Once you've done that, it becomes clear that National is nowhere near commanding a majority in the 2009 Parliament, unless you assume that either (a) all the undecideds are not going to vote, or (b) when they do come to vote they will all neatly fit into the patterns of those who have already made up their minds.
Below I've reinterpreted the July TV3/TNS poll. I can't speak to whether it is any better or worse than any other polling company, but they're the only company that I can find their full poll results from Scoop (unfortunately I can't yet find August 08 though).
|July||Original||Actual# ||Correct %|
|Don’t know/ undecided||11.9||119||11.90%|
|Would not vote||1.7||17||1.70%|
I also ran the median support percentages across the Feb, April, May, July 08 polls and the numbers come up pretty close to the July results - Labour on 30.7%, National 42%, undecideds on 11.55%. It is where the undecideds will fall that will determine the election, or perhaps more importantly whether they go out to vote at all. My money is firmly on those undecideds being made up primarily of previous centre-left voters who have lost faith in Labour coalitions, but don't believe National offers anything better for them (except perhaps vague platitudes).
NB: On reflection, II should expand on the last point about turnout. I don't think it is safe to assume that the people who are undecided are necessarily more likely to not vote. There are a multitude of reasons why people don't vote: for instance, if it's raining on polling day, turnout will be down; ditto if there is a major rugby test. Heck, there are a great number of opinionated people out there who will give you there preference for Government, but won't actually get out of their chair to put there preference into practice (To illustrate, turn-out last year was 77%, but the proportion of undecided will not have been more than 10% the week before polling day); as long as National looks like a clear winner (because of the dodgy reporting on poll-results that are the subject of this post), I suspect the more that National supporters might be lulled into laziness in the belief that other voters will do their work for them. Nonetheless, my feeling is that Labour is the one that will be most damaged by low turnout.