.

I have taken a different approach. For each of the 11 new districts, I have done my best to calculate an overall party vote based on the 2011 votes, and then to guesstimate the likely party strengths in each district, had those votes been cast on the new boundaries. I report also the 2011 census returns on community background, since that is usually the first question people ask about the statistics.

The scores on the doors are that six of the new eleven districts have a clear Unionist majority, and in three of those six the DUP starts within a seat or two of having an outright majority. Four of the eleven have a clear Nationalist majority, and I make Sinn Fein within spitting distance of a majority on their own in two of them, and the largest party in the other two. Belfast is more evenly split: I give Nationalists 27, Unionists 24 and the Alliance Party 9 seats of the new 60-strong council, based on the 2011 votes.

I emphasise that this is not about predicting the results of next month’s elections; this is about establishing a baseline against which the election results can be measured. With that caveat, I would say that the nominal total number of seats on the new councils for each party, which would mark an equivalent performance to 2011, are:

DUP 145 (20 Lisb&Cas, 19 ND&Ards, 19 M&E Ant, 17 Belfast, 17 Ant&Nby, 15 CC&G, 13 ABC, 9 Mid U, 8 D&S, 5 F&O, 3 NM&D)
SF 115 (19 Belfast, 18 F&O, 17 Mid U, 16 D&S, 16 NM&D, 9 ABC, 9 CC&G, 5 Ant&Nby, 3 M&E Ant, 3 Lisb&Cas)
UUP 77 (12 ABC, 9 F&O, 9 Ant&Nby, 8 M&E Ant, 7 ND&Ards, 7 Lisb&Cas, 6 Belfast, 6 CC&G, 6 Mid U, 5 NM&D, 2 D&S)
SDLP 67 (14 NM&D, 13 D&S, 8 Belfast, 6 ABC, 6 Mid U, 6 F&O, 5 CC&G, 4 Lisb&Cas, 3 Ant&Nby, 1 M&E Ant, 1 ND&Ards)
Alliance 34 (9 Belfast, 8 ND&Ards, 6 Lisb&Cas, 6 Ant&Nby, 4 M&E Ant, 1 CC&G)
TUV 3 (2 M&E Ant, 1 CC&G)
Greens 1 (ND&Ards)
PUP 1 (Belfast)
Others 18

Projections for the 11 individual councils below, along with wee maps which are taken directly from the DoE site at and are therefore Crown Copyright, used here I hope with the shield of fair use.

North Down and Ards

(The whole of Ards district, and almost all of North Down district apart from the very small Cedar Grove corner which goes to Belfast – the North Down figures below are adjusted for the removal of Cedar Grove.)

PartyArds  N Downtotal%ge
DUP11,7328,991  20,723  41.8%
Cty Pshp8008001.6%

Goes down from 48 seats to 40.
61.8% to Unionist parties; 4.0% to Nationalist parties; 34.2% for the rest.
2011 census: 75.05% P, 13.12% C, 11.83% Oth/None.
Edited to add: small corrections made to Ind and Community Partnership totals.

“Irish Observer” projects 20 DUP, 8 Alliance, 7 UUP, 1 SDLP, 3 Inds, 1 Ind U.
(“Faha” hasn’t got round to this one yet.)

I’d say it’s a stretch for the DUP to get so close to an overall majority on less than 42% of the vote (they won 22 seats out of 48 in 2011), and the Greens somehow always manage to pull something out of the hat, if not quite what they wanted. So my gut feeling is that a reasonable projection of the 2011 votes to the new boundaries would give the DUP at least one less and the Greens a seat in North Down.
My call: 19 DUP, 8 Alliance, 7 UUP, 1 SDLP, 1 Green, 3 Inds, 1 Ind U.

Mid and East Antrim

(The whole of the current Ballymena, Carrickfergus and Larne districts.)

Party  Ballymena  CarrickLarnetotal%ge
DUP10,9995,356  3,374  19,729  42.0%

The three councils currently have 56 seats, going down to 40.
68.1% to Unionist parties; 11.7% to Nationalist parties; 20.2% for the rest.
2011 census: 72.88% P, 19.34% C, 7.77% Oth/None.

“Irish Observer” projects 19 DUP, 8 UUP, 4 Alliance, 3 SF, 2 TUV, 1 SDLP, 3 Inds.
“Faha” projects 20 DUP, 9 UUP, 4 Alliance, 3 SF, 2 SDLP, 2 TUV.

Larne and Carrick in particular have a tradition of voting for independents, so I think the former is nearer the mark. It would have been impressive for the DUP to get half the seats on 42% of the vote (they won 24 out of 56 real seats in 2011).
My call: 19 DUP, 8 UUP, 4 Alliance, 3 SF, 2 TUV, 1 SDLP, 3 Inds (agreeing with Irish Observer).

Lisburn and Castlereagh

(The current Lisburn district, losing about 20% of its voters to Belfast, and the current Castlereagh district, losing about 40% of its voters to Belfast as well.)

Party  Lisburn  C’reaghtotal%ge
DUP15,9126,305  22,217  47.4%

The two councils together currently have 53 seats, to be reduced to 40; but the drastic transfer of 20% of Lisburn and 40% of Castlereagh to Belfast actually means, uniquely, a net increase of councillors per voter in the remaining territory (if I have my sums right).
66.1% to Unionist parties; 15.4% to Nationalist parties; 18.5% for the rest.
2011 census: 66.90% P, 23.95% C, 9.15% Oth/None.

“Irish Observer” projects 20 DUP, 8 Alliance, 6 UUP, 3 SF, 3 SDLP.
(“Faha” hasn’t got round to this one yet.)

The SDLP surprised me with their performance in Lagan Valley in the 2011 election. I think Irish Observer is too pessimistic for them and the UUP, and very optimistic for Alliance. But the DUP getting half the seats on 47.4% looks about right.
My call: 20 DUP, 7 UUP, 6 Alliance, 4 SDLP, 3 SF

Antrim and Newtownabbey

(Simply the two current districts amalgamated.)

Party  AntrimNbytotal%ge
DUP5,210  11,947  17,157  38.6%

Goes down to 40 councillors from 44.
60.8% to Unionist parties; 22.1% to Nationalist parties; 17.1% for the rest.
2011 census: 61.10% P, 29.74% C, 9.17% Oth/None.

“Irish Observer” projects 19 DUP, 8 UUP, 6 Alliance, 4 SF, 3 SDLP.
“Faha” projects 17 DUP, 9 UUP, 6 Alliance, 5 SF, 3 SDLP.

The latter looks right to me.
My call: 17 DUP, 9 UUP, 6 Alliance, 5 SF, 3 SDLP (same as Faha).

Causeway Coast and Glens

(The whole of the current Ballymoney, Coleraine, Limavady and Moyle districts)

Party  Bmoney  Coleraine  LimavadyMoyletotal%ge
DUP5,1826,9843,191955  16,312  32.8%
SF2,3441,7594,626  1,38110,11020.3%

Goes down brutally from 68 councillors to 40.
52.6% to Unionist parties, 32.4% to Nationalist parties, 15.0% for the rest.
2011 census: 54.79% P, 40.21% C, 5.00% Oth/None.

“Irish Observer” projects 15 DUP, 9 SF, 6 UUP, 5 SDLP, 2 Alliance, 1 Ind U, 1 Ind Nat, 1 Ind.
“Faha” projects 15 DUP, 9 SF, 7 SDLP, 6 UUP, 1 Alliance, 1 TUV, 1 Ind U.

“Faha” is very optimistic that the SDLP could take 17.5% of the seats with 12.1% of the vote. Moyle in particular is a fissiparous electorate, used to voting for independent candidates, and “Irish Observer” looks to me to be closer to the mark in that regard. However, I agree with Faha that the TUV should be projected as defending a retainable seat.
My call: 15 DUP, 9 SF, 6 UUP, 5 SDLP, 1 Alliance, 1 TUV, 1 Ind U , 1 Ind Nat, 1 Ind.

Armagh, Banbridge and Craigavon

(The whole of Armagh and Craigavon districts, almost all of Banbridge minus most of Ballyward ward, and the Caledon corner of Dungannon district. The Banbridge figures below are adjusted for the removal of Ballyward.)

Party  Armagh  Banbridge  Craigavon  Dgntotal%ge
DUP5,3265,7299,974265  21,294  27.8%

Goes down to 41 councillors from 65.
56.3% to Unionist parties; 38.0% to Nationalist parties; 5.7% for the rest.
2011 census: 51.74% P, 42.95% C, 5.31% Oth/None.

“Irish Observer” projects 12 DUP, 11 UUP, 10 SF, 7 SDLP, 1 Ind U.
“Faha” projects 14 DUP, 11 UUP, 9 SF, 6 SDLP, 1 Ind U.

I don’t think you can easily get 17 Nationalist seats on 38% of the vote, and I am also struck by how close the UUP and DUP are (this being one of the former’s hidden bastions of comparative strength).
My call: 13 DUP, 12 UUP, 9 SF, 6 SDLP, 1 Ind U


(The whole of the current Belfast district, plus about 20% of Lisburn, 40% of Castlereagh, and the very small Cedar Grove corner of North Down. I have tallied the two parts transferred from Lisburn, Dunmurry and Drumbo, separately since they are dramatically different but – with apologies to my friends form there – have treated the transferred parts of Castlereagh as a block.)

Party  Belfast  D’m’y  D’bo  ex-C’reagh  C Grtotal%ge
SF28,2345,946595  34,280  31.2%

Old Belfast Council has 51 seats; this will increase to 60.
35.7% to Unionist parties, 46.9% to Nationalist parties, 17.4% for the rest.
2011 census: 42.47% P, 48.82% C, 8.71% Oth/None.

“Irish Observer” projects 20 SF, 17 DUP, 9 Alliance, 7 SDLP, 6 UUP, 1 PUP.
(“Faha” hasn’t got round to this one yet.)

It may seem odd to give the DUP 28.3% of the seats on 24.0% of the votes, but I’m with “Irish Observer” here. There’s still a small systemic bias against SF, in that population drift even in the short space of time since the new wards were drawn up has left them with a relatively greater number of votes needed per councillor elected. (Meaning that the Nationalist electorate is increasing, but increasing most in places where they were already strong; while Protestant population drift is taking voters out of what were already low turnout areas; I’m not suggesting any deliberate design effect in the process, which was transparent, democratic and professionally implemented.) So I’m projecting:
My call: 19 SF, 17 DUP, 9 Alliance, 8 SDLP, 6 UUP, 1 PUP

Mid Ulster

(The whole of the current Cookstown and Magherafelt districts, and almost all of Dungannon apart from the Charlemont triangle, which goes to the new ABC council.)

Party  Cookstown  Dungannon  Magherafelt  total  %ge
SF6,1358,2049,732  24,071  40.6%

Goes down from 52 seats to 40.
38.2% to Unionist parties; 56.4% to Nationalist parties; 5.4% for the rest.
2011 census: 33.46% P, 63.77% C, 3.77% Oth/None

“Irish Observer” projects 18 SF, 9 DUP, 6 SDLP, 5 UUP, 2 Ind Nat
“Faha” projects 19 SF, 8 DUP, 6 SDLP, 6 UUP, 1 Ind Nat.

Those TUV votes are rather thinly spread but will transfer back to the DUP, so the Unionist parties are on 38% rather than 34% of the vote collectively – which should mean a fifteenth seat somewhere. And that number of votes for independent candidates should deliver two of them.
My call: 17 SF, 9 DUP, 6 SDLP, 6 UUP, 2 Ind Nats

Fermanagh and Omagh

(The whole of the current Fermanagh and Omagh districts)

Party  Fermanagh  Omaghtotal%ge
SF11,27610,353  21,629  40.9%

Reduced from the current 44 councillors to 40.
37.5% to Unionist parties; 53.8% to Nationalist parties; 8.7% for the rest.
2011 census: 33.08% P, 64.23% C, 2.69% Oth/None

“Irish Observer” projects 18 SF, 9 UUP, 6 SDLP, 5 DUP, 2 Ind Nat.
“Faha” projects 19 SF, 9 UUP, 6 SDLP, 5 DUP, 1 Ind Nat

Faha sometimes overlooks independents; I’m with Irish Observer on this one.
My call: 18 SF, 9 UUP, 6 SDLP, 5 DUP, 2 Ind Nat (same as Irish Observer).

Derry and Strabane

(The whole of the current Derry and Strabane districts)

PartyDerry  Strabanetotal%ge
SF  14,011 6,834  20,845  35.5%

Down from 46 councillors to 40.
24.3% to Unionist parties; 67.7% to Nationalist parties; 8.0% to the rest.
2011 census: 25.40% P, 72.16% C, 2.44% Oth/None

“Irish Observer” projects (with uncharacteristic vagueness) 16-20 SF, 10-13 SDLP, 7-11 DUP, 1-2 UUP, 0-1 Ind
“Faha” projects 15 SF, 15 SDLP, 8 DUP, 2 UUP.

I make SF 6% ahead, which surely reduces the chance of the SDLP tying them for seats. And with all those votes for independents floating around, I would have thought there must be a chance of one making it though. (In 2011, two were elected in one of the Strabane DEAs.)
My call: 16 SF, 13 SDLP, 8 DUP, 2 UUP, 1 Ind.

Newry, Mourne and Down

(The whole of the current Down district; the whole of the current Newry and Mourne district; and the Ballyward corner of Banbridge.)

Party  DownN & M  B’wardtotal%ge
SF5,806  17,877260  23,943  36.4%

Goes down from 53 seats to 41.
23.4% to Unionist parties; 67.3% to Nationalist parties; 9.3% to the rest
2011 census: 23.91% P, 72.32% C, 3.77% Oth/None.

“Irish Observer” projects 15 SF, 15 SDLP, 5 UUP, 4 DUP, 1 UKIP, 1 Ind Nat.
“Faha” projects 16 SDLP, 15 SF, 4 DUP, 3 UUP, 2 Ind Nats, 1 UKIP.

I have to say it’s difficult for me to imagine SF failing to convert a 5.5% lead in the overall vote into a margin of at least two council seats, let alone falling behind. Those numbers look to me more like:
My call: 16 SF, 14 SDLP, 5 UUP, 3 DUP, 1 UKIP and 1 Ind Nat.

There is plenty of room for disagreement, and in any case the real verdict on the numbers above will be rendered by the voters on 22 May.

  • Great stuff Nicholas,
    I’ll have Faha’s analysis for Lisburn and Castlereagh up tomorrow. He is completing one a week in the lead up to the May elections and saving Belfast until last 😉
    One of the recurring themes of his articles is the repidly falling nationalist turnout and whether this trend will continue.

  • Seamuscamp

    SDLP total for Armagh etc overstated by 10000?

    Very interesting article. Thanks

  • Brian Walker

    Formidable piece of work Nicholas by you and others. It puts the so-called mainstream media to shame – and free at the point of use!

  • Rab12345

    Failed Alliance politician predicts Alliance success. Yawn.

  • Thanks to three of you. Good catch, Seamuscamp, I’ve corrected the error.

  • Gopher

    Interesting NI21 figure no where in the calculations.

    My questions for the experts on whether we will see marked effect from

    1/ Higher Quotas
    2/ More runners
    3/ Migration out since the crash
    4/ Higher turnout or Lower turnout

    On the basic figures it seems my assumption is correct the boundary changes to Belfast favoured Alliance rather than the SDLP. Although the figures are Lab figures if the unionist vote dont leave Alliance they are heading to overtake the SDLP in Belfast.

  • Larne man

    Can’t see any chance of 7 Nat seats in Lisburn and Castlereagh. Four seems most likely looking at the DEA’s.

  • Wabbits

    Nicholas your figures for Derry/Strabane are pretty accurate. My reckoning is that this will be interesting on two counts. 1. Overall turnout (it will probably be on the low side). 2. How the Strabane vote pans out.

    I would say it will end up Sdlp 14 or 15 and SF 16 or 15. With Unionists sharing the other 10. I don’t expect to see any independents scoring enough first preferences to carry them through this time. They really would need to be passing the quota on first preferences or extremely close to it.

    I would imagine the quota, even on a fairly low turnout of say 50% would see most (admittedly not all) candidates needing to be over or near the 1100 mark by the end of the process to be elected. That is why the turnout is vital.

  • Larne man

    The idea of low unionist turnout doesn’t appear to bear much to reality; in practically every council the unionist vote appears to either match or even significantly overperform the protestant community background %. The one obvious exception is Belfast. Perhaps this is due to differential turnout between republican and loyalist areas of the city, along with the high protestant Alliance vote in East Belfast? Can anyone explain this?

  • Gopher – no figures for NI21 because these are the 2011 votes and they weren’t around yet!

    Larne Man – I’m calling 2 Nats in Killultagh, Lisburn Town North and Castlereagh South, and 1 in Downshire West. Yes, three-quarters of SF’s vote in Lisburn, and 40% of the SDLP’s, is transferred to Belfast, but that still leaves a fair bit.

    Wabbits – I accept that most of those votes for independent candidates in Strabane in 2011 will make their way back to the SDLP one way or the other. That still leaves the SDLP behind SF in votes.

    Effect of increased quotas: makes it tougher for candidates with just one strong area of support (which normally means independents). Otherwise hits everyone equally.

    Demographics: Don’t forget the age differential (the figures I give include under-18s as well) and the None/Other category.

    Turnout affects parties as well as communities. My gut feeling in 2011 was that the SDLP voters in particular just weren’t motivated. Part of any election campaign is maximising your own supporters’ presence at the ballot box. (Except here in Belgium, where voting is compulsory.)

  • GEF

    Nicholas Whyte, many thanks for this post. The info is extremely useful coming up to the election in May 2014.

  • redstar2011

    Really thought Belfast would be a lot tighter

  • Gopher

    I thought it would be be also on paper it looks like the higher quotas will bring the micro party and indie transfers into play.

    The interesting thing is the twin jewels in the SDLP crown Derry and Down looks like they be lost this election. A SF victory in this election and you would really have to fear for Durkan. The SDLP will have to fight like tigers or face extinction in ten years

  • Morpheus

    Very interesting analysis, well done.

    So much has happened since 2011 so I suppose the obvious question is how do people think events since then – the performances of the parties, the notorious leaflets ,the flag protests, the emergence of NI21, the recent ‘revelations’ in The Alliance Party, UPRG campaigning for the DUP, Willie Frazer, Welfare Reform etc. – will impact the above analysis.

    Another interesting factor will be the addition of those who were too young to vote at the last local elections into the voter pool. Will they have the ‘fresh attitude’ that Michelle Obama talked about at the G8…

    …or will they be apathetic?

  • mjh

    Hi Nicholas
    That was an impressive – and impressively well-presented piece.
    Using a different method I got the same totals as you for TUV, PUP and UKIP. I got 5 fewer SF (110), 2 fewer DUP (143), and 4 fewer Independents (14). But I had 5 more SDLP (72), 2 more Alliance (36), 1 more UUP, and 1 more Green (2).
    What struck me forcibly is that this cannot be a purely mathematical exercise. It is impossible to avoid having to use one’s judgement at some points.
    My method was to start with the total share of seats won by each party in a new council area at the last election, adjusted to eliminate the differences in electorate in the old District Electoral Areas. This served as a rough proxy for the efficiency in party’s vote – where their ability to attract transfers or the geographical spread of their support might result in them winning a larger or smaller share of the votes than their share of the 1st preference vote.
    Adjustments were then required to account for Independents, or small parties that might not win a seat because the new quota was higher. This involved judgement based on known transfer patterns.
    In some cases the calculations suggested that more than one outcome was almost equally possible. For example a significant difference between a party’s vote share and seat share could suggest that a party had been “lucky” or “unlucky” under the old boundaries, and that this could be different under the new. Or the calculation could come to, say, 1.5 seats – indicating that it could go either way (with a consequent impact on another party.)
    So here is my back projection of the 2011 results onto the new boundaries. I show a range where there Is reason for doubt – with my best call in brackets.
    Antrim & Newtownabbey
    DUP 15-16 (16), UUP 8-9 (9), All 6-7 (6), SDLP 3-4 (4), SF 5

    Armagh, Banbridge & Craigavon
    DUP 11-12 (11), UUP 11-12 (12), All 1, SDLP 6, SF 10, Ind 1

    DUP 17-18 (17), UUP 4-5 (5), All 7-8 (8), SDLP 9-10 (10), SF 19-20 (19), PUP 1, Ind 0-1 (0), eirigi 0-1 (0)

    Causeway Coast & Glens
    DUP 13-14 (14), UUP 7, All 1-2 (1), SDLP 5-6 (6), SF 7-9 (9), TUV 1-2 (1), Ind 1-3 (2)

    Derry & Strabane
    DUP 7-8 (8), UUP 1-2 (1), SDLP 14-15 (14), SF 15-16 (16), Ind 1
    Fermanagh & Omagh
    DUP 6-7 (7), UUP 8, SDLP 5-6 (5), SF 17, Ind 3

    Lisburn & Castlereagh
    This is the council with the greatest uncertainty in the calculation. I chose not to attempt to split the old DEA’s. Although this introduces more uncertainty into the calculation I judged it preferable to attempting to guess where each party’s votes were located in a DEA divided between two new councils. However we can see that with slightly different methods we get different results. Nicholas has SF on 3017 votes taking 3 seats – whereas I have SF on 1927 votes and no seat with a possibility of 1. In this respect Nicholas’s greater knowledge has probably resulted in a better judgement.

    But for the record I had:
    DUP 21-22 (22), UUP 6-7 (7), All 6, SDLP 4, SF 0-1 (0), Green 1

    Mid & East Antrim
    DUP 18-19 (18), UUP 8-9 (9), All 4-5 (5), SDLP 2-3 (2), SF 2-3 (2), TUV 2, Ind 2-3 (2)

    Mid Ulster
    DUP 8-9 (8), UUP 6-7 (7), SDLP 6-7 (7), SF 16-17 (17), Ind 1

    Newry, Mourne & Down
    DUP 3, UUP 4-5 (5), All 1, SDLP 13-14 (13), SF 14-15 (15), UKIP 1, Green 0-1 (0), Ind 1-2 (2)

    North Down & Ards
    DUP 18-19 (19), UUP 8-9 (8), All 8, SDLP 1, Green 1, Ind 3

    I think it best to express the uncertainty where it exists to avoid claims that a party has won or lost a “notional” seat that they may never “notionally” have had.

  • mjh,

    I think our differences are probably within the realm of reasonable variation between reasonable people. I do note that you are consistently giving the SDLP slightly better results and SF slightly worse than I am. We’ll see; my feeling is that in 2011 the SDLP slipped behind in a lot of places – they were the only party that consistently lost seats everywhere – and my projection reflects that. If they can find the ignition key, of course, it’s a different matter.

    Lisburn/Castlereagh gave me by far the biggest headaches, as I think is apparent. Not all of Dunmurry Cross goes into Belfast, so that accounts for most of the discrepancy between our tallies right away. SF got 6860 votes in Dunmurry Cross in the last election; I transfer 5946 of them to Belfast keeping 914 in Lisburn. But you and I differ by 1100 rather than 900, so I’m not sure where the other 200 come from. (Possibly Castlreagh West.) But I’m prepared to accept that there are other ways to crumble this cookie.

  • Gingray

    Great work Nicholas – I know Faha over in Bangordub thinks the DUP pulled a fast one and got their way over the councils, whats your own opinion on this?

    Should be an interesting election at any rate, do you come back for them?

  • Count Eric Bisto von Granules


    I heard this theory on the DUP getting their own way over councils before. On the face of it, it would be hard to disagree, but given that it is recognised that SF have a degree of strategic nous and an ability to play the long game, could it be that this is what SF want.

    Presumably, they ‘got’ something from the DUP in return for this, but if you look at ‘Armagh, banbridge & Cragavon’ together with ‘Causeway Coast & Glens) then you have 2 councils where catholic population is 40%.

    Given another census cycle, depending on the age profile of the unionist population, it is likely that the Catholic proportion will be nearing 50%. Once this happens the discrepancy between Catholic and Nationalist votes will disappear as loaned votes will go and nationalist parties will be able to show a ‘prize’ for increased turnout.

    It should also be figured with the continued greening of Belfast that 7 out of 11 councils could show a nationalist majority and this is not to be under estimated in terms of psychological impact on the 100th anniversary of the NIs foundation.

    Or am I ascribing an undeserved intelligence to SF in this?

  • Hi all,
    Faha’s 9th council analysis just up as promised here:
    For brevity his prediction is: SF 0 SDLP 5 Alliance 8 Green 0 UUP 6 DUP 21

  • mjh

    Hi Nicholas

    Yes my back projection calculates that the SDLP won more “notional” seats than your, and that SF took fewer. But I have not been able to identify any consistent pattern which might suggest why this should be.

    Lisburn is the biggest discrepancy, as already discussed, where my approach of not splitting DEA’s gives 3 fewer SF than yours – but the SDLP is unaffected.

    But there is only one council district where I have one more SDLP than you at the expense of SF – Mid & East Antrim – where you have 3 SF and 1 SDLP, while I have 2 each.

    They actually won 3 seats each in 2011. SF does have a very slight advantage in vote share of 5.9% compared to 5.8%. Weighting the seats won in each DEA to eliminate differences in the size of the electorate gives SF 2.4 of the new seats and SDLP 1.9. Which I rounded to 2 each.

    In Mid Ulster there was only 1 Independent elected in 2011. I kept the call at one, whereas your method gives Independents 2. Hence I have 1 more SDLP than you.

    Similarly in Causeway Coast & Glens I judged that none of the 4 Independents was likely to have been elected in Moyle. That is because I looked at their vote as individuals. Your approach may have judged that transfers between them would carry one through. (A perfectly valid point of view.) This gives me 1 additional SDLP.

    I have three other Councils each with 1 more SDLP than in your tally. In these cases it appears that my approach has calculated 1 more CNR seat than yours and that the beneficiary has been the SDLP. In Antrim and Newtownards this gives me one fewer DUP than you, in Belfast 1 fewer Alliance and in Derry and Strabane 1 fewer UUP.

    My main point is that on a Northern Ireland total we have very similar outcomes – but that we need to exercise caution when we analyse the eventual 2014 results at individual council level.

    By the way, Bangordub. I find your site very interesting. You have my admiration for all the work involved. So far I have been too lazy to comment on more than Slugger.

  • Morpheus

    Have you got an email address that you wouldn’t mind sharing bangordub?

  • Charles_Gould

    No party has overall control of any council by all the three number crunchers.

  • Gopher

    Charles I m not quite sure all our number crunchers have approach the subjective from a dispassionate position. We have seen the biggest housing boom and bust beginning in the last election cycle, we have new parties and ones like the greens testing their novelty. We have the imponderables of the UUP and Alliance vote and no one is sure the SDLP have touched bottom and add to this new boundaries. It’s not the most precise science.

  • Charles_Gould


    Nicholas Whyte has clearly stated that his projections are just a benchmark for comparison, not a prediction.

    The SDLP under Alasdair McDonnell’s leadership and strategy has put a lot of work in raising the profile of a number of SDLP candidates in a number of key targeted areas. I am hoping for a good performance.

    Not quite sure how Alliance are going to do now that they have taken a more unionist-hostile approach.

  • Morpheus

    Anna Lo stating her personal preference while reinforcing her commitment to the GFA and the principle of consent is a more unionist hostile approach? Hardly.

  • Morpheus,
    Sure, it’s

  • Charles_Gould

    There is currently a Green in Castlereagh it is interesting to see that none of the number crunchers forecast his survival.

  • mjh


    I have expressed no forecast on the chances of a Green being elected to Castlereagh in 2014 – and neither has Nicholas.

    It is simply that we have both concluded that had the 2011 election been conducted on the new boundaries it would not have happened then. (Although please note that for my part I said that there was an element of doubt. I called it Green 0 or 1 with 0 being more probable.)

    Whether a Green is returned this year will depend on how voters’ opinions have developed in the last three years.

  • Charles_Gould

    Thanks mjh sorry for my sloppy use of the word “forecast”. I guess yours is a “pastcast”!

  • mjh

    Love your new word, Charles. Very useful.

    Maybe Mick will be running a thread on it in three years time ☺

  • Charles_Gould

    I see that Ballymena is being merged with two towns that have a lot of Alliance voters – Larne and Carrickfergus.

    It will be interesting to see if this has an effect on Ballymena Alliance prospects – will Alliance start to compete more vigorously across the entire district? Will the higher profile of the party extend its brand?

    (I always thought Alliance could do better in Ballymena if they tried a bit harder).

  • Coll Ciotach

    Alliance in Larne council depends on nationalists voting for them as the least worst option, in Ballymena there are alternatives so my answer would be no, and also expect Sinn Fein to get a gunk in the Glens before the day is out

  • socaire

    Sometimes you would get the impression that ‘Sinn Féin’ think they have their market made and have no reason to court the CNR population any more. This could be dangerous at a time when people are becoming a bit apathetic about elections. I mean why bother to elect anybody to any thing when they are powerless – just another well paid tier of ‘govt’ to temporarily placate the natives. As Coll Ciotach says, a gonnc could be in store in more places than the Glens

  • Charles_Gould

    I am not so sure about these predictions of problems in the Glens.

  • Gingray

    How do you see things panning out in Lisnasharragh? I know your analysis is based on previous results, but do you have a personal opinion on how it may pan out?

    Count, I can see what you are saying about SF potentially playing a long game, and I think it would be hard to have seriously messed the review around to the degree that Faha is hinting at. I think that with 11 councils what they have come up with is as fair as possible.

    Against that, does anyone have any reasonable justification for Castlereagh and Lisburn? The Galwally map on the council page is quite funny, showing the lengths that people have went to to place forestside in Lisburn.

    I wonder, for any of you who made submissions for Local Goverment reform, did you have Castlereagh in Belfast or another council?


  • Charles_Gould


    Don’t you think it odd that Carrowreagh is in Lisburn? It is a few hundred metres east of Stormont and seems part of Belfast not Lisburn. I also spotted the odd shaped line round forest side – that looks so unreasonable I can’t see it lasting long.

  • Charles_Gould


    From a Ballymena point of view, it seems odd to be joined with Carrick and Larne – over the hills and far away.

    Ballymena – Antrim is a much more obvious and natural pair with transport linkages.

    Instead of Newtownabbey-Antrim and Larne-Ballymena-Carrick it would (to me) have been so much more compelling to have Ballymena-Antrim and Newtownabbey-Carrick-Larne.

    Still, what do I know?

  • Gopher

    Charles I dont think many of the Boundaries will make geographic sense they were just gerrymandered for religious balance. Lisburn/Castlereagh is the most bizarre LGD one could imagine with its salient into Belfast, Ards and North Down.

    A 16th century Map of the Holy Roman Empire makes more sense, infact the architects of the new Local government districts must have been Hapsburg fanbois becasue they have thrown in Dundonald as a concessionary town or Bishopric of Lisburn. Total nonsense

  • RyanAdams


    I think support for SF is very much concentrated in the area of Lisburn going into the Belfast. This boundary is very similar to the one that is used for Westminster/Assembly elections and removed the SF seat altogether in Lagan Valley when it was amended. If I read correctly you had overcast SF on boundaries before here (albeit, with a slightly different context this time)

  • Nicholas,

    Thanks for all the work you put into this, especially the maps. I have a question. The default position in most of the new 11 is simply to amalgamate the old 26. But when pieces were taken off of the old councils and given to a new council what was the rationale for this? Why not just go with a straight reduction from 26 to 11 using the old borders?

  • Gingray

    I think gerrymandered is a bit of a harsh term, when you look at the concentration of nationalists across NI, it is actually quite hard to get a perfect balance. Outside of the oddness around east belfast and the weird map with forestside, the rest seem largely ok.
    Charles, I do agree with you on the Ballymena/Antrim, but that is personal opinion rather than anything dodgy going on. Would you agree?

  • Coll Ciotach

    Charles – as predicted – check out the news for Margaret Anne McKillop, she was the chief strategist for SF in the middle glens – now standing for the SDLP. Quare gunk.

  • Gopher

    @Gingray When I say “Gerrymandered” the sin is against geography rather than against nationalism or unionism per se. The compromise just shows up are politicians for what they are quite useless.

  • Ulidian


    Who’s the “independent” in Causeway Coast & Glens for whom you’re predicting election?

  • Charles_Gould

    CC Thanks. That is very interesting!

  • Tochais Síoraí

    Ulidian – Read. They’re projections not predictions.

  • Red Lion

    It doesn’t even seem right to have Down in with Newry. When I’m in the lovely village of Strangford it seems so far removed from Newry area. I hope this doesn’t have a negative impact of the St.Pat’s festival in Downpatrick which is very well run and non-polarised.

    Anyway-the analysis misses out the impact of NI21 which I understand as it is based on past performance.

    What do the statisticians make of NI21’s chances and where, and who might lose to them. Or is this near impossible to predict?

  • Sorry to be slow in responding – I’ve been on the road for the last few days.

    Red Lion, as I said to Gopher earlier, these are 2011 numbers – a “pastcast” in Charles Gould’s elegant formulation – so no NI21. There simply are no statistics to go on. If I had to guess, I’d expect them to pick up a handful of seats in Lisburn/Castlereagh, south and east Belfast, and maybe County Down. Getting into double figures would be a very good result.

    Ulidian – all I’m saying is that there were enough votes for independents in Moyle in 2011 that I would have expected one to get in if the election had been held on the new boundaries. Of the four independent councillors elected in 2011, Padraig MacShane got the most votes, for what that is worth.

    Gingray – specifically on Lisnasharragh – I don’t have a strong view, but Irish Observer’s projection of 2 DUP, 2 Alliance, 1 SDLP and 1 UUP from 2011 looks sound to me. I understand that both the Greens and SF have hopes there for this year.

    Charles Gould – back in the day, I was always a bit mystified that Alliance did not do better in Ballymena. On the basic demographics it’s similar to a lot of places which are good territory for the party. But I am not a Ballymena expert.

    mjh – thanks for your detailed analysis. I can’t really challenge you on any of the specifics, but I’m grateful for your thought-provoking efforts.

    Long reply follows about the overall process.

  • On the discussion of the boundaries – I don’t like them much myself, but I think it is wrong to use the word “gerrymandering”, which means unfair manipulation of boundaries to ensure a particular electoral result. This was a bureaucratic process which was specifically instructed to prioritise conservatism ahead of service delivery, hence the result.

    No matter how you draw boundaries for X number of councils in Northern Ireland, with roughly equal population, you will find that you have a little over half of them with a Unionist majority and the rest with a Nationalist majority, and maybe one in the middle. And given current electoral trends, that is going to put the DUP and SF in pole position – but that is because of the way people vote, not because of where the lines are drawn on the map. So allegations of gerrymandering (or indeed of a “carve-up” from people who are still aggrieved that they got fewer votes in the last election) frankly go too far, and weaken the argument.

    Anyone familiar with Belfast must concede that there was an overwhelming case for it incorporating both Twinbrook/Poleglass and inner Castlereagh, which are socially very closely linked to the rest of the city; and the case for Glengormley, Holywood, Dundonald (and even Minnowburn, which is surprisingly far out if you are driving) etc is much weaker. The whole process was pretty transparent for those who cared to pay attention at the time. The Commissioner’s reports have vanished from the government website but are archived here (see Belfast boundaries specifically here).

    I do agree about the internal links in County Antrim (and indeed wrote about this at some length in my submissions to the parliamentary boundary commission). At the same time, once you have been given the remit to merge 26 councils down to 11 with a roughly similar population and only minimal surgery (Belfast apart), there aren’t a lot of choices. I note that in the original consultation document (see pages 41 and 43), not a single one of the proposed 7-council or 11-council models merged Antrim and Ballymena.

    I also agree that Lisburn/Castlereagh looks odd, though with all due respect to people who have been there more recently than I have, the Forestside decision looks about right to me – the remit was to include residential areas that look to Belfast in the city boundary, and Forestside isn’t exactly residential. The alternative to Lisburn/Castlereagh is to put Dundonald (including Carrowreagh which should be understood as northeast Dundonald) in with North Down and Ards (which is sensible enough) and then Lisburn has to take in Antrim (which makes a lot less sense, and also see above about internal linkages in County Antrim as a whole).

    Really I’d have preferred a much more wholesale revision of the districts, which would certainly have given us less sprawling output. To take three examples: the Dundonald salient, as discussed above; the north-south v east-west split in Country Antrim, also as discussed above; the inclusion of the Clogher Valley in a district which stretches up to Magherafelt, rather than the new Omagh/Fermanagh district. But the political choice was made pretty early on to stick to existing boundaries by and large, and we have what we have.

  • mjh

    Red Lion

    You asked if it was possible to calculate a forecast for NI21′s chances and where, and who might lose to them.

    The quality and quantity of information is limited – which means that the quality of any forecast is likewise limited.

    What we have is the performance of the two independent unionist candidates in the 2011 Assembly election. Extrapolating those to the whole of NI would have given an independent liberal unionist vote in the region of 5.6%. However NI21 may not be seen now as identical to those independent unionists.

    Since then we have only had one opinion poll with NI21, but that is now seven months old. It showed NI21 on 4.7%.

    It looks as if the party may fight fewer than half of the Electoral Areas. It has not yet announced its candidates r – so has lost precious time to organise, canvass and get them known. Moreover it appears that there will not have any sitting councillors among them, which means they start without the incumbency advantage the other parties will enjoy.

    For these reasons the party will poll below its potential. I guess at between 2% and 3%.

    As for the number of seats: as a general rule a party will win a similar percentage of seats to its percentage of votes. But this does not hold true if a party earns a very low percentage of votes. This is because it wastes a higher proportion of its votes in Electoral Areas where it does not win a seat. At the last council elections UKIP, PUP and TUV all won only half the number of seats equivalent to their share of the vote.

    So 2% of the vote would probably give a party 4 or 5 seats and 3% 7 or 8 seats. At 4% they may get nearer their vote share, say 14 to 15 seats. At 5% they might get the full 23 which their vote share would suggest.

    As to who would lose out: the Assembly independent unionists took the vast majority of their votes from the UUP. Although in both constituencies Alliance grew their vote, the evidence suggests that the independent unionists did intercept a much smaller number of votes that would have otherwise defected from UUP to Alliance. This effect was more marked in the constituency where Alliance had been weak, and much less so where Alliance was already strong.

    This does not mean that only the UUP would lose seats. It is possible that in a number of DEA’s an NI21 success could rob a UUP candidate of a surplus which would have been transferred to a DUP. So the UUP candidate may still get elected but a DUP would lose out.

    Where would the NI21 gains take place? Look first at Councils and District Electoral Areas where the UUP have an above average share of the vote. The (limited) evidence would suggest that NI21’s best prospects would be those of them where Alliance has failed to establish a strong presence – Armagh Ban & Craig, Ferm & Omagh, Causeway Coast & Glens, Mid Ulster. Intuition says the opposite – Ant & N’abbey, Mid & E Ant, N Down & Ards, Lis & Castl.

    Politically, however, it needs to take at least 4 within the Lagan Valley constituency and 4 within South Down in order to indicate that Basil and John are on track to retain their seats.

  • Gopher

    Ill disagree about Dundonald, it is always associated with Belfast and is much the same as Ballyhackamore. City bus runs right through it (unlike Holywood). Holywood was always associated with Bangor and North Down but Dundonald is part of Belfast, no geography separates it, like in the case of Glengormly. Dundonald has no association other than the hospital with North Down and little with Ards and absolutely none whatsoever with Lisburn. It is comical that if you are going to your “County Capital” you have to drive through Belfast which is proof enough. Though the “Dundonald Salient” does have a certain ring to it a blind man on a galloping horse can see its part of Belfast.

  • Fahas penultimate blog is now up on my site covering North Down and Ards. Interestingly he predicts a gain for the Green Party but still no joy for the SDLP.

  • Gopher

    Think this is the hardest one to predict basically for two reasons the first is you never know if this crowd will actually vote and secondly you never know who this crowd will actually vote for.

    This problem is compounded by a few things that will have changed since 2011. Alliance whatever people think of them nobody knows what North Down think of them. North Down and Strangford tend to be silent assassins. Secondly, North Down was a very flabby constituency James McKerrow UUP getting elected with 288 FP votes ELEVEN! stages later.

    Add to that NI21 and anything could happen.If I were a bookie North Down would be the one place I would be happy to lay (apart form the SDLP seat in Peninsula, nationalism join in the fun in that part of the world)

  • Big Boss

    Im late to the party here but id like to offer a though on the Mid-ulster figures,

    Espeically Dungannon and Torrent area… Iv looked at my figures as well as the candidates and i have a few observations to make

    Dungannon, with the new wards from Moy being added would surely give the SDLP an advantage, given that thier candidates personal/family history in that area. I think its more likely the SDLP will take a seat in that ward than say 2 SF. I think the independent republican will be safe.

    Torrent area is also unlikely to give up an independent nationalist. The area hasa history ofit, but the independent vote in 2011 was achieved by 2 candidates both on either end of the nationalist political spectrium. I think itsmorelikely that SF will take 4, or the SDLP 2 in this area than an independent. Of course it all depends on what independent(s) will be running.We also have to figurein here alot of big name candidates (at least on local council) are leaving the scene for both SF and SDLP. Just be interesting if the people will still follow the party line.

    In a boring Mid-Ulster council election i would look tro these areasfor a bit of excitiment.

  • Sean Og

    You are missing the point of this thread Big Boss.

    The votes from 2011 are simply being projected on to the new boundaries as a baseline.

    This isn’t about predicting the outcome of the elections in May.

  • Big Boss

    Fair enough!

  • IJP


    I was with you almost to the end there.

    If I follow you correctly, you suggest that NI21 will run in roughly half the DEAs, and that thus 4.7% becomes 2-3% overall (i.e. they average 4.7% where they stand but, obviously, zero elsewhere). However, you then add the (lack of) incumbency factor, thus suggesting this will be more like 2% overall, thus averaging 4% where they stand.

    The problem is, even this would in all probability deliver zero seats, assuming it’s relatively evenly spread (i.e all candidates score within the same general range, say in NI21’s case 3-5% – this happens with the NI Conservatives). Even 5% just never gets you in the running.

    It is a possibility they would get more than that around Lisburn, but I’ve been campaigning there and have honestly heard no hint of it whatsoever (that said, I heard no hint of Nicholson’s vote while campaigning in 2009, yet he was comfortably re-elected).

    Four seats in Lisburn is a hell of an ask – the Alliance Party has never managed it in its history (though it may do this time…)

    By the way, for a new party with no experienced election managers, merely getting 50 candidates correctly nominated would be some achievement, not least since we are now into injury time!

  • mjh

    Not quite, IJP

    I’m saying that we have two aging (and therefore increasingly less relevant) bits of evidence for the level of NI21 support. The first suggesting 5.6% and the second 4.7%.

    For the three reasons I gave – not standing in half of the Areas, handicapping themselves by starting their campaign well behind the other parties, and not having the benefit of sitting councillors who have acquired personal votes through the years – they will not reach their true potential (whatever that may be) in the Council elections.

    My guess at 2% to 3% of the total Northern Ireland vote is still a guess. That is why I also give possible seat projections for 4% and 5%.

    It seems reasonable to assume that if they fight 40 to 50 District Electoral Areas that these will by and large those where they enjoy an above average level of support. Therefore I would expect their vote share in those Areas to be MORE THAN double their potential Northern Ireland share. So MORE THAN 4% to 6%.

    You are too pessimistic to suggest this would give them zero seats. It is extremely unlikely that their vote will be evenly spread across 40 to 50 District Electoral Areas. No other party’s is. The reason the Conservatives did not win any seats was not because their vote was evenly spread – they just did not get enough votes. In the handful of Areas they fought in 2011 their vote share ranged from 0.7% to 4.2%.

    You make a good point that Alliance wins an Assembly seat in Lagan Valley with only 3 local councillors. On the other hand SDLP has 3 but no MLA, SF has 6 and no MLA, UUP 7 and only 1 MLA (now NI21). DUP has 16 and 4 MLA’s. Also under the new boundaries the number of seats in the Lisburn part of Lisburn & Castleagh district will be 20% greater. But you are right -it will still just be possible that 3 council seats will be sufficient to suggest that an Assembly win is not impossible. But they will need 4 to be reasonable confident.

    By the way those last numbers are for Lisburn and Dromore.

    Thank you for helping me to clarify my previous posting.

    I entirely agree that fighting half the Areas is a considerable achievement for a party only one year old.

    Whether it is wise we shall know in 6 weeks time.

  • SeaanUiNeill

    I’m still trying to work out why an “economy of scale” re-organisation of local councils and services is going to raise my rates bill!

  • Morpheus

    Interesting feature in The Belfast Telegraph asking the question “Are you voting for the right party?”

    Food for thought…

  • boondock

    Im a 90% match with the PUP ffs

  • Morpheus


    I was 75% NI Conservatives….had to look them up to see who they were!

  • mjh

    And apparently I’m Sinn Fein

    Bet they tell fjh he’s Alliance!

  • SeaanUiNeill

    I did my Tele Test as “Barack Obama” and discovered I should vote KKK.

    Next time I’ll try a really serious paper such as The Sun, and leave looking at the Belfast Telegraph for important information on Belfast nightlife.

  • IJP


    And most bizarre of all, I am actually Alliance… weird!

  • Charles_Gould

    Portadown looks an interesting ward. Back in 2001 it was a UUP town with 2 councilors, but DUP then had a strong surge so the UUP lost one. At the same time David Jones (an orangeman) was an independent councillor. He stood down in 2010.

    Now D Jones is back this time for UKIP, and my guess is that he must be in with a good chance, given that he is will known, especially in the Orange, but also more generally as a past councillor. He is talking the talk of “moving beyond old NI politics”.

    There is also a strong entrant for the DUP: Doug Beattie who is a decorated military man who has written some books on his experiences. He seems like a very likeable person, too, so a great catch for the UUP.

    Portadown has quite a few DUP councillors, one of which was co-opted when their top vote getter on council (Anderson) was elevated to being an MLA.

    So its possible DUP may lose out to strong challenges from UKIP and the UUP.

    Thats before we factor in challenges from NI21 and TUV.

  • oakleaf

    Are the bbc or utv going to do any decent coverage of the local government elections? There previous effort was an embarrassment.

  • Drumlins Rock

    The chances of good coverage are slim, they will prefer to talk about “topping the poll” and who is related to who rather than turnouts, transfers and swings that make the difference. They should be looking for those un expected results that either forecast change or hint at an interesting local story.

  • Charles_Gould

    In my 9.49 I meant that to say “There is also a strong entrant for the UUP” (not for the DUP).

  • Charles_Gould

    To be fair only a very small audience share would be interested in council elections – most people don’t know the names of their local councillors.

  • DC

    I reckon Alliance will fail to get its vote out in core unionist areas especially in Carrick, particularly so now Sean Neeson has retired and left the stage, not sure about how things will pan out in east Belfast.

    The lack of an effective liberal unionist party at the moment, one robust on constitutional identity and robust enough to handle SF, while all the same liberalising social policy, is making things tricky, unionist voters left with just the DUP. I think this will lead to poor turn out in unionist areas perhaps people feeling exactly the same way as Alex Kane and just not voting, as such voters no longer sure who is best to represent them and may not come out.

  • “To be fair only a very small audience share would be interested in council elections – most people don’t know the names of their local councillors.”


    In America for a serious channel like PBS or CNN the solution would be to have the political reporter run that information on his blog. Doesn’t Mark Devenport do this type of analysis on his blog?

  • Final Faha prediction article now published (Belfast):

  • Charles_Gould

    Comparing with Nicholas Whytes projections, we see that DUP and SF are the biggest under performers in terms of damage done:

    DUP projected 145. outcome: 130. (-15)
    SDLP projected 67. outcome: 66 -1
    SF projected 115. Outcome: 105. -10
    UUP projected 77. Outcome: 88. +11
    Alliance projected 34. Outcome 32. -2
    Green projected 1 Outcome 4 +3
    TUV projected3 Outcome 13 +1’s p0
    PUP projected 1 Outcome 3+2
    UKIP projected 1 Outcome 3 +2
    NI21 projected 0 Outcome 1 +1

    Combined Nationalist: 11 lower than projected
    Combined Unionist: 16 more than projected