Friday, April 30, 2010

Back in the UK

The British election is now a week away and the Conservatives seem all but assured of victory - the real question is whether or not they'll be able to get their majority. Sound familiar?

The real twist is the rise of the Liberal Democrats, which has buggered up most seat projections. But hey, that just makes it more fun. Here's the skinny on UK polling numbers and projections:

1) The UK Polling Report has all the latest polls.

2) The gambling line at Betfair sets the Conservatives at 318, Labour at 216, the LibDems at 87, and other parties at 29.

3) You can play around yourself with BBC's swingometer. Just plug in the numbers and, Bob's your uncle, you get projected seat counts!

4) A more advanced projection model can be found here. My only qualm is that I just don't think there are enough data points to customize vote patterns on a riding by riding basis. Regardless, it projects the Conservatives at 291 seats, Labour at 209, the LibDems at 120, with 8 seats going to other parties. It sets the odds of a Tory majority at 8%.

5) Electoral Calculus projects seats using a regional swingometer and the betting markets. They have the Cons at 283 seats, Labour at 238, the LibDems at 97, with 32 seats going to other parties.

6) Finally, this brings us to our good friends at 538. Their model is a bit too subjective for my taste, and I'm not a fan of the geometric swing. The problem is, if you assume Labour will only hold 60% of its vote and you extrapolate that to every riding, you'll be taking the most votes away from their strongholds. For example, if you did this in Canada and had the Liberals down in the polls, their biggest losses would come in Toronto. Which I'm not sure is what would happen in reality.

On the flip side, their model will probably account for the rise of the LibDems better than most. So I guess we'll have to wait and see. Regardless, they project the Conservatives at 299 seats, Labour at 199, and the LibDems at 120.

Labels: ,


  • Plus, it's never good to bet against Nate at 538 ;).

    I'd trust his overall counts over the others just because of his pretty amazing track record.

    By Anonymous Anonymous, at 8:49 p.m.  

  • 538's coverage of the UK election has been terrible - nobody on their staff including their "expert" know anything about parliamentary elections. At one point their guy was going off on how the Tories needed a 10 point win just to get a plurality.

    As for Nate's accuracy - what accuracy? 538 is only as accurate as the polls fed into its voting model. Well actually it is a bit less accurate, except where there is a dearth of data or good polling. As a result, the 2008 primaries really allowed Nate to shine - largely because it provided better predictions than polls showing a Clinton lead rooted in name recognition.

    Nate pooh-poohed a Republican win in MA till the last second (and boy did he hedge). He didn't think Christie was going to pull it off, either - he was wrong.

    538's model largely makes consensus predictions, and even then, they are lagging ones since the model only changes after polls come out. It has not functioned well since the 2008 election.

    I believe that prediction projects like that of Ray Fair's presidential vote equation are of infinite more value than the likes of 538. Fair uses economic data to predict the next election result. As a result he can make fairly accurate predictions a couple of years in advance. True, on election day he is usually less accurate than a 538 - but he is more accurate for the majority of the year.

    In the UK, I predict that the Liberal Democrats will win substantially more seats (~150) than most are predicting. If you look at polls focused on Labour-to-Conservative marginals (eg. Angus-Reid's polls), the Lib Dems run a close second to the Tories.

    By Blogger french wedding cat, at 1:42 a.m.  

  • Shorter hoser: Blah blah blah grumble growl.

    Whatever . . . you were the loser who kept blabbing during the US Presidential election, resorting to latching on to false reports floated the day after about turnout only to be proven flat wrong when the actual numbers came out.

    The intensity with which you babble didn't make you right then so I'm pretty immune to your chortling now . . . though nothing would please me more to see the Liberal Democrats pick up more seats than predicted.

    By Anonymous Anonymous, at 12:06 p.m.  

  • Angus Reid's top line polling has been very chaotic. I wouldn't trust their results.

    By Blogger kevinsutton, at 5:32 p.m.  

  • "Whatever . . . you were the loser who kept blabbing during the US Presidential election, resorting to latching on to false reports floated the day after about turnout only to be proven flat wrong when the actual numbers came out."

    Do you mean false reports like this one from the Census showing that voter turnout was unchanged between 2004 and 2008? Or is the census just blabbing too?

    By Blogger french wedding cat, at 7:02 p.m.  

  • "Additionally, voters 18 to 24 were the only age group to show a statistically significant increase in turnout, reaching 49 percent in 2008 compared with 47 percent in 2004. Blacks had the highest turnout rate among 18- to 24-year-old voters — 55 percent, an 8 percent increase from 2004. The increased turnout among certain demographic groups was offset by stagnant or decreased turnout among other groups, causing overall 2008 voter turnout to remain statistically unchanged — at 64 percent — from 2004."

    By Blogger french wedding cat, at 7:03 p.m.  

  • Grumble, grumble double growl.

    Read it and weep . . .

    Incidentally, GMU is recognized as a rather conservative clan in US university circles.

    Grumble away, mate, and I'm still rooting on your prediction about the lib dems though the fact that you're ready to sever limbs about it makes me less optimistic.

    By Anonymous Anonymous, at 11:28 p.m.  

  • Besides, I just read your linked report. Now let's see what might be a better study. One that actually LOOKS at the amount of votes cast or one that reports on a census count of "reported voting" (i.e., basically asking who voted).

    Hmmm, I think I'd have to go with the study looking at the ACTUAL vote count over the poll. But its nice to see you dug up something that kinda, sorta supports the argument you were spewing rather definitively with 24 hours of the polls closed (and which matched the republican spin of the day).

    Carry on.

    By Anonymous Anonymous, at 11:37 p.m.  

  • 1. I was able to claim that voter turnout did not increase close to election day because I knew how many people voted. An increase of 5 million is not very high since the voting age population rose as well (by about 11 million).

    2. My source has a less accurate numerator, yours has a less accurate denominator. Macdonald's calculates turnout by dividing by voting age population. However, there are a number of factors (eg. change of address, whether somebody is a felon) that are difficult to account for. Macdonald mentions this in his FAQ. The Census study, by contrast, has accurate demographic data at the individual level.

    Moreover, the increase Macdonald shows is only 1% - hardly the remarkable turnout trumpeted by the media. Lets split the difference at a 0.5% increase in voter turnout - that is far smaller than the increase we saw in 2004 of about 5%. However the 2004 result didn't make the headlines because it didn't fit the story of a boring Kerry and an evil Bush.

    As to GMU's bias, I don't think a professor would fudge their numbers anyway (or that bad numbers would pass through the peer review process). Nonetheless, you have found yet another place in which to be wrong:
    If you look at donations from GMU employees, they donated $52,950 to Obama and $15,050 to McCain.

    3. I am not a Republican (I do prefer to be on the unpopular side of an argument though), nor am I an especially angry person. Given the lack of data we have about one another, it might be a good idea not to jump to those conclusions - I mean you probably aren't as dumb as I think you are :).

    By Blogger french wedding cat, at 1:18 a.m.  

  • The Angus Reid UK polls have been attacked for being biased - a lot of questions were asked before vote intent.

    By Anonymous Sean, at 12:13 p.m.  

  • By Blogger mmjiaxin, at 8:17 p.m.  

  • By Blogger fattoma, at 8:54 p.m.  

  • By Blogger rehabgad1, at 10:10 a.m.  

Post a Comment

<< Home