That’s Numberwang Special: Test cricket’s 10-year challenge

Ben Stokes Headingley England Australia Ridiculous Ashes PA

If Test cricket were on Facebook, it’d be an old school pal who occasionally posts something mind-blowingly brilliant but mostly gets by on a steady diet of Likes from rehashing the same old memes you’ve seen 100 times before.

You wouldn’t dream of unfriending it, even if you’ve got newer and more exciting friends now – and the odd blast of genius reminds you of why you liked it so much in the first place.

But a lot of the time it’s going through the motions, sometimes even resorting to attention-seeking Vaguebooking like “maybe I need to shed a day, then more people will like me” or “no-one wants to play me any more, I guess it’s the beginning of the end”.

So if we stretch this analogy even further, what would Test cricket’s 10-year challenge tell us? How did 2019’s Tests compare with 2009’s? And, for that matter, other years ending in 9?

Firstly, some admin. I’ve split Tests by results into five categories: Thrashings (wins by 200+ runs, or by 9+ wickets, including innings victories); Comfortable (wins by 101-200 runs, or by 5-8 wickets); Close (wins by 51-100 runs, or by 3-4 wickets); Thrillers (wins by 1-50 runs, or by 1-2 wickets); and Draws (draws).

This obviously creates the odd absurdity whereby the Stokes/Perera masterpieces of 2019, and Lara’s similar epic of 1999, are lumped in a category with, for example, Pakistan falling 50 runs short at Galle in 2009. Shrug emoji. Genuinely knife-edge games like that trio of one-wicket victories are too rare to deserve a category of their own; a team getting within 50 runs or two wickets of victory usually wins, and if they don’t then it was very close.

If you’re doing an exercise like this, you have to draw the line somewhere – that’s also why, for example, England’s Ashes-clinching 197-run win over Australia at the Oval in 2009 falls in a different category to Sri Lanka’s 202-run triumph over New Zealand the day before, even though neither was really more comprehensive than the other.

Secondly, there are anomalies. This isn’t a scientific data-mining study, it’s a brute force look at results from years that end in 9; as such, the fact 2019 and 1999 happened to feature three of the all-time great fourth-innings nailbiters is not cancelled out. Just as the real 10-year challenge might gloss over the nine years you spent in an iron lung, so Test cricket’s version is not a rigorous statistical scan of the game’s general health.

Caveats aside, what’s the story?


Of 2019’s 39 Tests – 35 of which produced a result – 25 were not even close. India won seven consecutive Tests by massive margins; Australia walloped Sri Lanka, Pakistan and New Zealand; pretty much everyone benefited from England’s unparalleled ability to lose Test matches by a lot. Of the remaining 10 matches, the only tight ones were in Durban and Leeds; every other result in 2019 was by at least 107 runs or seven wickets.

Stuart Broad England Kevin O'Brien Ireland Lord's

That’s not to say they were all dull games. The other four Ashes Tests ebbed and flowed, even if they didn’t converge into a Headingley-style climax; England’s 143-run win over Ireland in July was a lot of things, but very definitely not dull. The emergence of Ireland and Afghanistan as Test nations is a good thing, but will lead to mismatches at first; similarly, it’s hard to describe India’s relentless pursuit of excellence as bad for the game, even when it turns out to be very bad for the opposition.

But the difference with the past is striking. The 41 Tests in 2009 produced 26 results, of which 12 fell into the Thrashings category; 1999 had slightly more, proportionally, with 15 from 29 (out of 43 Tests); there were four Thrashings out of nine results in 1989, five from 14 in 1979 and five from 15 in 1969. Last year’s 83.3% Thrashing Quotient (TQ) is a clear outlier.

This may be because of a greater number of mismatches as the game expands in reach and contracts in quality; it may be because of a worldwide shortage of the technique and concentration required to turn a game around following a bad start, meaning teams who gain the upper hand are more likely to increase their lead as the game goes on.

Or perhaps Tests that in older, slower times would have ended in a draw – because of the sheer length of time it used to take to build up an impenetrable lead – are now more likely to play out to their natural conclusion.


Of the many innovations proposed for Test cricket, none has so far involved changing the number of teams participating in a match. As such, last year’s 39 Tests were contested by a total of 78 XIs. And 41% of them – 32 – were from England, Australia or India. There are nominally 12 Test nations these days, although Zimbabwe did not play any Tests last year – so the Small Eight had to scrap it out for 46 slots between them. To numberwang this even further, that’s 5.75 each, compared with 10.67 each for the Big Three.

In 2009, the Big Three had a similar feast – an average of 11 slots each – but there were more leftovers for the rest, who gorged on 8.17 slots apiece.


The closest of our sample years to parity was 1999, which saw the Small Six given nine slots each, versus 10.67 for each of the Big Three.

But back in 1989, the gap was far wider – with only 21 Tests played, the Big Three claimed more than half the slots, 8.33 each, with the Small Four making do with 4.25.

Doing this for 1979 and 1969 would be anti-historical #maths of the highest order. It would ignore the explosion in Indian power over the global game that happened following the 1983 World Cup, and write off the 1979 iteration of possibly the greatest ever Test side – Clive Lloyd’s West Indies – as “Small”.

That doesn’t mean I haven’t done it – and the trend is still there, with the Big Three still getting a lot more games – but I’m not going to pretend it’s part of the same pattern.

It’s worth noting that several of these years featured Ashes series, which automatically block off 10 – or sometimes back in the day even 12 – slots for England and Australia.

But that’s a symptom of the same process that saw Sri Lanka, Bangladesh, Afghanistan and Ireland claim just 18 of last year’s 78 slots – it’s very much from the Alan Partridge “people like them, let’s have more of them” school of scheduling.

People in England, India and Australia love Tests. Especially against each other. But so might people in the “lesser” countries, if only they were allowed to play the bloody things.


Common sense suggests it’s easier to win a game by taking the last wicket. A dominant side in the third innings of a Test will often either be setting an unassailable target, or bowling themselves to an innings victory. Batting gets harder as the game goes on, hence batting first still being the default option at the toss – and hence the follow-on becoming increasingly rare.

This explains, to an extent, how rare it is for sides to hit the winning runs in Test cricket these days. In 2019, it happened eight times. At the sub-Thrashing level, it’s actually even – five winning runs, five winning wickets – but the big wins are 22-3 in favour of victory being sealed with a clatter of stumps or a rock-and-rolling of DRS rather than a clout through the covers (thanks again, Ben).

But in 2009, with Thrashings thinner on the ground, winning runs were no more common. There were 14 sub-Thrashing wins that year – only two of them were by the side batting last.

1999 saw a much more even split – 17-12 in favour of the side bowling last overall, 7-7 where the margin of victory was more narrow.

But set your DeLorean for 1989 and you’ll emerge in a world where winning runs are the most common way of winning Test matches overall, by five to four. The same is true in 1969, when it was 9-6, while 1979 saw 10 sides fielding last claim victory, compared with four chasing sides.

So there’s very little in the way of pattern, but what we can discern is that as victory margins have grown larger in recent years, the side fielding last – whether it’s in the third innings or the fourth – has assumed more and more control.


Lots of modern Test cricket would be unrecognisable to a visitor from 1969 – or at least, to a visitor from 1969 who’s taken a short cut rather than plodding along at the usual rate of one year per year, watching lots of Test cricket on the way.

They’d see more positive results, for sure – that one’s a definite pattern, which has been well covered elsewhere. And while I can’t stress enough how unscientific this whole thing is, I’m pretty confident the trend towards bigger margins of victory is a real one.

But despite all the changes, what our time traveller would be watching would still be recognisably the same basic game.

Nothing has killed Test cricket so far, despite the regular bouts of agonised fretting over its future.

Maybe that’s the moral of the story – patterns come and go, but Test cricket is still the same lovable, infuriating, stupid, brilliant game it ever was. Contorting its peculiar, anachronistic schedule to fit a championship format won’t spoil it, nor will getting rid of the fifth day.

Then again, perhaps there is no moral to this story, and it’s just a bunch of stuff that happened. And long may it continue.

By Tom Evans