Survation and ComRes least accurate pollsters for European Election

There is some justified crowing going on over at You Gov following the European Election results. The pollster highlights that:

YouGov was the only company to get the top two, UKIP and Labour, right to within 1 point; YouGov was the only company to get all parties right to within 2 points; YouGov was the only company to rank all five parties in the right order (with Greens coming 4th and Lib Dems coming 5th).

It appears You Gov is setting the standard with the way it handles the data in order to generate the most accurate prediction of voting intentions.  Survation and ComRes gave the least accurate predictions as can be seen from the graph below:

In fairness to Survation and ComRes, they at least both correctly identified UKIP as the election ‘winner’. ICM on the other hand understated UKIP’s vote and overstated Labour’s by four points in calling the election for Labour.

We won’t go as far as to suggest Alan Bown should consider getting his money back for Survation’s polling for UKIP in marginal constituencies. But perhaps UKIP should be careful to note the performance of all the pollsters and their handling of responses where voters refuse to say who they plan to support, or say they are undecided about who they will support, before deciding which constituencies the party ‘throws the kitchen sink at’ in 2015.

An additional factor to bear in mind is that the results above were generated using larger sample sizes than the ones in the constituencies that have been carried out for the parties, and therefore the scope for error is likely to be even larger in those very targeted polls.

When looking ahead to the 2015 elections, observers can clearly do worse than treating You Gov’s polling outputs as the most likely indicator of voter intentions.