Last lap

Hi,

I’m out of the jurisdiction at the moment (and posting this from a hot-spot that refuses to recognise Firefox!), but got a text last night of figures for the last MRBI poll, which I presume you’ve all read. This is an important one, as the final MRBI in 2007 out predicted all the other polls, including a later effort from RedC, with the MRBI poll being to the nearest percentage point for FF, FG and LP.

The percentages (along with seat projections) are as follows;

FF 16% (18 seats)

FG 37% (74 seats)

LP 19% (38 seats)

SF 11% (16 seats)

GP 2%

OTH 15% (20 seats)

Unsurprisingly, it reflects the swing against LP last week, and if LP are to finish north of 40 seats, they will need to have some sort of modest recovery in the final week. This level is almost exactly what they got in 1992, although there was must less of the vote to their left in that election. In ’92, had they run sufficient numbers of candidates they would have got about this figure, so it is open to argument as to whether this would be a better or worse result.

FG, perhaps slightly surprisingly, appear to be approaching a ceiling. My own expectation was that they’d break the 40% barrier in this poll, with a bandwagon effect (coupled with the fear of Indos holding sway) pushing them over the line towards overall majority territory. While this would represent an outstanding result for them, it is again, like LP, no more than their best performance under Garret FitzGerald in the 1980s.

SF will be happy to see another poll that has them over 10%, and will be hoping that they hold this vote better than they did in ’07. Given how much of their vote appears to be rooted in anger over the bailout, I suspect that they will. The Greens, while still in the margin of error, will hope that they can push up to 3-4% in the final week and salvage up to 3 seats. OTHs remain an enigma, and one that won’t be resolved until Saturday when the votes start coming out of the boxes.

FF on the other hand seem stuck on 16%, and will be hoping that the methodology of all 3 polling companies is understating their support. They may be, however it’s hard to see this being more than a couple of percentage points, and they probably have to hit 19% or so to make the breakthough in a large number of 3 seats in the West.

And 19% undecided at this stage? Will they even make it to the polling stations, I wonder? Time will tell

I’d like to remind you that the prediction competition is open to the close of polls, when I understand that details of an Exit Poll may start to emerge. As soon as we have that out, I’ll run the figures, and on the day of the count, I’ll be using a spreadsheet I’ve developed to take the tallies as they come , compare them against the expected results, and use the variance to estimate the votes in un-tallied areas. The early results will be skewed by incomplete tallies in perhaps an unrepresentative sample of constituencies, but by 10-15 tallies in I expect us to have a reasonably good steer on where it’s going.

Laters….

D