By Charles Davis
For some years now, run out credits to fieldsmen have been recorded in Test Match scores. It is worth asking, who have been the most prolific fieldsmen when it comes to run outs? Thanks to sources such as Ray Webster’s First-Class Cricket in Australia, and to research for overseas Tests, it is possible to answer this, up to a point, for Tests involving Australia. The limitation is that about 7% of run outs have no known credited fieldsmen, mostly involving Tests played long ago, so figures for early players may be a little low.
The most run out credits for non-wicketkeepers are:
Run Outs Tests per 10 Tests
JB Hobbs 12 41 2.9
SR Waugh 12 168 0.7
RN Harvey 11 79 1.4
DG Bradman 9 52 1.7
DR Martyn 9 56 1.6
CL Hooper 8 25 3.2
IM Chappell 8 75 1.1
RT Ponting 8 88 0.9
It is remarkable that an Englishman heads the list, given that we are limited here to Tests involving Australia. Surprising, too, that modern players do not feature more, given the profusion of matches today, and the fact that there are no gaps in modern data.
The most successful keepers have been Rod Marsh and JJ Kelly (a 19th-century player) on 14, with Gilchrist, Healy, Oldfield, and Grout credited with 13 run outs each.
The unbroken 10th-wicket partnership of 22 between Stuart MacGill and Nathan Bracken to win the Sheffield Shield (PC) final was the highest last-wicket stand to win a first-class match in Australia since 1972, when South Australia’s tailenders McCarthy and Hendricks put on 51 to beat NSW at the SCG, with 17 minutes to spare.
When Glenn McGrath took the astonishing figures of 8 for 24 against Pakistan in Perth, he still did not win the Man of the Match. How often does this sort of thing happen?
Batsmen are routinely favoured when it comes to MoM gongs, and wicketkeeping, for most, must remain its own reward. Of the last 282 MoM awards, 60% have gone to batsmen and only 32% to bowlers. 8% went to all-rounders or keepers, but these tended to favour batting feats. If a bowler takes the best bowling figures in a match, he only has a 26% chance of the award, whereas scoring the most runs in the match gives you a 47% chance.
In terms of its rarity, McGrath’s 8 for 24 is equivalent to a batsman scoring a triple century, taking into account the extremely low cost of the wickets. Yet the award went to Justin Langer, who scored 191 and 97. I’m sure the judge could justify his choice; its just that these justifications too often seem to work in the favour of batsmen, over bowlers and keepers.
Curiously, as performances even out over a full series, bowlers seem to get more reward. They win about 45% of “Man of the Series” awards, a far better return than the 32% figure for “Man of the Match”.
It’s funny how prejudices can colour our memories. Jack Fingleton, Test player and journalist, was often a voluble critic of his contemporary Don Bradman. Among his criticisms of the Don was a claim that Bradman was a strike hog. He wrote, in Batting from Memory, that “it was not possible to have much of the strike while Bradman was there. He was such a fleet and superb runner between the wickets that he always managed to manipulate the strike.” This is not a trivial charge. The implication that a teammate plays selfishly is a grave allegation in the eyes of Australian cricketers.
Time for a reality check. Ball-by-ball Test records show that, during Fingleton’s Test partnerships with Bradman (mostly in 1936-37) it was Fingleton who received over 58% of the strike to Bradman’s 42%. In their world-record sixth-wicket stand of 346 at Melbourne, Fingleton faced 412 of the 698 balls bowled (59%) to Bradman’s 286 balls. It was much the same story in their other partnerships.
The imbalance probably has more to do with Fingleton’s very slow scoring (hence fewer strike-rotating singles) compared to Bradman, than with any conscious attempts to gather the strike. It is possible that the Don sometimes attempted to manipulate the strike to restore some sort of balance, but if so, he was mostly unsuccessful.
Where records survive, Bradman averaged 51-52% of the strike with all partners. The slight excess above 50% can partly be put down to Bradman’s batting with tailenders, when he was within his rights to manipulate the strike. In his two great partnerships with Bill Ponsford in 1934 (388 at Leeds and 451 at the Oval) Bradman received 50.8% of the strike.
You would think from reading about the game that today’s international cricketers are playing more than ever, and that the workload is getting too much for some senior players. But how does the amount of cricket played by today’s players compare to earlier eras?
Here is a selection of Australian players from different eras, at the peaks of their careers, and the number of days of senior cricket (first-class and “List A’ one-dayers) they played:
C Hill 1899-05 54 days per year ML Hayden 2001-04 101
DG Bradman 1929-33 60 days per year JL Langer 2001-04 96
RN Harvey 1953-57 80 AC Gilchrist 2001-04 100
WM Lawry 1961-66 75 RT Ponting 2001-04 100
GS Chappell 1972-77 86 GD McGrath 2001-04 79
AR Border 1986-91 108 JN Gillespie 2001-04 83
ME Waugh 1995-99 114
In short, cricket workload has increased, but only over the long term, and not as much as you might think. None of today’s senior players play quite as much cricket as Mark Waugh ten years ago, or Allan Border almost 20 years ago.
Mark Waugh actually played 173 days of cricket in one calendar year, 1995. He was able to combine his international duties with a full season of county cricket for Essex. The fact that, even in recent years, Australian players have chosen to play county cricket whenever there was a break in the international program, tends to undermine their claims that they are being overloaded.
The workloads today are only about 20% higher than they were 50 years ago (Harvey). And note that players like Harvey, Hill and even Bradman, probably played a lot more minor cricket than today’s stars.
If you look at the English professional game, high workloads extend even further back. As early as the 1890s, Tom Hayward of Surrey was playing more than 100 days a year, while in the 1940s, Dennis Compton averaged almost 120 days per year. In Hayward and Compton’s time, a day’s cricket in England usually meant more than 130 or 140 overs. Today, 90 overs is a typical output.
The Spoils of Duckworth/Lewis
The Duckworth/Lewis system for deciding the winner, or modified targets, in interrupted One-Day games, should be an even-handed system that doesn’t change the odds either way. Theoretically, it should share victory between teams batting first and second in the same proportions as complete games.
However, when we look at all the cases where D/L has applied, a trend emerges. There have been 34 matches decided by D/L since its introduction in early 2001. Of these, one has been tied, 24 have been won by the team batting first, and nine won by the team batting second. This strongly suggests an imbalance; in uninterrupted 50-over matches, the team batting second wins 50.3% of the time. In other words, ODI results are very even when no interruptions occur, but D/L favours the team batting first by a ratio of almost three to one!
It is possible that this imbalance is a statistical fluke, which will even out over time. However, the binomial theorem suggests that the probability, of getting a 24 to 9 imbalance by chance alone, is 1% or less. It is hard to say what could be wrong with the D/L method, which is based on sound principles, but if the imbalance continues, then the method may need to be evaluated.
An interesting new fashion in Tests has seen captains turning away from enforcing the follow-on. Once a rarity, batting again when leading by over 200 is fast becoming normal. Stats tell the story: from 1975 to 2000, the follow-on was enforced 93% of the time. In the Tests since then this figure has dropped to 77%, and in 2004 it was below 50% (six out of thirteen). How successful is the change in strategy? Overall, since 1975, teams enforcing the follow-on have won 81% of the time, while teams in similar positions, but choosing not to enforce, have won over 93%. There have been two losses by teams enforcing the follow-on, both well known to Australian supporters - Leeds 1981 and Kolkata 2001. The last time a team lost after choosing not to enforce the follow-on was in 1950.
With Tests being played so frequently, captains are probably becoming more aware that "burn-out" of bowlers by enforcing the follow-on can not only affect teams within matches, but in subsequent matches also.
Michael Clarke feathered the Australian selectors’ caps by becoming only the fourth player to score centuries in his first Test innings both home and away. The others are Azhar Mahmood of Pakistan in 1998, Kepler Wessels (playing for Australia) in the 1980s, and a name from the very distant past, Henry “Harry” Graham, who played in the 1890s. One wishes Clarke a happier career than Graham, who played only six Tests and died young. Azhar Mahmood, also, failed to live up to early promise, all three Test centuries coming early on. Kepler Wessels, of course, had a much more substantial career, and even scored another century on “debut” when he changed countries to represent his native South Africa.
If second innings of debuts are included, Lawrence Rowe of West Indies (1970s), and the Indian prince Kumar Shri Ranjitsinhji (1890s) also scored hundreds and their home and away debuts.
After taking 6 for 9 in a freak bowling spell in Mumbai, Clarke missed out on a bowl in his next Test in Brisbane. The last player to bag 6 or more wickets in one Test, and then not get a bowl in his next (barring cases of injury, or abandoned matches), was Clarrie Grimmett in 1932, when he took 14 wickets against South Africa at Adelaide, only to be surplus to requirements in Melbourne when South Africa were bowled out for 36 and 45.
In the classic Sydney Test of 1894, Australian keeper and captain Jack Blackham was injured during England’s second innings and had to leave the field. George Giffen took over as captain, put himself on to bowl in the very next over, and then bowled 50 consecutive overs, taking 2 for 102, before England were finally dismissed. This is the longest-known bowling spell in any Test in Australia. Australia lost the match by 10 runs. Giffen holds another little-known record: he faced over 60 balls before he scored his first run in Test cricket.
When Australia and England played a tie in the NatWest series final at Lord’s in early July, the result kept alive – by the barest margin – a golden run for Australia under the pressure of finals cricket. The last time Australia lost a tournament final was in 1999, in the Aiwa Cup in Sri Lanka. Since then, Australia has won thirteen finals, with one washout and one tie. Australia performs much better in finals that in run-of-the-mill ODIs, where they lose about 20% of the time, sometimes, as was seen against Bangladesh two weeks earlier, to weak opponents. Their peerless record when the pressure is on suggests a psychological component: it is mirrored by the fact that Australia, while losing the occasional Test, lost only one “live” rubber in 51 Tests from 2001 to March 2005. That loss was to India at Adelaide 2003-04.
Australia’s sequence of four losses early in the 2005 Ashes tour, including a Twenty20 game, had historians struggling for precedents. Finding another example of an Australian Ashes touring team losing four in a row took some detective work, but it did happen, just the once, over a century ago. The year was 1890, and the Aussies lost four first-class matches consecutively in June of that year. And even that team did not have the burden of two scandals to throw into the mix. It was in an era when Australian cricket was at its all-time lowest ebb, so the losing sequence was a good deal less surprising than the 2005 case. There is a more recent precedent if tours to other countries are considered: Australia lost six matches in a row on tour to India in 1997, most of them One-Day Internationals.
Australia’s loss to Bangladesh in the ODI at Cardiff was widely described as the biggest upset in ODI history. Perhaps it was reminiscent of a one-day defeat suffered by Australia in the days before ODIs, when Bob Simpson’s team lost to Holland by three wickets in 1964.
There are various factors behind the increasing success of bat over ball, but certainly improving bats and smaller fields (due to boundary ropes) must be having an effect. These are difficult to analyse statistically, but one measurable factor is the increasing proportion of boundary hits. Apart from a brief spell at the peak of the Golden Age (1895 to 1905) where the incidence was over 50%, for most of Test history, about 46-48% of runs were scored as boundaries. In the late 1980s, this proportion began to rise, slowly at first, but passing 52% in the late 1990s, recently reaching 56%.
The change can be illustrated anecdotally. When Chris Gayle hit 317 at St. John’s (tiny ground , fast outfield) against South Africa, he hit 37 fours and only 2 threes. When Bob Cowper hit 307 at the MCG in 1966 (big ground, very slow outfield, no boundary ropes), he hit 20 fours and 26 threes. Overall the conditions probably lost Cowper over 50 runs. The ratio of fours to threes at St John’s is almost 8 to 1, whereas in Australia it is closer to 4 to 1. Another recent report tells how Hershelle Gibbs used at least 47 different bats in a single year. These bats were needed for no more than 47 innings in senior cricket in 2004, scoring just 1547 runs. Apparently the bats are not rolled or hardened, thus offering more spring and power, but wearing out incredibly quickly.
Cricket has become more entertaining, so perhaps we shouldn’t complain, but if scoring levels change permanently, the game loses contact with its traditional standards, and that would be a real loss. The place in history of our leading players will become uncertain.
Australia’s top-order collapse in the Twenty20 match against England at Southampton on June 13 was actually one of the most extreme in international cricket history. The Australians at one stage were 7 for 31: to find an instance of Australia losing so many wickets for fewer runs, in any full international match, one has to go back to the last day of a Test at the Oval in 1896, when Australia lost 7 for 14 on an unplayable wicket. It is even worse when we consider that Australia at one stage had been 0 for 23, so losing its first 7 wickets in the space of 8 runs and 20 balls. No team, in Test, ODI, or Twenty20, has ever given up its first seven wickets in so short a span. The previous worst was the West Indies, who went from 0/16 to 7/25 in 36 balls in an ODI against Zimbabwe (of all teams) in 2000-01.
Charles Bannerman’s 165 (retired hurt) in the very first Test match in Melbourne in 1877 remains one of the most famous Test innings. However, its statistical features have largely remained a mystery, so readers may be interested in the facts of the innings, derived from an over-by-over analysis:
Bannerman score 27 runs before lunch and 99 afterwards on the first day. His scoring was extremely variable in speed, and depended on the match situation. He faced about 100 balls going from 25 to 50, and hit only two 4s in that 50, and at one stage did not score for 14 (four-ball) overs. However, he then hit eight more boundaries in the next half hour, going from 50 to 100 in barely 30 minutes off only 40 balls. He reached his 100 out of only 136 off about 190 balls in the 93rd over. He became more circumspect after the loss of two quick wickets and finished the day 126 not out off about 235 balls in 115 overs. He scored over three-quarters of Australia’s runs on the first day.
Bannerman played cautiously the next morning, moving to 159 at lunch after a short session. At 7 for 240, he was struck on the right hand by Ulyett and retired hurt shortly thereafter (his glove was faulty). He finished with 165 in about 295 minutes off about 330 balls in 167 overs. He hit 18 boundaries, 15 of them in the arc from cover to mid-on.
The England team for the first Ashes Test in 2005 was, in one sense, the least experienced team in Ashes history. The players were able to boast only 25 Ashes Tests between them, the smallest total for any England team since the early 1880s, when Test cricket was just getting established. The only Australian team since the 1880s with less experience than this played in 1978-79 during the Packer upheaval. For England, some otherwise established players like Flintoff and Strauss were making their Ashes debuts. Flintoff, in fact, set an all-time record by playing 47 Tests before meeting Australia for the first time. The Australians, by contrast, went into the series with a combined total of 131 Ashes Test caps, although it is surprising to find that the batting stalwarts Hayden, Langer, Martyn and Gilchrist had each played only ten or eleven Tests against England.
The most experienced team in Ashes history, in terms of Ashes Test caps, was the Australian team of 1905, with a total of 239.
“When you win the toss, usually you bat. At other times, you think about sending the opponents in to bat, and then you choose to bat first.”
Generations of captains have followed, by word or deed, this piece of conventional cricketing wisdom. Traditionally, they have been supported by the fact that teams batting first have won more often than they lose. Yet, slowly but surely, this pattern has changed, and changed so gradually that few commentators have noticed. The fact is that, these days, Test cricket favours the team batting second. The traditional advantage of batting first, which before WWII gave rise to a 59:41 win:loss ratio, has virtually been reversed in the last ten years (400 Test matches). Since 2000, teams batting first have won 83 and lost 117 Tests.
There are various factors at play which may be deserving of a detailed analysis, for another time perhaps. However, one observation: there is a little-recognised problem in winning cricket matches for teams batting first. This is because teams batting first must usually score an excess of runs to secure victory, and this uses up precious time. (Tests won by teams batting first are, on average, 10% longer than those won by teams batting second.) When time constraints come into play, most captains will delay second-innings declarations until the probability of defeat is minimal. Quite often, this presents the team batting last with the chance to escape with a draw.
Bruce Mitchell’s 58 for South Africa v Australia at the Gabba in 1931 was hardly an earth-shaking event. However, it may well hold two world records. Mitchell’s innings started on Saturday, 28th November . Thanks to a rest a day and long rain delays, he was not out until the following Thursday, four days and 21 hours later. It was the longest innings, in elapsed time between first ball and last, ever played in a Test match.
On the Wednesday, there were only two hours of play, and Mitchell added only eight runs, failing to score for the first 90 minutes. Analysis of a detailed description in the Brisbane Courier produces an estimate of 95 balls faced in those 90 minutes, the longest standstill known in any Test. Trevor Bailey possibly faced more in a scoreless spell for England v South Africa at Leeds in 1955, but we don’t know for sure. J.T. Murray faced 80 balls before he scored in Sydney in 1962-63. Murray was batting with a serious shoulder injury in a defensive situation.
Mitchell’s eight runs in two hours is the fewest runs known for a batsman in a full 2-hour session, in Tests involving Australia. Arshad Khan scored only 5 in a session for Pakistan v Sri Lanka at Colombo in 2000-01, but, thanks to slow modern over rates, this involved far fewer overs than Mitchell’s session (27 overs vs 46 for Mitchell).
Bruce Yardley’s 74 off (about) 48 balls for Australia at Bridgetown in 1978 is not remembered as one of the great innings. Yet perhaps it deserves to be. In facing either 29 or 30 balls for his first 50 runs, Yardley recorded the fastest half-century, in terms of balls faced, by any Australian in Test matches (the uncertainty arises because one of his scoring shots, although recorded by the scorebook, probably should have been credited to his batting partner). Most remarkably, Yardley hit two sixes and five fours in three overs by Joel Garner, who was perhaps the most formidable bowler of the day. This was during a bowling spell where Garner took four wickets in six overs without conceding a run to any of Yardley’s batting partners. Yardley hit nine fours and two sixes in his first 53 runs. He had come to the wicket after a major batting collapse, and his 74 was the second-highest score of the match.
For a while there, the dubious tactic of using nightwatchmen in Tests was abandoned by Australia under Steve Waugh, but it has now returned. Captains persist with the tactic in complete absence of any hard evidence that it actually works. A good example of why it often fails was demonstrated in the Australia-New Zealand Test at Auckland.
Jason Gillespie was promoted up the order when Martyn was out at 4 for 215 shortly before stumps. This demoted Adam Gilchrist, who had just hit three Test consecutive centuries (the first Australian batsman to achieve this since Bradman), to #8 in the order. No batsman in such form has ever been forced to bat so low in the order. The predictable happened. Gilchrist, in spite of scoring at a run-a-ball, ran out of partners when he was 60 not out. Earlier, Gillespie had batted two and a half hours. In his rightful place in the order, he would in all probability have helped Gilchrist to a much larger score. As it was, Gilchrist’s chance of equalling Jack Fingleton’s record of four consecutive tons was dashed.
The issue of umpiring favouring home teams arose again during Pakistan’s tour of Australia in 2004-05. This is a contentious issue that would require a lengthy analysis, but in the absence of that, lets look at just one peculiar fact: Adam Gilchrist has never been given out lbw in a Test in Australia. And he has fallen lbw only twice in ODIs at home.
Two dismissals lbw in a total of 121 dismissals (45 Test, 76 ODI). That is a remarkable departure from the normal figures; a typical left-hander would expect about 13 lbws, allowing for the surprising fact that lbws in ODIs are much less frequent than in Tests. Away from home, Gilchrist is much more likely to be lbw, in 21% of his Test dismissals, and 8% of his ODI dismissals. Nothing unusual about those figures, but they do highlight the home game anomaly. Another unusual feature is that most of his lbws occur at low scores. He has never been given out lbw in double figures in a Test innings, home or away. When Gilchrist gets going, he is incredibly dominant. Could it be that umpires are influenced, perhaps unwittingly, by this dominance?