Does the Le Pen support explain the gap between the firms?

The above table has been adapted from the excellent list in Wikipedia and shows the range of opinion polling in the final week before the crucial first round of voting tomorrow. The top two tomorrow evening then go into a run off election on May 6th.

So we have a polling range from a 6.5% margin for Sarkorzy over Royal to just 1%. Both cannot be right.

In 2002 the best polling prediction of the share of the far right Le Pen was 4.5% lower than what he actually achieved in the ballot. Big efforts have been made by some of the pollsters to avoid a repetition and it appears that those that are giving more weight to those respondents who say they voted for Le Pen last time are showing the lowest Sarkorzy figures.

Ipsos, which owns Mori in the UK, has been showing much higher figures for Sarkorzy and usually has lower shares for Le Pen than the other firms. At the other end CSA has been showing the highest figures for Le Pen and lower shares for Sarkorzy.

Looking at the table above it appears that the surveys showing the highest shares for Sarkorzy also have lower shares for Le Pen. So if Le Pen does end up with more than the polls are suggesting then Sarkorzy might be affected more than the others.

My polling predictions are:-

Sarkorzy will probably do worse than the final polls are suggesting

Royal will equal or maybe exceed by a touch the final figures

Le Pen will do better than even the highest prediction from the final week.

Bayrou will equal or do slightly worse than the polls

In the betting Sarkorzy is still the firm favourite but has eased to 0.57/1. I think that the current 3.3/1 on Royal is a value bet.

Mike Smithson