Rob Hosking: Labour’s weird data obsession and that awkward poll

The problem is spreadsheets. Noses are buried so deeply into Microsoft Excel or whichever spreadsheet programme the Labour research team uses, they are losing sight of all political reality.

The party’s Parliamentary team, and certainly its back-office ?? or at least those who are left in its back office ? are obsessed with data.

Spreadsheets, graphs, data points, shifting trendlines and margin of error are all matters of weird fixation. They are treated like a combination of the Holy Grail and the Holy Bible: the subject of the ultimate quest and also matters of holy writ.

Now, it is not unusual ? in fact, it is common ? for political geeks to get incredibly excited about this stuff. Shifting data about polling trends as well as the economy and human geography generally is pretty much at the core of any political management tool.

No, that’s not the issue.

But it is backroom stuff. It excites those who are already excited by politics. It does little for anyone else.

Geeks who are obsessed with their data are needed in politics. And they’re not, as a group, bad people. Some of my best friends fall into that category.

Some of them are pink, run a blog and are mostly consumed by the arts, lifestyle and fitness.? But boy do they know their numbers.

Successful political leadership, in the end, is about judgement. It can ? and often is ? based on the kind of data provided by back room nerds.

But it is based on many other things as well: empathy, connection with other human beings, and often just sheer instinct and gut feel. All that appears to have been crowded out by Labour’s geeks.

In a different sort away, the infamous Chinese sounding names database Mr Twyford put together last year was a similar example of the same syndrome: the conviction that if data looks awesome on a spreadsheet in an office in Wellington it will be a powerful argument that will confound the party’s enemies and gobsmack the humble Kiwi voter with its brilliance and vision.

And then, of course, when data turns up which tells the party things it does not like, the response is a mix of denial and smear.

The attack on Statistics New Zealand a month ago has been mentioned: this week, there was the ? admittedly much less serious, but equally psychologically telling ? refusal to accept the result of the Colmar Brunton poll.

Now, a sensible reaction would have been to shrug, admit Labour is not doing as well as it would like, but perhaps add the classic, rather boring but politically prudent line about not commenting on individual polls.

Instead, Mr Little damned the poll as ?bogus.” To compound the matter he then released Labour’s own polling, which had several effects, none of which are helpful to the party.

One is it kept the story going for another 24 hours.

Two, while the junkies might know that UMR is a reputable pollster, undecided voters are only going to hear ?Labour?s pollster? and decide the whole thing is a bit of a jack up.

And even the political junkies are asking pointed questions about the weightings used in that UMR poll, what methodology, and precisely what questions were asked. Despite promising to release its methodology, Labour has not done so.

Only the commenters at The Standard and the Labour party itself consider the UMR poll to be a good poll.

But explaining is losing, and the worst mistake Little made is blow it up as an issue.? The response has been near universal disbelief.? And now, everyone wants to keep seeing the next UMR poll.? And the next.? And the next.? If Labour don’t, they’ll be “hiding something”.? If they do it sporadically, they’ll only release the “good ones”.

What a prize idiot.

Whaleoil has reset the Ballsup counter.? He may have set a new record of 21 days (previous was 13), but this does have all the hallmarks of another howler.