Friday, 15 January 2010

Some Electoral Arithmetic for Ynys Môn (updated)


      Prompted by the Ynys Môn Conservatives selection of Anthony Ridge-Newman, the Druid has been checking the odds William Hill is offering on the Anglesey election result:

                          Now             Nov '09

Plaid Cymru           1/3                1/3
Labour                  9/4                5/2
Conservative          12/1              10/1
Lib Dem                100/1            100/1

So, the Conservatives’ selection of an English candidate has lengthened their odds from 10/1 to 12/1, whilst the Labour odds have shortened very slightly to 9/4. But as far as William Hill are concerned at 1/3 Plaid Cymru’s Dylan Rees is Red Rum - he’s a racing certainty.

But are we really so sure that the results are this certain?

Wales-only Voting Intention polls are pretty thin on the ground, but YouGov published their first last November, and here are the headline figures for All-Wales (the comparison is with the 2005 General Election results in Wales):

Con 31% (+9.6)
Lab 34% (-8.7)
LD 12% (-6.4)
PC 15% (+2.4)

Now, in predicting the 2010 general election results in Ynys Môn, All-Wales polls aren’t actually all that helpful as Wales isn’t really a very homogenous place. North West, Mid and West Wales leans strongly towards Plaid Cymru, Labour is strong in the North and the South, with Conservative support picking up towards the borders with England. Luckily the November ’09 YouGov poll also include a break down of these results regionally. The North Wales results were:

Con 40%
Lab 29%
LD 10%
PC 16%

Now, if we compare these poll results with the actual 2005 election results in North Wales, we get the following figures (look at the bottom 2 lines of the chart):

click image to enlarge

The Plaid Cymru vote is pretty steady with the Lib Dems are slipping some 5%. However, the big news is a MASSIVE 13.3% swing from Labour to Conservative in North Wales.

What does this mean for Ynys Môn?

Of course the situation isn’t really homogenous in North Wales either but, frankly, its the best guide we can get from the data available. So, if for the sake of argument we assume that Peter Rogers decides not to stand (a very big assumption) and his support from 2005 swings behind the Conservatives, this is what this polling data predicts for Ynys Môn:

So what can we conclude?

  • Ynys Môn in 2010 is a three-way-marginal. If Peter Rogers decides not to stand, there only looks to be about 400 votes between Labour, Plaid Cymru and the Conservatives. 
  • But we know that Peter Rogers not only split the Conservative vote in 2005 - he also took some Plaid Votes. If those voters fall in behind Plaid Cymru then Dylan Rees will just overtake Albert Owen and will be our next MP.
  • If however Peter Rogers DOES stand this year then the Conservatives are doomed and it will be a straight fight between Labour and Plaid - and probably much, much closer than William Hill is calling it.

Ynys Môn in 2010 is on a knife edge...
UPDATE: Another YouGov Wales-only voting intention poll, commissioned by ITV, has just come out here. The headline All-Wales figures are:

Con 32% (+1)
Lab 35% (+1)
LD 13% (+1)
PC 13% (-2)

YouGov don't seem to have published the regional data yet, but as soon as they do the Druid will re-compute the Ynys Môn 2010 prediction accordingly.

UPDATE 2: YouGov how now published the regional data of their latest Wales-only poll here. The figures for North Wales are:

Con 33% (-7)
Lab 34% (+5)
LD 11% (+1)
PC 17% (+1)

The comparisons are with the previous November YouGov Wales poll we discussed avbove. The North Wales sample is only a very small 236 people so there is more than enough room for error - however if we calculate the change from the 2005 General Election results as before we get:

Using this swing data to recalculate the predicted Ynys Môn result this year (with all the same assumptions as earlier) we get:

The Conservatives now seem to be out of it - with or without Peter Rogers standing - and it looks like a pretty close fight between Labour and Plaid Cymru - Albert Owen still has everything to fight for.
UPDATE 3: Thanks to Meurig in comments for pointing out that YouGov does not include the Meirionydd Nant Conwy area in their definition of 'North Wales'. The Druid has recalculated accordingly to give a more accurate prediction. So the revised YouGov North Wales numbers compared with 2005 General Election results in North Wales (excluding Meirionydd Nant Conwy) are as follows:

And the revised Ynys Môn predicted result based on these updated numbers are:

The gap between Labour and Plaid Cymru is now down to less than 100 votes. Dylan Rees may or may not be Red Rum - but its going to be a photofinish!


Meurig said...

Just a short note. The regions YouGov are using aren't exactly the same as in your analysis. The new constituency of Dwyfor-Meirionnydd is included by them in Mid & West rather than North Wales. It's tricky to take account of boundary changes, but I'd suggest excluding Meirionnydd Nant Conwy from your figures would produce a more accurate comparison.

The Druid of Anglesey said...

Many thanks for this info - I've updated the figures accordingly in Update 3 above.

DomFisher said...

Just came over from Iain Dale. I've been looking for somewhere with detailed local knowledge of your area from a political point of view. V interesting, but to be honest it would be interesting if you stuck your neck out a bit more. ;-)

The Druid of Anglesey said...

Dom, if I was to stick my neck out - I'd say Plaid will probably take Anglesey. Thats despite the fact that Plaid's Dylan Rees is a weak candidate, not from Anglesey and not known for his charisma. I think he's a drag on the Plaid vote, but the dramatic recent job losses will probably more than do for Albert Owen's chances. If Peter Rogers were to stand, with the current anti-politician feeling, I wouldn't be surprised if he even overtakes Albert Owen to take 2nd place. Good enough?

DomFisher said...


mint tea said...

I wonder if the Druid would be so bold as to make such a prediction today (16 April 2010). I think not.....