Statistical Gizmos and the UK Election

The recent elections in the United Kingdom provided a really fun opportunity to see how extensively statistical decision evaluation and predictive modeling have penetrated popular culture.  The British press outdid themselves with online graphical gizmos that allowed readers to set the terms for outcome scenarios and let those spin out in true operations management style.
While The BBC offered an election seat calculator that really only translated voting percentages to number of Parliament seats won, the Guardian put up a Three-Way Swingometer.  With about 8, depending on which you count, parties in the fray, the Swingometer allowed readers to twiddle a dial to anticipate the effect of hypothetical party-shifting and election results.  
Next, Nate Silver, the election forecasting guru behind the website, produced what he calls the Advanced Swingometer to offset the statistical disarray introduced by the original version’s assumption of a uniform rate of "swing."  He backed this up with a demonstration of how elegant the statistical analysis  behind his model was. 
The Times came forward with a predictive map based on the predictions of gamblers in UK’s lively betting shop scene.  Who know where those risk assessments came from.  

None of the online descriptions of the methods behind the gizmos were very detailed.  There were no mentions of named statistical analysis procedures, and this turns out to have been a good thing–because none of the gambits proved up to foreseeing the muddle that resulted from the actual voting.  If you wanted to try to come to a clear view of that, you will need to consult the decision tree posted by the BBC. 

Leave a comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: