Which is why Labour leader Andrew Little’s attack on the One News-Colmar Brunton poll, as “bogus” – with its sham connotations – was so ill-advised.
Opinions are divided in the caucus over whether he should even have engaged – and hence highlighted – the poor 26 per cent result.
But given he did, he might have got away with “rogue”. There are occasionally polls that step outside the expected range of results. That’s precisely what the margin of error, normally expressed as a level of confidence at the 50 per cent level, accounts for.
There is a huge difference between calling a poll “rogue” and “bogus”. 1 out of 20 polls will be “rogue” in that the confidence level for the margin of error is 95%. That is just normal sampling error. But to call a poll “bogus” is to imply that the polling company has done something fraudulent. Little is lucky Colmar Brunton is not litigious.
That was, sort of, the point Labour was making by releasing the latest data from its pollster, UMR, with a warning over its finding National was on just 40 per cent. In other words, polls can be seriously wide of the mark at times. UMR’s probably was, and so was the One News poll.
But it didn’t stop some activists adopting a sort of post-truth polling stance, asserting the UMR poll much better matched their view of reality.
Sigh. Ice cream castles in the air.
Few things are more dangerous than only believing the polls which are good for you.
It also came at a bad time for the party as it contemplates that most fraught of MMP political processes; the shape of its party list and who will be high, low and shafted.
It all comes down to the party vote, of course, but with a twist for Labour.
It has pledged to gender balance its caucus by 2017.
When the policy was signed off in 2013, then-president Moira Coatsworth said the target would be achieved by calculating the gender mix at various different levels of support and taking into account the likely electorates Labour would win.
But a party vote of 26 per cent, in line with the TVNZ poll, delivers a very different scenario – and a political death sentence for many a male aspirant – than the 35 per cent-plus yardstick the party is assuming.
The problem starts with the imbalance in winnable electorates.
At the moment Labour has 27 electorates, but only 10 are held by women and 17 by men.
Throwing in a few seats it thinks it can win, such as Christchurch Central (lawyer Duncan Webb) and Auckland Central (Jacinda Ardern) doesn’t necessarily help
Neither did the Otaki selection meeting help by picking Rob McCann over pre-meeting favourite Penny Gaylor.
So applying the 50 per cent rule strictly, Labour faces a possible scenario where Little takes the top list slot and the next nine winnable slots go to women.
So good bye David Parker and Trevor Mallard and the chances of any other male candidate who doesn’t have an electorate.
That is being ruled out by Labour’s top table, because winnable places will need to be preserved for Trevor Mallard, as Labour’s nomination for speaker, and its policy brains trust David Parker (if he wants another term).
That doesn’t necessarily pre-suppose a top five place for both, because unless Labour gains enough seats to win the Treasury benches its nomination for speaker is academic.
Even so, it will no doubt raise questions among the activist base. Why adopt a gender balance plan if you are going to make exceptions based on the “need” to elect existing male MPs?
So they are going to ignore their own rule. Doesn’t that then show how stupid the rule is if they have deemed it unworkable?
But there is another issue complicating matters; Labour’s shameful lack of any MPs of Asian or Indian ethnicity in the current caucus.
Of course none of this is a problem if the polls improve. At 38 per cent everyone is in, everyone is happy, At 26 per cent its every “man” for himself.
No wonder Little and his team are hyper-sensitive to bad polls.
If Labour win say 30 electorate seats then on current polls they may not even get Andrew Little in on the list!