The Roar
The Roar


What are our chances of winning the Ashes?

Shane Watson may be out of the Test side for good. (AP Photo/Rick Rycroft, File)
25th June, 2013
1040 Reads

Like some revenge-fuelled fisherman in one of those godawful ‘I know how you batted last summer’ movies, the horror of the Ashes are bearing down on us, threatening to gut our hopes of victory with the bloodied hook of 2AM batting collapses.

Or some other, less strained metaphor not built on a 90s teen horror movie franchise best left forgotten.

But with Darren Lehmann now inserted as the BUPA healthier version of former coach Mickey Arthur, there’s plenty of fresh fodder for speculation in the last fortnight before the series actually begins.

Will Chairman of Selectors John Inverarity be replaced by a flea-ridden sock puppet? Will incumbent Test captain Shane Watson unveil some Machiavellian scheme to retain the team leadership he worked so hard to stumble ass-backwards into on the tour of India? Who, exactly, will be next to punch Joe Root upside the head?

And the big one: what are Australia’s chances of winning the Ashes (or ‘returning the urn’ as some rhyme-happy Twitter hashtags would have it), really?

I can’t answer any of the first three, but I can take a quantifiable stab at the last.

First, let’s note that of the 94 non-drawn Test series between the top eight Test nations over the past decade or so, only 31 have been won by visiting teams. That’s a little less than a one in three chance.

Next, let’s add a little bit more smartitude to this. The ICC publishes Test ratings, which attempt to quantify the relative strengths of the Test nations. With a little bit of digging I was able to find the ratings for all teams for all those series over the past decade or so. How useful are those in predicting the winners of Test series?

Pretty darn useful as it turns out.


Like Duckworth Lewis targets or the running between wickets of Phil Hughes, it’s probably simplest to skip the tedious details of the mathematical calculations and cut straight to the results.

(Although, if you must know, I derived the proportion of the visiting team’s rating to the sum of the two teams’ ratings, converted it to log-odds, then performed a linear regression against a similarly log-oddsified proportion of Tests won by the visiting team [using a Laplacian rule of succession for whitewashes, obviously], with the regression constant factor retained to adjust for home ground advantage. Because what else would one do?)

The result of this quick and dirty analysis? Turns out the team ratings allow you to predict the likely number of wins by the visiting team and get within 0.5 of the correct number a little over 76% of the time.

How does this translate for the Aussies in the upcoming Ashes?

The most likely result by this formula is that Australia will ‘win’ two Tests out of the five. Since draws count as half a point, those two ‘wins’ could be one actual win and two draws. Or four draws. Or, as strange as it may seem, two actual wins.

Whichever way it shakes out, England win the series by a one-Test margin and Australia fail to regain the Ashes (‘spurning the urn’).

What about a drawn series? To draw the series, Australia need to get 2.5 wins out of the five Tests. According to the formula, there’s roughly a 13% chance of that happening. But even if it does, the Poms keep the Ashes, at least until the Australian summer (‘adjourning the urn’).

And what are the chances of Australia winning the series and actually returning the urn? About 6%. Which, by any standard, isn’t great.


But, on the positive side, we’ve smashed the Poms in the pre-series fiasco-a-thon (‘upturning the urn’). And that’s got to count for something, right?