If you have polling that says you are well ahead of the other candidate, crowd sourcing a ground swell of support to create more momentum by using so-called "people power" might be a good tactic.
Or it might just expose how fundamentally flimsy your position is.
In the last few days there have been three separate polls that have all said that Kevin Rudd is preferred Labor leader to Julia Gillard and it is the basis of his pitch to caucus – "I am popular, Julia Gillard is not. I can beat Tony Abbott, she can't."
But the problem in these times of excitement is that polls often capture the volatility of public opinion, not its underlying strength.
What someone tells a pollster at 6:30 in the evening having watched the 6:00 p.m. news and whilst preparing dinner is not necessarily a firm conviction.
When Kevin Rudd issued his call for supporters to ring their local member and lobby for him we decided to host an online poll to make this "Celebrity Big Brother" more transparent and informative.
We gave respondents the option of nominating "Rudd", "Gillard", "neither" or "unsure" and of telling us why. We also asked about voting intentions and demographics so that we had an idea where respondents were coming from.
The survey was promoted to our usual online panel as well as via Twitter and Facebook.
The result is that we had 2,883 responses, far in excess of our usual response rate, and more likely to be from Greens and Labor voters than normal.
And of those respondents 38% wanted Gillard, 36% Rudd and 24% neither.
So Rudd's tactic fails if it is reflected in the calls going to Labor members' offices, which informal media surveys suggest it is.
It only works if Rudd is not just ahead, but by a significant margin. To be close or behind is a disaster, displaying the underlying weakness of your position.
Discuss in our Forums
See what other readers are saying about this article!
Click here to read & post comments.
15 posts so far.