BlogMenai, Hung Parliaments and Statistical Gobbledegook

  

Mandy Rice-Davies: "Well he would, wouldn't he?"

BlogMenai has taken issue with the Druid's recent post analysing Plaid Cymru's claims that in the event of a hung parliament they could hold real influence. In my post I relied on the polling data available on the non-partisan poll aggregating site electoralcalculus.com and concluded that, in the event of a hung parliament with the Conservatives the largest party but short of an absolute majority, there is very, very little likelihood that the Conservatives would ever need to rely on Plaid Cymru votes - thus contradicting Plaid's central argument to voters in the run up to the General Election. You can read it all here.

BlogMenai - an openly Plaid Cymru supporting blog - completely dismisses the Druid's conclusions ("well he would, wouldn't he?" as Mandy Rice-Davies might have said) and goes on to claim that I rely on statistical gobbledegook and that the case I made is false for the following reasons:

1. I have used polls selectively to support my point and ignored those which don't fit in with my preferred conclusion; and
2. that I have used outdated polls.

To prove his point, BlogMenai then lists the results of a number of polls (without any attribution) which show a much smaller gap between the Conservatives and Labour than an aggregate of the polls of the month ending March 4th on which my post relied (my post was made on March 9th). BlogMenai then goes on to state - again without any supporting evidence - that some of the polls he lists point to Labour being able to form a coalition government with the support of just SNP and Plaid Cymru alone without needing the Lib Dems at all.

My reply

BlogMenai is certainly right to say that the polls have tightened but I'm afraid the claim that I somehow 'cherry picked' only those polls which supported my argument is patently false. The Druid wrote the post on March 9th, in response to a letter by Dylan Rees in the Daily Post the previous day, and based it on the non-partisan polls aggregator electoralcalculus.com which on the day of writing was aggregating polls for the month ending March 4th (electoralcalculus.com is only updated once or twice a month). Now electoralcalculus.com aggregates ALL the polls over the stated period so BlogMenai's claim that I somehow maliciously chose only those polls which supported my argument is so obviously false to be laughable.

But its worse than that. Whilst accusing your humble Druid of only selecting polls which support his argument, BlogMenai then goes on to (a) disregard the results of Angus Reid polls because (he claims) they are "untested in the UK" and "give Labour a much, much lower share than other conventional pollsters" and (b) makes his unfounded conclusions on the basis on a selection of polls of his choosing - the very definition of selecting polls to suit one's argument. This is particularly ironic considering BlogMenai piously concludes his post by saying that although he may be biased towards Plaid Cymru he would never stoop so low as to use statistics selectively to make a point. Really? It doesn't look that way.

That said BlogMenai is right to say that the polls are tightening so lets crunch the numbers again using the latest aggregated numbers from electoralcalculus.com - those for the period February 5th to March 18th (although I'm sure our cherry-picking friend BlogMenai will complain once again that they aren't recent enough or that a selection of polls chosen by himself show an even smaller Conservative lead):

So with these more recent polls, electoralcaculus.com predicts exactly the same number of seats for Plaid and the SNP as previously:

Plaid Cymru - 5 seats
SNP - 7 seats

Therefore between them Plaid and SNP are still likely to control a block of just 12 seats. Labour, based on electoralcalculus.com's same calculations would win 262 seats - i.e. 64 seats short of an absolute majority. In this case Labour would need the support of the Lib Dems (53 seats) and both Plaid and SNP (12 seats) together just to get a majority of just one seat. Doesn't sound anything like BlogMenai's glib statement that Labour can form a coalition government with the support of just SNP and Plaid Cymru alone, does it? (Incidentally, BlogMenai also fails to explain how exactly Plaid Cymru and the SNP will be able to work together on reforming the Barnett Fomula when even Plaid's Dafydd Wigley has noted that they would be working at cross purposes and would need to hold separate negotiations with the Tories. You didn't mention that point in your critique of my post, did you, BlogMenai. Didn't it fit in with your argument perhaps?)

Anyway, returning to the figures for a moment: the Conservatives (302 seats) are still able to form a coalition with just the Lib Dems (53 seats) - giving them a comfortable majority of 29 seats (whether they call it a formal coalition or not). So still no need for Plaid's votes and still no extra influence for Plaid in Westminster.

The Druid's political allegiance

BlogMenai's criticisms don't stop there however. He goes on to say that the Druid must be a Tory because I only attack Labour and Plaid Cymru. I don't know how many times I have to write this but the Conservatives have not had any influence over Anglesey since 1997 -- 13 years ago. What's the point of attacking them? What I'm concerned with is the CURRENT state of Anglesey and those who bear political responsibility for it NOW. And, like it or not, the facts are:

• Ynys Môn has now been represented by Plaid Cymru's Ieuan Wyn Jones as either MP or AM for the past 23 years
• The UK has had a Labour government in Westminster since 1997 and Ynys Môn has had a Labour MP in the shape of Albert Owen since 2001
• The Welsh Assembly has been governed by Labour since 1999 until the emergence of a Labour/Plaid Cymru coalition in 2007

Therefore, BlogMenai, the truth is that Ynys Môn has been administered at some level by Labour and Plaid Cymru for almost quarter a century, yet you think that as an Anglesey resident I should somehow be directing my fire on the Conservatives for the Island's woes? Anglesey is now officially the poorest county in the UK - the very definition of political failure - and BlogMenai's partiality to Plaid Cymru is obviously blinding him to the conclusion that Ieuan Wyn Jones HAS FAILED the residents of Anglesey.

The Druid's Challenge to BlogMenai

Anyway, the polls will no doubt continue to change between now and the general election so I will continue to crunch the poll numbers once a month and if it ever gets to a point were statistically the polls point to Plaid Cymru having any real influence I will plainly write so here. Furthermore if following the election there emerges a Labour coalition with just Plaid Cymru and the SNP - as BlogMenai predicts - then I'll eat my 'wisg las'. If on the other hand such a coalition fails to materialise then I look forward to a full apology from BlogMenai. Okay?

Finally my message to BlogMenai readers is this: if you want to read about the world as Plaid Cymru would dearly like it to be - go ahead and read BlogMenai; if, however, you want to read about the world as it really is, read the Druid.

UPDATE (23 March): BlogMenai has now responded to this post and you can read his reply (in Welsh) here. My response is coming soon...