Poll predicts result, gets it wrong, so surely the previous critics of the series of polls were right all along: sounds simple, clear and logical.
Yet in the case of the Lib Dem Voice polls of party members, it’s more likely also wrong. Counter-intuitively, the very errors in the LDV poll on the Party President contest show that many of the previous critics were likely to have been even more wrong.
Here’s why: previous criticism has mostly (not completely, but overwhelmingly) been from the angle of ‘these results show a party that is too happy / loyalist / supportive of coalition and the truth is that the party isn’t really like that’.
Now it’s true that the Party President poll was wrong, and that suggests that previous results may not have been accurately reflective of the views of party members with things going off the rails somewhere between the point in the past when previous evidence suggested that the results were accurate and now.
But think about in what way the LDV poll was wrong.
The LDV poll greatly under-estimated the support for the candidate who was the most establishment, with the largest backing from senior figures in the party and who used high profile endorsements from the likes of Paddy Ashdown and Shirley Williams, who are pro-Nick and pro-coalition.
So if those previous LDV polls were wrong, were they likely to be under or over-estimating support in the party for Nick Clegg, coalition and related issues?
Despite the past criticisms mainly claiming that the polls must have been over-estimating it, if anything the President result suggests they were under-estimating it.