Illustration / Anna Crichton
Can the techniques economists use predict the winner of the Rugby World Cup?
Massey University economist Sam Richardson is prepared to give it a go.
At the New Zealand Association of Economists conference in July, he presented the results of some modelling he had done on the question.
After all, economists are frequently interested in why some countries outperform others by some measure.
The problem they have is that when they compare any two countries, there will be all sorts of differences between them that might make a difference to whatever the researchers are interested in.
So economists employ fancy mathematical techniques to try to isolate and quantify how much difference any one of those factors makes, for example, by controlling for other factors and seeing how much of an effect remains.
It is about quantifying the relative explanatory power of the variables.
They can then weight the variables in a mathematical model and compare the results it generates about the past with what actually happened. The better the fit, the more confidence they can have in the predictive power of the model.
They have to work within the limits of the data available, however. They might well be able to dream up more subtle versions of their models - factoring in a pool of death effect in the Rugby World Cup case, perhaps, but clearly not before the draw has been decided.
Richardson was able to trawl through the results of 25 countries over seven tournaments.
The factors he considered were whether there was a host nation effect, how important having a New Zealand coach was and whether it helped to have previously won the cup, or to be the current Six Nations or Tri Nations champion.
He also looked at a couple of economic factors, whether the size of a country's population or its wealth (proxied by per capita gross domestic product) mattered. Those factors, after all, might give New Zealand an advantage over Samoa or Tonga, but would favour the United States and Japan most.
It turns out neither is significant - just as well, given that New Zealand's per capita GDP is nothing to write home about.
Hosting a competition does seem to boost a country's medal tally in the Olympics.
But no similar effect is evident in the Rugby World Cup.
Research by PwC similarly found that population and per capita GDP were not significant factors in Fifa World Cup finals.
Having won the Rugby World Cup in the past does not seem to make much difference either. Past performance in that sense does not predict future success.
The more recent past is, as you might expect, more relevant. Being the current Five or Six Nations champions at the time of a Rugby World Cup has proven a plus.
But the southern hemisphere equivalent, not so much. Just as well, perhaps, as it is currently Australia, albeit from a competition truncated by the RWC itself.
One factor of particular interest to Richardson is whether having a New Zealander as coach of a national team makes much of a difference.
It turns out that it does.
"There is evidence to suggest New Zealand coaches positively and significantly improved win counts and tournament performance," he said.
And though they fall outside his data set, Japan's win over South Africa and Wales' over England add weight to the thesis. In all, six of the contestant nations this time have a New Zealand coach.
In 2011, the model Richardson estimated would have given the highest probability of winning the cup to New Zealand, as of course it did.
It would have also accurately identified the quarter-finalists.
But it would have given the second highest probability of winning to England (which in the event never made it beyond the quarter-finals) ahead of the eventual losing finalist, France.
"The model will explain a certain amount, but there will always be an element of uncertainty, an element of unpredictability," Richardson said.
And that is not only obviously right but also a good thing.
It would be a pretty poor sort of sporting contest if the dismal science, economics, could reliably predict the outcome.
And for the current tournament? The model picks Ireland to win, followed by New Zealand and then Wales, Australia and England.
Ireland has a Kiwi coach and though at the time of writing they had yet to be tested by the heavyweights in their pool, notably France, having won the Six Nations this year and last year, the odds of them making the knockout stages look pretty good.
But time, of course, will tell.
For too long, ‘accident’ has been a blanket term that absolves responsibility.