Numbers don't lie: Measuring the dominance of Steve Smith's Ashes

Steve Smith’s Ashes campaign was one of the most dominant performances with the bat in the Ashes or any other series.

An average over 100 for the second Ashes series in a row justifiably had the cricket scribes trotting out comparisons with the Don.

But just how best to rate his performance in the context of Ashes history? While up there with some of Bradman’s series, his series average of 110 didn’t quite match up to the Don’s best (139) or, indeed, Smith’s own 2017-18 series average (139).

Likewise, playing only four Tests, his series aggregate of 774 fell short of a few totals set by the Don, Wally Hammond and Mark Taylor.

However, the best way to measure relative dominance might be to compare the output of the best batsman with other batsmen in the same series, who faced the same conditions, rather than just aggregates and averages across different series.

For example, piling up runs against Stuart Broad and Jofra Archer on pitches doing a lot was arguably a tougher feat than churning out big scores against the revolving door of mostly journeymen bowlers who fronted for England in 1989, when six Australian batsmen averaged over 55 for the series.

At the same time, measuring Smith in comparison to his teammates and opponents begs the question: what if the other batsmen in the series didn’t rate very highly in historical terms?

Was it a little easier for Smith to shine in comparison to the likes of Stokes, Root and Labuschagne than, say, for Bradman in 1930 whose contemporaries in the series included such legends as Hammond, Hobbs, Sutcliffe, Ponsford and McCabe?

That’s a fair point, but not one that can be easily resolved by statistics. It’s still worth measuring how well a player did compared to others in the series and assessing that in historical perspective.

With these points in mind, we at CricTragic.com wheeled out the big wooden abacus and crunched the numbers for Ashes series going back to 1900 to compare the most dominant performances by batsmen in terms of series averages and aggregates.

While averages are better than aggregates for judging batsmen’s output over a few years or a whole career, in a single series a couple of not outs can lend an unwarranted impression of superior performance to averages (for example, Steve Waugh averaged 127 in 1989 compared to Mark Taylor’s 84, thanks to four not outs, despite scoring 336 fewer runs than Taylor.) We need to take both into account.

So we’ve looked at the series where one runscorer outstripped the rest by a big amount and looked at the margin by which he did so: by what percentage were his aggregates and averages bigger than the next best?

We then averaged those two figures to get a composite score for “dominance”. E.g. exactly doubling both the next best batsman’s aggregate and average would give a dominance score of (100 + 100) / 2 = 100.

If the next best bat was on the other side, we also asked how the premier batsman compared to his next best teammate, who faced the same bowlers and conditions?

Also, what proportion of total runs in the series did the top batsman make? And if they missed a game like Steve Smith in 2019, what proportion of total runs in the games they actually played did they score?

The results show a mix of both expected and possibly surprising results. Bradman’s phenomenal 1930 series (974 runs at 139) still tops the list in terms of relative dominance.

His dominance score of 92 reflects how far he outscored Herbert Sutcliffe, who compiled 436 at 87. But it only just pipped Steve Smith’s dominance score of 88 in 2019, in comparison to Ben Stokes.

However, Bradman’s dominance compared to the next best Australian, Bill Woodfull, was significantly higher than Smith’s margin above Labuschagne.

Bradman did amass the highest percentage of total runs scored by batsmen in a series, with 18.2 percent in 1930. Smith’s share in 2019 was 15.6, but it rises to 18.2 percent as a share of the four Tests that he actually played in.

And the biggest share, if we include series where a batsmen played only three Tests, was by Geoff Boycott in 1977, when he made 18.6 percent of runs in the matches he played.

Boycott had missed the first two Tests before ending a three year self-imposed exile from the England team. He still managed to top the aggregates in a low scoring series, and thanks to an average of 147 compared to the next best (Bob Woolmer on 56) ranks third on our table with a dominance score of 87.

The next most dominant Ashes batting performances were also by Englishmen – Len Hutton and Wally Hammond, both with scores of 74, in 1950-51 and 1928-29 respectively.

It is notable that four of the five most performances were on tours away from home. Special mentions also go to Steve Smith in 2017-18 (70 percent above the next best, Shaun Marsh, lest we forget).

Also to Peter May (62 in 1956) and Bradman (59 in 1936-37).

Beyond the raw numbers, how else should we compare these performances in terms of quality? Bradman’s and Hammond’s 900-plus series were high-scoring affairs: e.g. seven batsmen averaged 50 or more in 1930.

On the other hand, the biggest series of Smith, Hutton and Boycott were low-scoring affairs. Only four players averaged over 35 in 2019: by contrast, in 40 Ashes series going back to 1921 at least top order batsmen averaged more than 35.

Whereas only five series saw fewer than four average better than 35: 1981, 1978-89, 1953 and 1956 (which makes May’s 1956 series average of 91 look pretty phenomenal).

Arguably, Hutton faced the most difficult bowling opponents: Ray Lindwall, Keith Miller and Bill Johnston, all of whom rank in the top ten of Australian bowling averages since 1920, and Jack Iverson, who took 21 wickets at 15 over that 1950-51 series, which were in fact the only five Tests he played.

Jofra Archer might make it into that bracket one day, but Broad, Woakes and Leach will not.

However, Hutton’s runs were in a losing cause – England went down 4-1. It’s hard to put Boycott at the top, simply because he only played three of five games and not outs contributed to an outsize average. Smith’s effort, on the other hand were crucial to achieving a drawn series and retaining the Ashes.

While he didn’t quite equal the record of seven consecutive scores of 50 and over in Tests, I would be very surprised (though haven’t sought to check) if anyone else has ever scored six consecutive scores of 80 and above.

Of course, Bradman in 1930 was instrumental in securing a 2-1 series win for a team with a fairly ordinary bowling line-up apart from Clarrie Grimmett.

And he shone in a field that included no less than twelve batsmen who finished with career averages above 45: Bradman, Hobbs, Sutcliffe, Hammond, McCabe, Ponsford, Hendren, Leyland, Woodfull, Duleepsinghi, Fairfax and Archie Jackson.

Only three of the 2019 crew have achieved that mark to date – Smith, Root and Warner.

Here are the figures showing the most dominant series batting: