October 19, 2013

Is Test cricket in better health than ever before?

We tend to believe the game was better and more competitive in the past. But the truth as seen through the numbers is more complicated

Brian Statham averaged five runs fewer per wicket than fast bowlers of the '80s and '90s, but that only indicates that batsmen have become more attacking over the years © PA Photos

How has Test cricket evolved over the past 80 years or so? Where does it stand today? There are a number of opinions about this. Many of these views have acquired the status of truisms. Contemporary fast bowling is not as good as the fast bowling in the '70s, '80s and even '90s. Batting techniques have declined. The quality of cricket is not what it used to be in Tests. Fielding has improved out of sight. Tailenders bat better these days compared to earlier eras. Test cricket is dying because wickets are consistently dead and boring. You hear all these points and more from pundits, partisans and pressmen alike. To the trained ear, they contain familiar amounts of nostalgia - the tendency to view the past only in terms of its highs, setting aside its lows.

The figures suggest a more complicated picture of the evolution of the contest between bat and ball.

A close look down the list of Test cricket's most successful bowlers at their respective peaks suggests that a certain type of bowler does not exist these days in the game's foremost format. This bowler took a Test wicket once every ten or 11 overs (that is, a strike rate of 60-66), but conceded less than 25 runs for each of these wickets. This was not a shock bowler like Malcolm Marshall or Waqar Younis or Fred Trueman, all of whom took their wickets at the rate of one every seven to eight overs. The art of this type of bowler was to keep the runs down, to give the fielding captain control on good wickets.

Brian Statham was perhaps the foremost exponent of this art. He was genuinely quick, bowled over a 100,000 deliveries in his 18-year career in first-class cricket, took over 250 Test wickets and 2200 first-class wickets (this meant he bowled nearly 1000 overs per year!). He took a Test wicket every 63 balls, and conceded 24.8 runs in return. Alan Davidson, the Australian left-armer is another example of this type of bowler. He took a wicket every 62 balls, at a cost of 20.5 runs. Davidson occasionally also bowled offbreaks. Keith Miller and Neil Adcock are two other examples.

In a later era, Kapil Dev took his wickets at about the same rate at Statham. But each of Kapil's wickets cost him five runs more than Statham. By Kapil's era, a fast bowler who took a wicket every 10-11 overs was conceding about 30 runs per wicket. Chris Old, Heath Streak and Rodney Hogg conceded about 28 runs per wicket, and took a wicket every 62 balls. At the higher end of this range, Danny Morrison took his wickets at 35 apiece. Michael Kasprowicz 33 apiece. In this era, bowlers whose wickets cost less than 25 runs each tended to have strike rates in the low 50s or high 40s. Elite fast men like Waqar Younis, Allan Donald and Malcolm Marshall were taking wickets at the rate of one every seven or eight overs

Today, we have Dale Steyn, who takes a Test wicket every 41 balls, at a cost of 22.6 runs each.

This does not mean that Statham was more accurate than successful bowlers from the '80s or '90s. It suggests that batsmen in more recent times have been more willing to play strokes. A friend of mine argued, persuasively, that stock deliveries that were guaranteed to be "dot" balls in earlier eras are no longer so. These days, there are more batsmen capable of playing more strokes in more Test teams than ever before. The proliferation of limited-overs cricket with its tremendously skewed contest between bat and ball must have something to do with this. Limited-overs cricket provides batsmen with an incentive and an arena in which they can cultivate attacking shots. The reverse sweep makes a regular appearance in Tests these days. It is a matter of time before batsmen are playing the scoop.

In one sense, I agree with those who argue that Test cricket has become a batsman's game over the decades since World War II. Bats have become heavier and better, protective equipment has improved, laws have been changed slightly in favour of batsmen. But this has had a counterintuitive effect as the figures below show. They have meant that batsmen have been willing to play more shots, take more risks. Batsmen are getting out more frequently than they used to, with the result that there are fewer draws. It is true that the larger number of results has something to do with superior drainage facilities at contemporary cricket grounds (resulting in substantially lesser time lost), but the strike-rate figures should not be affected by superior drainage. It has also been suggested that tailenders tend to be better batsmen these days than in earlier eras, but there is no evidence to suggest that this is the case.

Decade-wise results
Decade Drawn matches Result matches % drawn
Matches Average Matches Average  
1870s 0 0.0 3 18.7 0%
1880s 4 31.4 25 17.9 14%
1890s 6 29.2 26 24.4 19%
1900s 10 30.4 31 23.9 24%
1910s 4 27.6 25 27.5 24%
1920s 16 41.3 35 31.2 31%
1930s 36 39.9 53 28.8 40%
1940s 22 39.5 23 32.5 49%
1950s 51 36.2 113 25.9 31%
1960s 88 36.5 98 29.1 47%
1970s 84 39.5 114 28.9 42%
1980s 122 39.9 144 28.0 46%
1990s 124 39.5 223 28.5 36%
2000s 114 44.7 350 31.5 25%
2010s 39 42.4 113 31.2 26%

While over rates have declined, and the laws, the conditions and equipment have all developed to make batting easier, batsmen are getting out more frequently and outright results are more frequent. Between 1960 and 2000, two out of five Tests were drawn. Since 2000, one in four has been drawn. Between 1960 and 2000, on average, 355 overs were bowled in a Test. Since 2000, this figure drops to 331. The cost of a wicket in a non-draw Test in the 1960-2000 period was consistently about 28 to 29 runs. In the 2000 to 2013 period, this rises to 31. Between 1960 and 2000, one Test wicket fell on average every 66 balls. Between 2000 and 2013, this figure was down to 60 - a difference of ten overs per completed innings, and more than one full session per Test match. See figures for wickets taken by bowlers here (i.e. excluding run-outs). While fielding standards have undoubtedly risen, this has not resulted in more run-outs. In fact, the number of run-outs per Test match has declined from 1.2 in the '50s and '60s to one since the 1980s. Batsmen are fitter too.

This makes comparisons across eras very difficult. Harbhajan Singh, for example, took his wickets more frequently than Erapalli Prasanna or Bishan Bedi. Yet, few will agree that Harbhajan is a more attacking bowler than Prasanna or Bedi were. If we were to ask connoisseurs of Test cricket to say when, in their view, bowling was superior in Test cricket, they would probably be quite unanimous in choosing the period 1960-2000 over 2000-2013. Brett Lee, for example, took a Test wicket every 53 balls. Consider the bowlers who didn't manage to do as well as Lee (let alone significantly better) - Imran Khan, Wes Hall, Curtly Ambrose, Wasim Akram, Jason Gillespie, Andy Roberts, Peter Pollock, Craig McDermott, Courtney Walsh and James Anderson. All these bowlers have better bowling averages than Lee. Darren Gough took his wickets more frequently than all these bowlers, including Lee. He also did slightly better than Dennis Lillee, Jeff Thomson, Ian Bishop and Glenn McGrath! Yet, if all the fast bowlers named in this paragraph are ranked by experienced cricket watchers, Lee and Gough are likely to be nearer to the bottom than the top of the list.

The influence of one-day cricket has to be acknowledged here. There is less evidence to suggest that T20 has had an impact. This could be because, while Test and ODI XIs were more or less the same across the cricket world for many years, T20 and Test teams have tended to include different players from the outset. T20 came of age after the era when the ODI specialist had already arrived.

Test matches today see far more action on a ball-by-ball basis than ever before. Batsmen take more chances, wickets fall more frequently, and results are achieved with greater regularity. This makes bowlers who can control the game even more valuable than they used to be. McGrath was probably the last great bowler who could bowl in the Statham-Davidson mould. Steyn is a strike bowler who concedes runs at about the same rate as Lee did.

Evolution in Test cricket is essentially evolution of the contest between bat and ball. It can be considered to something akin to "evolution" over a period of time, because it occurs due to intended and unintended consequences of adjustments to the laws and the conditions made by many different people over a period of time. While batsmen have apparently been ascendant, they have also been getting out more frequently, thereby producing greater numbers of outright results.

It is one of the paradoxes of our age that those of us who are doing better than others also complain the loudest. So it is with Test cricket. In an age when more Tests are being played than ever before, with fewer draws than ever before, for more money than ever before, predictions of Test cricket's demise are also just as loud. Tests in England have been sold out, close finishes to Tests in India have brought the country to a standstill (and it's a big country!), and India are scheduled to play Tests more frequently in the next few years than they have at any time in their history (with the possible exception of the 1978-80 period). Today there are more competitive Test teams than ever before.

Test cricket is thriving on the field. Off the field, it's another story.

Kartikeya Date writes at A Cricketing View and tweets here