Slugger NI Election prediction competition!

One for the political anoraks: Stratagem are running an NI Election Prediction Competition in which you have to say which party will win each individual seat. In the spirit of complementarity rather than competition, we’ve got a few extra questions for you (below).

NB: You will need to read the update as a few entrants haven’t fully understood part of this competition.

Specifically, we want you to predict the turnout and percentages of overall votes cast in NI for each of the major parties. Then, you have to tell us which seat each of the major parties will have the most positive performance relative to 2005. Where there have been boundary changes, the performance will be related to the nearest relevant seat. Ask for clarification in the comments (below) if you have any questions.

How do we score it?

We’ve not finally decided yet – hey! We’re all Web 2.0 here – we don’t impose our judgements! We crowdsource your wisdom instead. We’re sickeningly smug that way. Here’s our initial bid though:

  1. Percentages: If you predict that the DUP will get 20% and they get 25%, you will get a score of -5 for coming in five points below the right answer. If you guess that they will get 27% and they get 25%, you will get -2 for being two points above the right answer. You lose a point for every % point you are away from the correct answer either way.
  2. Best performances: You have to guess which seat the main parties will enjoy the best / least worst swing since 2005**. In the case of the TUV, it’s just about guessing their best constituency %  but if you think that a party will make no improvements, then pick the seat where they will suffer the least damage. You will get …. 5 points … for every correct answer.

Is this scoring system fair? Let us know in the comments. We think it’s a good opening bid, but there’s no rush in deciding, after all, the scoring system won’t affect your predictions. So go for it, and let us have any questions or suggestions on the scores in the comment thread (below). Good luck!

(NB: one qualification on our enthusiasm for crowdsourcing: We’ll look at comments on the points system but the final decision will be ours. This madness can only be taken so far…)

**Update: Just for clarification, it is the best % swing in relation to the 2005 % that the party scored in that particular constituency we are looking for. It’s the seat where the their vote went up by the largest number of percentage points (or declined by the least number of points). It’s quite possible in this scenario that the party you pick to have it’s best performance in a particular seat isn’t going to be the overall winner of that seat. Stratagem’s competition fulfils that role. If you want to re-do your entry, follow the guidance I’ve given to Drumlin’s Rock in the comments.


  • Chris Donnelly

    Point of order (just before any other anorak gets in)

    Regarding the ‘Best Performances’ criteria. Am I assuming correctly that the swing will be in terms of % of overall vote in constituency and not in relation to just own party’s vote in 2005?

    It’s important because, if it’s the latter, then entrants will hardly look past Strangford or East Antrim for nationalist parties. If it’s the former, then it becomes more challenging (and fun.)

  • orly

    I read it as the constituency where the party will get the highest % of the vote.

    So for the DUP I punted for Lagan Valley, for example.

  • Lionel Hutz

    What does Rodney Connor count as?

  • fitzjameshorse1745

    Mr Donnelly,
    Id hate to give away my sure-fire tips LOL.

    But I think (if Im reading the same thing right) Nats might not plump for those two.

  • Chris Donnelly

    FJH and Orly
    Think you’re right.

    He don’t count, nor does Hermon. Party designated individuals only (gets too messy otherwise.)

  • Drumlin Rock

    Hermon & Connor count as “others” which has just made me realise I forgot to factor Hermon into the percentages d’oh.

  • Picture this

    How many candidates from the ‘big 4’ will lose their deposits?

  • Chris,

    Your interpretation is the correct one. I said ‘swing’. It’s only the TUV where simply the best % is the measure because they didn’t run last time.

    I’d be interested to hear if anyone thinks that 5 points is the fair reward for getting each ‘Best Party Performance’ right? Should it be 3, 4 or even 6 points? Any views?

  • If you want to do it again, I can delete your previous entry? Do it and send me an email using this link – – saying what email address you used for the first one and I’ll delete it?

  • Orly, No – it says ‘best swing’. Some of the big parties have constituencies that are pretty-well nailed-on victories. In those seats a mediocre campaign can still result in a resounding victory because the others don’t bother campaigning. I’ll clarifiy the text on the post though to make it clear that it’s a swing in % terms.

  • Chris, just to clarify – re-reading your comment – it is the best % swing in relation to the 2005 %. It’s the seat where the their vote went up by the largest number of percentage points (or declined by the least number of points). It’s quite possible in this context for it to be a seat that the party doesn’t win overall. It’s about finding the best local improvement

  • redhugh78

    forgot to factor Connor in as other….doh!!

  • L. Costello

    I’ve tried a few times now to send my predictions but it won’t send! What is wrong? I’ve entered all of the required data in the respective fields.

  • L – I’m not sure – it’s working in Google Chrome and Firefox on a PC and in Firefox and Safari on a Mac.