No, I read and understood it.
It's obvious that you have have no experience in interpreting statistics. Of the 6.2% that voted BNP in June, a certain percentage of those would have been "possible" and "probably" leaners which were converted into actual votes on the day. Given the small sample group and the BNP's very localized level of support, never mind the fact that admitting to being a BNP voter is only marginally more socially acceptable than admitting to being a child molester, it would be quite absurd to expect a survey like this to accurately match the June election numbers, even if you had taken it on the day after the election. What it does do, however, is give us a number of indicators, which in even the most hopeful interpretation would suggest an increase in support for the BNP. For the record, I don't think that's a good thing, I'm just giving you the benefit of my experienced analysis.