first ever Test between South Africa and Pakistan at Johannesburg in 1994-95, the injured John Commins
came in to bat using a runner (perhaps the fastest runner in world cricket,
Jonty Rhodes, who had been out the previous ball). Off his first ball, Commins completely forgot about the runner and set off on
a run, and was hopelessly run out going for a second as confusion reigned. He
collapsed on the ground, having aggravated his injury. In a final insult, the
umpire called the run a leg bye, so Commins was out
for a golden duck.
UPDATE: although the newspaper report that this is based on specifically describes a leg bye, the ball-by-ball record shows that no run was completed, and the published score confirms this. Commins was run out at the striker’s end. There was a leg bye earlier in the over, before Rhodes was out.
Have I discovered a new Test cricket record?
The New Zealand Herald reported that, on the fourth day of the Bulawayo Test in 1992 (Zim v NZ), "play started in front of one paying spectator". If true, this is the smallest recorded crowd at a Test match where spectators were permitted. There has been a small number of other Tests where spectators were excluded for security reasons on one or more days.
Of course, it is possible that the reporter was exaggerating for effect.
A curious moment in cricket history that Wisden may prefer to forget: In March 1982, Wisden announced that the matches then being played by a ‘rebel’ English team against South Africa would be recognised as Test matches. The announcement met with immediate criticism, and Wisden never followed through.
In 1996, the Association of Cricket Statisticians and Historians reported that pre-1970 Wisdens contained errors in 70 per cent of their scorecards of county matches. They determined this by comparison with official county scorebooks.
A couple of points in Wisden’s defence: 1) many of the errors were trivial, for example confusing byes with leg byes. 2) I have found plenty of ‘official’ scorebooks that contain errors if you look closely enough.
On the fourth day of the Test at Ahmedabad in 1987, Pakistani fielders were pelted with stones by “hooligan” spectators. Pakistan captain Imran Khan refused to continue and about an hour of play was lost. Kapil Dev and Sunil Gavaskar made appeals for calm in Gujarati and Hindi. When order had apparently been restored and play was re-started, eight of the Pakistan players came onto the field wearing helmets. According to The Hindu newspaper, this demonstrated the players’ “sense of humour”; personally, I very much doubt that they were trying to be funny.
The most popular days of the Test year: If I calculate correctly, there have been 113 Tests with play on January 3, and 103 on January 4. I haven't excluded washed out days, but rest days are excluded. Boxing Day is fourth place on 90.
Those 113 Tests occurred in 71 different years, since some of them had concurrent Tests.
In Australia, 'New Year' Tests have a longer tradition than Boxing Day, although they often did not start exactly on New Year's Day.
There have been only 3 Tests with scheduled play on May 9, and in one of those play was washed out.
There have been 87 Test innings with just two individual centuries, where the two centurions did not bat together in partnership. The smallest such innings was 374 by New Zealand at Hamilton 1990-91, where John Wright scored 101 and Andrew Jones 100*, but they did not bat together. There has been only one case where both batsmen reached 150: Auckland 2003-04 where Scott Styris made 170 and Chris Cairns 158.
A Long Day’s Batting
The most balls faced by a batsman in a day’s play, that I know of, is 424 deliveries faced by Alec Bannerman on the third day at Sydney in 1891-92. He scored just 67 runs. There is a small uncertainty about the balls faced (plus or minus) which may be important because Bradman faced 421 in a day during his triple-century in 1934 and 420 in a day during his triple in 1930. There are 3 others in the range 410-420 so it is all very close.
This is a category that is really 'set in stone'. All these marks were set in the days when more than 140 overs could be bowled in a day. There is no possibility that any modern batsman could rival these numbers.
Most balls Faced in a Day of Test cricket: Individuals
I was looking at Test series where one bowler completely outclassed his team mates. The ultimate example (excluding one-off Tests) was Bangladesh in England in 2005, where Mashrafe Mortaza took 4 wickets at 49.5 (in itself, pretty ordinary), but his team mates combined managed only two wickets at 375.5, a difference in average of 326.
For a more substantial series, with a bowler who took ten wickets or more in a minimum three Tests, Ray Bright took 15 wickets at 23.8 for Australia in Pakistan in 1979-80, while the other bowlers (including Dennis Lillee) put together took just 8 wickets at 92.4. In a six-Test series in Pakistan in 1982-83, Kapil Dev took 24 wickets at 34.6 while his team mates averaged 85.
At the other end of the scale, Graham McKenzie took one wicket for 333 runs in South Africa in 1969-70. His fellow bowlers took 64 wickets at 40.5 – still not very good, Australia lost the series 4-0.
A few years ago Benedict Bermange kindly sent me lists of the longest-serving #1 players in the ICC Test ranking lists. He has now updated them in comments on an Ask Steven post. I am copying them here since Facebook posts soon get hard to find.
Benedict has been
involved in the development of the ratings systems for many years and has a
copy of the algorithm. Without that, these numbers would be jolly hard to
Most Matches Ranked #1 Batsman
Most Days Ranked #1 Batsman
Most Matches Ranked #1 Bowler
Most Days Ranked #1 Bowler
I am of the opinion that the ratings system is very sound, but becomes less reliable further back in time. There were probably enough teams and enough Tests played from about 1955 onwards to produce useful rankings; before that, I am a bit dubious. The suspension of Test cricket in wartime, in particular, produces some funny results. Aubrey Faulkner did not take a single wicket during his period as “#1” bowler from 1914 to 1920. The reason is that S.F Barnes, who was #1 (by a country mile) up to 1914, did not play after the war, so his career ratings were terminated. Faulkner, on the other hand, played a single Test after the War (in which he took no wickets) and so is considered one of the few bowlers whose careers straddled the War. Of these bowlers, he had the best career record in 1914, even though he took his last Test wicket in 1912.
There is certainly a case for arguing that Dale Steyn is the greatest bowler of all time, based on these figures.
A report of a Sri Lanka/India Test from 1993 describes Sachin Tendulkar as India’s vice-captain. Tendulkar was barely 20 years old at the time. I don’t have any info on vice-captains, but I wonder if he was the youngest. “Vice-captain” is not an official position, and teams are not required to have one.
Tendulkar filled in as acting captain when Azharuddin was indisposed for part of the third Test.
During the first Test between Sri Lanka at Morutawa in1993, Sri Lanka appealed against Richie Richardson, simultaneously for caught behind and a stumping. Umpire Samarasinghe gave the stumping not out, but umpire Francis gave Richardson out caught behind. Replays showed that both decisions were probably wrong; the upshot was that Richardson was out anyway.
At Lucknow in 1994 against Sri Lanka, Sachin Tendulkar went from 88 to 100 in four balls (4, 0, 4, 4 off Wickremasinghe). Others have done it faster, but Tendulkar’s effort was notable because it was off the first four balls of the day.
Indian paceman Atil Wassan broke a stump when he bowled Mark Greatbatch to take his first Test wicket, at Christchurch in 1990.
I was surprised to see on Ask Steven that Peter Siddle went the most Tests without taking a catch (25). In fact the 25 Tests were the last 25 of his career. Siddle dropped three catches during that sequence.
At Nagpur in 1994 (Ind v WI), the first 16 wickets fell to catches. Only two of them went to wicketkeepers.
Just a few bits and pieces, while I search for inspiration…
Four dropped catches in an innings, since 2001
No one has recorded more than four drops in an innings in this period.
A recent Item in Ask Steven dealt with the longest gap between Tests for a player, in terms of number of matches missed. I wondered which players had missed the most total matches in his career, not just between two appearances. I should have been able to guess the answer.
Most Tests missed during whole career.
Here are the players who both played more than 75 Tests, and missed more than 75 Tests, during their careers
Most number of separate gaps in career
Longest Test careers without missing a single Test.
AR Border missed only one Test during a 156-Test career. Kapil Dev (131 Tests) missed only one Test, and that was for disciplinary reasons.
Debutants who top scored and had best bowling return for their teams (first innings only)
At the start of play on the third day at Christchurch in 1987, Gordon Greenidge, 16 not out overnight, went out to bat without his batting gloves. After five balls (faced by Haynes) he interrupted play to go and get some. After a delay, Haynes was out next ball, to Chatfield. Haynes evidently had something to say to Greenidge about the interruption. Greenidge was out to the next ball after that, to Hadlee’s first ball of the day.
Most balls faced between dismissal in ODIs: Chris Harris in 1997-98. He faced 465 balls between dismissals (9 innings) and batted for 150 overs.
Wellington 1990, a ball from Danny Morrison slipped from his grip and rolled to square leg. Alan Border claimed it, walked over and hit it to the boundary, with Morrison standing (glaring at him, I daresay) a few yards away.
During the first Test at Karachi in 1988-89, the Australians were furious at the umpiring of Mahboob Shah, claiming a number of unfair lbw decisions. The year previously, England had almost curtailed a Pakistan tour over similar issues. As they had in the previous season, Pakistan authorities resisted the Australian complaints and appointed Shah to the second Test as well. The Australians were a little mollified when Shah gave opener Ramiz Raja out lbw, second ball to Bruce Reid.
Shah also gave Javed Miandad out lbw in the same Test (for 107). This was a rare experience for Javed in home Tests.
Most Expensive First Over in Test Cricket (Runs)
Sohag’s over included four leg byes, so his figures at the end of the over were 1-0-14-0. Bartlett’s over would have cost 17 runs using modern counting (a no ball was scored from) but also included four byes. DD Ebrahim scored 15 runs in the Anderson over, the most by a single batsman off a bowler’s first over in Tests. Anderson first four deliveries in Tests were: no ball, FOUR, no ball, FOUR. “…not the greatest of starts for Anderson” said Cricinfo in the ball-by-ball text.
Allan Border’s two highest scores in Tests (at the time) both included incidents where the batsman believed himself to be out, but was recalled.
At Lord’s in 1985, Border, on 87 (the co-called Devil’s number for Australians) had turned towards the pavilion when Mike Gatting juggled a catch at short leg, then appeared to throw the ball in the air in celebration. Gatting, however, made a hash of the whole thing and umpire ‘Dickie’ Bird ruled that he had not controlled the ball sufficiently. Border went on to make 196.
At Adelaide in 1987-88, Border, on 66, appeared to be caught at mid-on by Jeff Crowe. The fielder, however, was not sure if he had caught it cleanly and called Border, who was heading for the pavilion again, back.
In that innings, Border was also given not out to an lbw appeal that Robelinda calls the “Plumbest ever LBW turned down in cricket”. I tend to agree. Even Border looked bemused. At the time of writing, the video is here.
Fastest ODI Centuries on Debut
Perhaps the most enduring important ODI record that could still be broken is Desmond Haynes’ highest score on ODI debut: 148 against Australia in 1978. Since then, almost 3900 ODIs have been played without anyone topping Haynes’ mark, representing almost 99% of all ODIs. Scoring a century on ODI debut is still fairly rare, and there have been only 14; one reason is that with so many matches played, debuts are not particularly common.
Here’s a list of most of the debut centuries ranked according to fastest to reach the century.
For the innings as a whole, Haynes’ remains the fastest century on debut (108.8 runs per 100 balls).
To get a first-100 figure for Haynes, I had to get a copy of the original scorebook, kindly supplied by Colin Clowes at Cricket NSW. Even this did not give a balls faced figure for first 100, but by re-scoring the scoresheet, I was able to get a reasonably accurate figure. The scoresheet was not very easy to read in places, but I am happy with the final figure.
There are a handful of ODI early debut centuries for which balls faced are not available for the first 100 runs, including the first by Dennis Amiss. However, these were not scored at a speed that would rival the highest-ranked.
[I forgot to add that none of these innings compare for speed with Shahid Afridi’s first ODI innings (100 off 37 balls), but that was scored in his second match.]
Pakistan at Abu Dhabi in 2011-12, Graeme Swann took a wicket with the last
ball of an over, but was then taken off because the new ball was available in
the next over. When Swann was brought back on, he took a wicket with his
first ball, and thus took wickets with consecutive balls in the same innings,
but 14 overs apart.
In an ODI
in Guwahati in 1993 between Sri Lanka and South Africa, Rumesh
Ratnayake, who was batting with a runner, was run out when
he set off for a run while his runner did not move from the crease. He
remonstrated with the umpire, who changed his decision to ‘not out’, only to
finally – and correctly – give the batsman out when the South Africans
fielders had something to say.
On the final day of the Sydney Test of 1985-86, against, India, Australian captain Allan Border was out twice, and only scored 10 runs. He was out at 11:18am (for 71, having been on 64 overnight) and again at 5:46pm, for 3. In between, at 3:40pm in Brisbane, his wife gave birth to a daughter.
347 for 4 in their first innings overnight on the fourth day,
managed to nearly lose the match, being all out for 396 against India’s 600
for 4, and losing six wickets for 119 in the follow-on. There had been just
eight wickets in 321 overs on the first four days, but there were 12 wickets
off 104 overs on the fifth.
At Old Trafford in 1948, the Australian batsmen (batting for a draw) did not change ends for the first 38.5 overs. All the runs, more than 50, were in twos and fours. Alec Bedser's first spell was 8 overs: 48 balls, and Arthur Morris faced all 48. Bedser’s spell finished, and there were, of course, other bowling changes during those 38.5 overs.
Ian Johnson was out for 6 off 32 balls, but only faced Pollard.
When Bedser came on for a second spell, Morris was still at the same end and faced the first 23 balls; so he faced the first 71 balls that Bedser bowled in the innings.
At Melbourne in 1883, Bill Murdoch faced up to 48 consecutive balls from Barlow without facing anyone else. He did not score.
Bowling to Exhaustion
I have tried to nail down the most balls bowled in a Test without a bowling change. I already knew that it was probably Ramadhin and Valentine at the Gabba in 1951 (final innings), but there is no surviving score. I think I can say now that it was 69.7 eight-ball overs (plus or minus one, perhaps) which translates to 93 six-ball overs (rounded). The next best known is 86 overs, by the same bowlers at Lord's in 1950.
Ram and Val came on immediately after a wicket in the fourth over (1 for 8), and the score was on 12. They bowled the last 11 overs before tea (very probably). There was a change of ends immediately following tea (one over for two runs by Gomez, in place of Ramadhin), then the 69.7 by Ram and Val extending into the next day. Ramadhin bowled from the Stanley St end before tea and the Vulture St end afterwards.
Australia won by 3 wickets. Note the lack of imagination on the captain's part not to try any other options, especially as Gomez had taken a wicket, and Valentine took 1 for 117.
I found an interesting line about Ramadhin in the match. It says that when called on to bowl, he rolled his sleeves DOWN. I have read somewhere that he once admitted that he bowled in long sleeves to disguise the possibility of chucking.
Australia had some luck. Hassett, on 11, played a ball from Ramadhin onto his stumps without the bails falling.
A quote from Wisden. [Note: the "first five overs" comment is incorrect; it was the first four.]
"Goddard's tactics caused considerable comment. Although Gomez took a wicket in his opening spell, he and Worrell bowled only the first five overs of the innings – they conceded fourteen runs – before Goddard switched to Ramadhin and Valentine, who were called upon to bowl unchanged to the end. Between them they sent down over 80 overs and both gave signs of having been over worked. Valentine lost his usually splendid length and Ramadhin suffered reaction in subsequent games. After calling for the new ball Goddard rubbed it on the ground to remove the shine and asked his spinners to continue."
Sreeram added the following: Frank Worrell said that Goddard was tactically naive and did not know anything besides bowling Ramadhin and Valentine. He relied heavily on the players for his tactics (I don't now remember who these were apart from Worrell. Stollmeyer, Gomez etc possibly)
In the 1951-52 Australia series, the players fell out with Goddard and they stopped providing their feedback to Goddard. So when Australians began to work Ramadhin & Valentine out, Goddard did not know what else to do.
[I would add that team discipline fell apart when Johnston and Ring were playing their match-winning last-wicket stand at Melbourne. Bowlers and other fielders changed the field around when the captain’s back was turned. Ramadhin just walked off the field. Goddard withdrew from the next (final) Test, citing “nervous exhaustion”.]
Two bowlers unchanged (six-ball overs equivalent):
93 overs, S Ramadhin and AL Valentine, Brisbane 1951
86 overs, S Ramadhin and AL Valentine, Lord’s 1950
79 overs, W Rhodes and WE Astill, Georgetown 1930
75 overs, DA Allen and GAR Lock, Calcutta 1961/62
73 overs, S Ramadhin and AL Valentine, Christchurch 1952.
65 overs Iqbal Qasim and Tauseef Ahmed, Karachi 1979/80
Rutherford Does an Alletson (well not quite)
I have carried out an unusual re-score, on a partial score of Ken Rutherford’s 317 off 245 balls at Scarborough in 1986. I have the batting page for this innings (kindly supplied by Jamie Bell at the Cricket Museum, Wellington), but not the bowling page. However, I was able to come up with something akin to a ball-by-ball analysis after I realised that the scoring strokes on the batting page were colour-coded according to the bowler who conceded the runs. Eventually, I was able to identify the bowler and tease out the scoring for each over, although the placement of dot balls within each over could only be estimated. Even so, it was possible to reproduce exactly all the published bowling figures, and the balls faced figures for batsmen (including milestones) seen on the batting page. It was made easier by the extremely fast scoring, which meant there were few maiden overs.
Rutherford’s scoring was a close rival for Charlie Macartney’s 345 in 1921, and especially remarkable after a regulation early phase: he did not score until his 14th ball and took 64 balls to reach 50. Rutherford scored 199 runs between lunch and tea, and at one stage scored 100 runs (213 to 313) in the space of nine overs – in just under half an hour. (Compton scored more than 100 in nine overs at Benoni in 1948-49, but they were eight-ball overs.)
These latter stages of Rutherford’s innings rival Ted Allettson’s 189 off about 106 balls in 1911 (the last 142 off about 51 balls) – perhaps.
Kings of the No Ball
On I briefly reviewed the bowlers, with the longest careers, who had bowled very few no balls in Tests. To follow up, here is the other extreme: bowlers who bowled no balls most frequently. This sometimes has to be done from incomplete data, as oftentimes in the past no balls in scores were not ascribed individually to bowlers. The ball-by-ball database helps here.
Wherever possible, I have used the number of deliveries that were called no ball or wide, rather than the number of runs conceded, so a delivery that goes for ‘four no balls’ counts as only one against the bowler, something that has been enshrined in official scoring protocols only in the last few weeks. No balls that were scored from are also counted wherever possible; these were not added to no ball totals in conventional scorecards before 1998, and have to be derived from original scoresheets.
The bowler with the most no balls in the database is Bob Willis with 762. However, Wasim Akram certainly bowled more: I have 745 in the database, covering 82% of his bowling career, and so have used an estimate of 838 no balls in total.
Bowlers with highest rates of no balls (Tests)
*Estimated from incomplete data. Qualification 100 wickets.
The bowler with most recorded wides is Jacques Kallis with 104, followed by Mitchell Johnson with 90 and Ian Botham on 77. Currently, Morne Morkel is on 75 with probably more to come.
Most consecutive dot balls in ODIs:
in the available data for a single bowler is 39, across two games by Ray
Price of Zimbabwe in 2008 (against Ireland and Kenya).
Jimmy Anderson bowled 31 in a row, including five consecutive maidens, in one game against Australia at Adelaide in 2003.
These sequences did not contain byes or leg byes.
only have data for about two-thirds of ODIs, and almost none before 1985.
About half of the cases of bowlers bowling six or more maidens occurred
before 1980; almost all of them were in 55- or 60-over games. I do have data
for Phil Simmons 10-8-3-4 in 1992, but his longest sequence was shorter than
In a Colombo Test in 1985-86, Roy Dias (4) was out caught second ball, after being dropped first ball by Viswanath at slip, with the ball going to the boundary.
To supplement the Test Match Database Online, I have posted a small number of notable first-class innings that I have been able to re-score.
- Trumper 293 in three hours in 1914 (Cricket NSW Library)
- Woolley 305 in 3.5 hours in 1912. (National Library of Australia collection)
- Macartney 345 in under four hours in 1921. (Cricket NSW Library)
- Bradman 452* in 1929-30. (Cricket NSW Library)
- Compton 300* in three hours in 1948-49. (M.C.C. Library, Lord’s)
- Lara 501* in 1994. (Warwickshire C.C.C. Archive)
The sources are original scorebooks found in various locations, listed above. They have been re-scored into linear form, along the same lines as the Test matches. As is sometimes the case with old scores, the early ones have occasional anomalies, with slight corrections required. The Trumper re-score produces a score of 294 not 293, while Macartney comes in at 343 not 345. Careful analysis has isolated an over where Macartney’s missing 2 runs were probably scored, and that has been added.
Comparing Bowling Performances: a Simple but Effective Way?
I have found an interesting and simple way to compare great bowling performances, which I would suggest has some validity. (The idea was suggested by Michael Jones.) It looks at the value of the batsmen dismissed. Simply add up the career batting averages of the batsmen dismissed by a bowler. A performance where a bowler dismisses five top-order batsman will be rated much higher than one who dismisses five tailenders, and this method allows for strength of opposition almost automatically. Here are the results for innings bowling.
Figures are sum totals of batting averages of batsmen dismissed in the innings.
Although it wasn’t in a Test match, Lillee's 8 for 29 against the World XI in 1971/72 beats this lot, but only just, with a total of 346.2.
I was surprised
not to see a bigger presence of bowlers who had dismissed Bradman, but Bowes
in 1934 is the only one. It was surprisingly uncommon for a bowler to dismiss
Bradman along with the other top batsmen in the same innings.
Highest Total Batting Value, Allowing for Runs Conceded (Test Innings Bowling)
A few of these performances are provisional, since the final career batting averages of the batsmen are yet to be resolved. Steve O’Keefe makes it onto the list with a ‘mere’ 6 for 35, but it was a remarkable return for very few runs. All six of his victims had batting averages over 32.
Well, there are limits to the usefulness of this. Judging those with very short careers by their batting average is a bit dubious. It’s an interesting exercise, nevertheless. One thing that I like about it is that it does not require a battery of ‘adjustment factors’. I don’t much care for ratings that bring in too many factors. Ultimately, the choice and weighting of the factors can be more important than the raw performances.
I will do an analysis of match figures at some stage. Since both of Laker’s innings returns in 1956 are found in these lists, I know what would be #1.
Over a whole career, the value of batsmen dismissed tends to even out, but pace bowlers do tend to dismiss higher-value batsmen, on average.
for Australia v Pakistan in an ODI at the MCG on 16 Jan 1997, Anthony Stuart
(5-26) took five wickets for 2 runs in the space of 14 balls, including a
hat-trick and all of them top- or middle-order batsmen. It was his last
appearance for Australia; he was not selected again.
(and rare) anomaly in the score of the 1st Test of the 1982-83 India
tour of West Indies at Kingston. In India’s first innings, Venkataraghavan is listed as “b Roberts 0”.
However, multiple independent reports, from the West Indies and India, say
that Venkat was out hit wicket. Both the West Indies Cricket Annual and Jamaica
Daily Gleaner say that Venkat’s helmet came off
and fell onto the stumps.
Against Sri Lanka At Wellington in 1982-83, New Zealand wicketkeeper Warren Lees took five catches in the space of 70 minutes on the fourth day.
last Test series in 1983-84, champion slipper Greg Chappell was keen to
overtake Colin Cowdrey’s record for most catches in
the field. Going into the third Test, he needed only one to tie; however the
third and fourth Tests passed without any catches for Chappell, and in fact
no catches at all were taken by Australia in the
slips in the fourth Test.
I have reached a milestone in uploading the Test Match Online Database, with the uploading of the 1913-14 England tour of South Africa. This completes the pre-WWI Test matches and means that all Tests from 1877 to 1960 have been completed with as much data as I can easily muster.
Some upgrading of the 1920-1960 material will take place. For instance, I now have info on the locations of catches for about 97 per cent of catches. I am working on carrying this data through further and have reached 1984 (the first 1000 Tests). There is a gap to 1999 and from then on I have quite good data. I hope to continue the post-1960 uploading before long.
In the meantime I might polish up and upload a handful of first-class matches that I have re-scored ball-by-ball. The list is short but interesting and will include the following, if I can find them among my computer files –
- Trumper 293 in three hours in 1914
- Woolley 305 in 3.5 hours in 1912.
- Macartney 345 in two sessions in 1921.
- Bradman 452*.
- Compton 300* in three hours in 1948-49.
- Lara 501*.
A question on Ask Steven got me thinking about the effect of no balls and wides on bowling averages. Historically, the counting of no balls and wides against bowling analyses has varied, from not at all before 1983, to a complete counting since 1998, even when runs are scored of the no ball. This has had an effect on bowlers’ averages; not a great effect, but for some bowlers there is a considerable change when it comes to rankings. This can be seen in the following table, which shows the best Test career bowling averages in the last 100 years. The shows the bowlers ‘official’ bowling average alongside the averages those bowlers would have obtained if their performances were counted according to pre-1983 protocols.
The biggest differences are recorded by relatively modern bowlers who bowled a lot of no balls, such as Wasim Akram and Sean Pollock. By deleting these runs conceded from no balls and wides, their averages improve. Wasim Akram, who bowled more no balls than any other bowler, actually gains 12 places on the all-time list.
Best Test bowling averages of the last 100 years, with no balls and wides not counted
Minimum 100 Test wickets.
This is not necessarily a ‘fair’ adjustment. No balls and wides are the bowler’s fault of course, and should be counted against them. But this table does give a more level historical comparison. I find it interesting how tightly bunched the averages are, more so than the averages of batsmen.
Incidentally, I would expect that doing the operation in reverse, that is, counting pre-1983 bowlers by post-1998 counting, would have less effect on the original rankings, because the earlier bowlers in this list did not bowl a lot of no balls. Some benefited from the back-foot no ball rule prior to the late-60s.
It is a curious thing that, prior to the adjustment, Johnny Wardle, the English spinner, has the best average in the past 100 years. This is not something that many people would guess. Wardle was the ‘junior’ spinner to Jim Laker and in the 1950s was in and out of the England team, which also had periods of strongly favouring pace bowling. Wardle never played more than seven Tests in a row, played only 28 Tests in all, and barely qualifies with his 102 wickets. He tended to be selected when conditions favoured spin bowling, and that may be why his average is so exceptional.
Most ‘Total’ Test dismissal credits. Wickets + catches + run out credits.
Caught & bowled count as only one each.
in 1977, Geoff Cope, on Test debut, came as close to a hat-trick as it is
possible to come without actually getting one. After Abdul Qadir and Sarfraz Nawaz were
out, Iqbal Qasim edged Cope’s hat-trick ball to Brearley
at slip, and the umpire upheld the appeal. Wild celebrations had already
begun and Iqbal was heading for the pavilion when Brearley
indicated that he thought the ball not carried to him, and recalled the
Kapil Dev, 94 not out overnight at Delhi in 1978-79, reached his first Test century by hitting the first two balls of the day from Norbert Phillip for four and six. His century came off 101 balls.
The third day of the Karachi Test in 1980-81 was delayed 20 minutes because umpire Shakoor Rana had left his bag in the car that dropped him off at the ground.
At Sabina Park in 1981, Michael Holding had three batsmen caught and two catches dropped, all in the space of two overs.
At Banglaore in 1974-75, Alvin Kallicharran (124) was involved in ten partnerships in one innings even though he did not open. He came to the crease at 38/0 after Roy Fredericks retired hurt, and was last out. (Fredericks returned at 264/8 and was ninth out.)
Here’s some data on the bowlers who bowled the fewest no balls and wides in Tests. Naturally, data on this subject is not complete, but with about 90% available, we can make some comparisons. About 80% is from scoresheets, the other 10% from published data.
Wherever possible, I have used the number of deliveries that were called no ball or wide, rather than the number of runs conceded, so a delivery that goes for four wides counts as only one. No balls that were scored from are also counted wherever possible. The data is something of a hybrid, in that even when no ball numbers are published, no balls scored from are not always. You need a full scoresheet for that; but make of this what you will.
There have been reports floating around the internet that bowlers like Kapil Dev and Michael Holding never bowled a no ball. These are nonsense. They bowled scores of no balls and wides, hundreds in the case of Kapil.
Zero Recorded No Balls and Wides in Test Career
While we can be pretty certain that Tayfield never bowled a no ball or wide, Mankad and Venkat are no so clear. Unfortunately, Indian sources for this sort of data are often weak.
Grimmett was already well known as a bowler who almost never strayed. In 1989, Greg McKie analysed Grimmett’s first-class career and found that he did not bowl any no balls at all among 50,000 deliveries. McKie found just five wides.
Wides but Zero No Balls in Test Career
Swann did record one no ball in ODIs, but it was called for having an illegal leg side field! Another bowler reputed to have never bowled a no ball was Lance Gibbs, but there is one in a scorebook from the 1965 series against Australia, and another reported from a Test in Pakistan in 1974-75. Tate, in addition to bowling no no balls, was never hit for six in a Test match, the longest such career.
No Balls but Zero Wides in Test Career
Certainly an extraordinary contrast in no balls and wides from Underwood.
Maiden Over in the Final Over of an ODI (50 overs)
I don't have all matches, but I know of eight occasions. Two are in the first innings
AA Donald SAf v Ban, Benoni 6-Oct-2002
CJ Anderson NZ v SAf, Mt Maunganui 24-Oct-2014
The others were in the second innings
C Pringle Aus v NZ, Hobart 18-Dec-1990
SK Warne Aus v Pak, Colombo 7-Sep-1994
AJ Hall SAf v SL, Adelaide Oval 24-Jan-2006
RN ten Doeschate Ber v Ned, Potchefstroom 8-Apr-2009
N Deonarine WI v Zim, Grenada 22-Feb-2013
GJ Maxwell Pak v Aus, Abu Dhabi 12-Oct-2014
In the overs by Warne, ten Doeschate, and Deonarine, the batting team had no chance of winning.
In the Pringle and Maxwell overs, the batting team needed only 2 runs to win and 1 to tie. Bruce Reid faced the Pringle over, and was unable to put bat to ball. In the Hall over, 11 runs were required.
Jerome Taylor's first first-class century was 106 in a Test match. His previous highest score was 40, and his first-class average prior to the century was 12.5. To date he has played over 50 non-Test first-class matches, and has still not made a score over 50. His first-class average outside Tests has dropped to 10.5.
Bruce Taylor of New Zealand scored a century in his first Test innings. It was only his fifth innings in first-class cricket and his previous best was 49.
Ian Healy scored four first-class centuries, all of them in Tests. He is the only player with more than 2, if I have calculated correctly.
At Auckland in 1977, Greg Chappell’s innings of 58 was interrupted when he was menaced by a streaker. Chappell attempted, apparently successfully, to hit the intruder on the backside with his bat. The incident appeared to distract Chappell, as he was run out off the very next ball.
Another snippet on early use of the reverse sweep: during that innings Chappell was described as playing a “back-handed” sweep.
Dropped Catches Report 2016-17
I have surveyed more ball-by-ball texts for dropped catches and missed stumpings, covering the period Feb 2016 to April 2017. This extends the missed chance data to about 700 Tests going back to year 2000.
There was increase in the percentage of catches dropped in the recent data. The 2016-17 figure was 26.5% chances missed, about two percentage points higher than the average over the previous four years. However, the miss rate was very similar to other years, including 2003 and the four years from 2008 to 2011.
It is not entirely clear why the rate was higher this year. One factor is an increased number of Tests involving Zimbabwe (30% misses) and Bangladesh (32.2% misses) which drags the overall percentage upward. If it is apparent from their match results that Sri Lanka is in decline, this is also borne out in the catch stats; SL’s stats have increased from 25% misses in 2015 to 30% in 2016-17.
Unlike the all-country data, the combined average for Australia, New Zealand, South Africa and England is steady at 23%. South Africa recorded only 18% misses, one of the lowest single-year figures for any country since surveys began in 2000.
The recent figures reverse a slight historical improvement trend seen over the previous decade.
individual countries in 2016-17 are
On the individual front, Alistair Cook has overtaken MS Dhoni to claim the most career misses of any modern player. He has 70 misses to Dhoni’s 66; the latter is probably a career final, while Cook’s figures do not include the current England season. Rahul Dravid is one recent player who may have dropped more catches than Cook; I only have data for about 70% of his career; from that I would estimate about 75 misses in total.
Most expensive miss of the year occurred when Azhar Ali was 17 against West Indies at Dubai; he was missed by Leon Johnson in the gully, and went on to 302 not out. KK Nair, who against England made one run more, was missed on 34, 217 and 246. Steve Smith was missed four times in his 109 against Sri Lanka; the innings, under difficult conditions, was nevertheless rightly acclaimed.
Kane Williamson had an exceptional year in the field, taking 14 catches and dropping only one. Johnny Bairstow recorded the most misses, 15.
At the time of
writing, Jimmy Anderson has probably become the first bowler in this century
to have 100 catches dropped. I have 97 for him, but that does not include the
current England season. Harbhajan Singh had more
than 100, but some of those, an unknown number, occurred before 2000. Stuart
Broad is not far behind Anderson. It will take further analysis.
I have posted pdf
versions of a couple of articles by me that have appeared online in Cricket
2) “End-over” jitters, that is the effect of having to bat in the final overs of the day (very little as it turns out). March 2017.
SOME SNIPPETS FROM THE 1970s…
In the final over of
the drawn third Test of 1974-75, bowled by Tony Greig
with Australia needing 14 runs to win, umpire Bailhache
(officiating at point) called no ball on the grounds of three fielders behind
square leg. According to reporters, three previous balls in the over had also
been bowled with this (illegal) configuration, but not noticed by the
umpires. There were six runs and one wicket off the over, and the match was
drawn with Australia finishing on 238 for 8, eight runs short of victory. The
penultimate over, bowled by Derek Underwood, had been an eight-ball maiden.
Test at Auckland in 1975, England’s Keith Fletcher took five ‘brilliant’
catches at slip according to The Times, but the Manchester Guardian reported that he
dropped another five.
In the Christchurch Test that followed, Barry Hadlee (12th man) fielded as a substitute, joining his brothers Dale and Richard on the field, while their father Walter watched from the stands. Barry never played in New Zealand’s first XI in A Test, although he played two ODIs for New Zealand alongside his brothers.
On the first day of the Kanpur Test of 1972-73, two thousand police were assigned to the ground. There were 30,000 spectators. There was supposed to be play on the following day, but it was a public holiday, and the police force could not find enough available officers, so it was declared a rest day.
In the Adelaide Test in 72-73, Talat Ali, on debut, suffered a broken hand in Pakistan’s first innings. He was the first batsman since Charles Bannerman to be listed as retired hurt in his first Test innings. He was not expected to bat again, but came out to bat in the second innings, late in the day, with Pakistan 214 for 9 and on the verge of an innings defeat. Holding the bat with one hand, he faced nine balls and forced play into the fifth day, but only thanks to a dropped catch by Ian Redpath on the very last ball of the fourth day.
What a finish that would have been if it had been a four-day Test!
The hoped-for rain did not arrive, and Talat was out in the second over next morning: 0 off 16 balls. A fourth-day photo shows him batting right-handed, but a report from the fifth day says he switched to left-handed. If so, he was the first known batsman to bat both ways in a Test innings, Salim Malik being the only other known case (see entry for 23 Oct 2016).
On the subject of ‘switch hitting’, one of the claimed inventors of the reverse sweep was Mushtaq Mohammad. Here’s some proof that he used the stroke in Tests – a quote from the Otago Daily Times during the Dunedin Test of 1972-73.
“After tea, both batsmen produced as fine a hitting display as seen on Carisbrook in memory. Off the first four overs, 46 runs came from spinners Pollard and O’Sullivan with Mushtaq showing incredible footwork by switching from right-hand to left-hand during the flight of the ball to sweep boundaries.”
Mushtaq made 201 off 407
balls and his partnership with Asif Iqbal (175 off 299) produced 350 runs off
about 575 balls.
The later stages of the partnership included 100 runs in the space of 37 minutes.
Ashru Mitra has done a nice piece of research on Test umpires who, on their Test debuts, gave a batsman out, or saw a wicket fall, on their very first ball (at the bowler’s end). I had heard of Bill Alley and maybe one other, but I was a little surprised at the number of names that came up. Although, technically, all dismissals require a decision by an umpire, not all of these necessarily required an active decision.
Ashru has kindly allowed me to publish his findings here.
Debutant umpires giving dismissal verdict
on the very first ball of the Test: A preparatory list
Umpire Debut: Thomas Burgess and Richard Torrance
Both umpires were making their Test debuts. Which ends they took is not known but wickets fell on the first balls of both the first and second overs.
Batsman dismissed: Herbert Sutcliffe cwk James b Badcock (UPDATE: Umpire Torrance)
Batsman dismissed: E Paynter b HD Smith (UPDATE: Umpire Burgess)
England at Brisbane, 1936-37
Batsman dismissed: Stanley Worthington cwk Oldfield b McCormick
India v West Indies at Calcutta, 1974-75
West Indies v Bangladesh at Gros Islet, 2004
First ball of the second over of the match:
Australia vs England at Melbourne (5th), 1958-59
Umpire Debut: L Townsend (Confirmed: His decision)
Batsman dismissed: TE Bailey c Davidson b Lindwall
The following umpires were making their debuts when a wicket fell first ball, but the decision either fell to the other umpire or is not known:
Australia vs England at Sydney, 1903-04
Umpire Debut: AC Jones (Confirmed: Not his decision)
Batsman dismissed: VT Trumper c Foster b Arnold
South Africa vs England at Cape Town, 1922-23
Umpire Debut: GJ Thompson (No information
West Indies v.
Pakistan at Port of Spain, 1957-58
New Zealand v Australia at Auckland,
Pakistan vs West Indies at Karachi, 1990-91
Umpire Debut: Riazuddin (Confirmed: NOT his decision)
Batsman dismissed: CG Greenidge
lbw b Waqar Younis
South Africa v
India at Durban, 1992-93
Sri Lanka v West Indies at Pallekele, 2010-11
Umpire Debut: Bruce Oxenford (Confirmed: Not his decision)
Batsman dismissed: Chris Gayle lbw b Lakmal
In Tests, the practice of having another bowler complete an over, when a bowler was injured mid-over, seems to have started in 1981. At Kingston in that year, Graham Dilley was unable to complete an over, and Robin Jackman bowled the last two balls.
I can’t find any earlier cases in ODIs. Perhaps the need never came up.
A questioner on Ask Steven asked if there were completed first-class matches where no one made 50 and no one took five in an innings. I looked at the last few years and found (only) one: Badureliya Sports Club v Bloomfield Cricket and Athletic Club in 2014. The highest score was 45 and the best bowling was 4/17. It was an interesting scorecard: Bloomfield led by 2 runs on first innings and also won by 2 runs when both teams made 175 in the second innings.
Test of 1905 featured a strange set of high-scoring strokes. Joe Darling hit
a six, but that was thanks to overthrows. He also cleared the boundary with
another stroke, but that counted only five. Kelly and Hill hit similar
strokes. Two batsmen, Hayward and Spooner, hit all-run fives without
overthrows (the only Test where this happened twice), while Rhodes hit a five
By this time, Australian (and South African) authorities were being more sensible and awarded six runs to all hits clearing the boundary. This did not come into universal use in England until 1912.
‘Century in a session’ is a familiar record category. How about most runs in two consecutive sessions? Not so familiar, but here is a list. I did this calculation after Shikha Dhawan’s tour de force in Galle.
Most Runs in Two Consecutive Sessions (Test Matches)
Where a player qualifies twice for the same innings, the higher value only is listed. Dhawan’s 190 is the highest innings to be contained entirely within two sessions. His 126 between lunch and tea is a record for the first day, and the third-highest between lunch and tea on any day, after the 173 by Compton and 150 (or 151 or 152) by Hammond.
Of course, the old-timers had the advantage of higher over rates, and so comparisons must be considered with that in mind, but they did also play with much inferior bats and on larger grounds.
Five wickets in fewest balls in an ODI, where known
11 balls J Garner, WI v Eng, Lord’s 23-Jun-1979 (World Cup Final)
11 ACI Lock, Zim v NZ, Napier 3-Feb-1996
11 Shoaib Akhtar, Pak v NZ, Auckland 18-Feb-2001
12 Mohammad Sami, Pak v NZ, Lahore 1-Dec-2003
12 M Morkel, SAf v NZ, Napier 29-Feb-2012
Zahoor Khan took six wickets in 15 balls, Dubai 2-Mar-2017.
KAJ Roach took 5 wickets in 9 balls, over 2 games in 2011.
B Lee took 8 wickets in 27 balls, over 2 games in 2003.
I wrote an article on the statistics of DRS a while back, which was published online by Cricket Monthly. I have posted it in my longer articles section here.
A Note on Score Reconstruction
I have been including, in the database for pre-1915 Tests, ball-by-ball reconstructions for certain innings and/or matches, made in the absence of complete scorebooks. I just wanted to make some points clear about this process, especially as the Database approaches the 1902 Tests played by Australia in England and South Africa.
It is a great misfortune that no scorebooks are known to exist for these matches. However, greater resources are becoming available in terms of match reports in newspapers, to the extent that it is possible to construct over-by-over (sometimes ball-by-ball) versions of some innings, particularly those that involve rapid scoring or frequent falls of wickets. The British Newspaper Archive now boasts dozens of titles, available in full and online, for the year 1902. In addition to this I have accessed, from libraries, copies of other newspapers that are not in this Archive. These reports vary in detail, but when distilled together, and taken with information from other sources (such as a partial score available for the Old Trafford Test), it makes possible a ‘best rendering’ of important innings that in turn allow estimates of balls faced and other important statistics.
This has also been done for the series in South Africa. Although sources are fewer, the South African papers of the time often used a strictly narrative style (old-fashioned at the time) of reporting that mentions almost all scoring shots in sequence. Australian papers prior to 1894 often used the same style; after that, a more interpretive style of reporting came into vogue that makes it much harder to reconstruct innings statistically.
1) Even with combined sources, gaps occur that must be filled using educated interpolation,
2) The sources sometimes conflict.
3) It is not always possible to come up with a sequence of overs that is perfectly consistent with every source.
Generally, however, the broad structure of innings are clear (who was bowling when, and in which overs wickets fell), and many of the scoring details can be accurately placed. It is just certain passages of play that must be filled in. Periods of slow play with occasional singles and maidens are particularly difficult; by the same token, they are often not important. Some other detail from the sources can assist, such as the reporting of the number of ones, twos, threes and fours for major innings, and this can be found in certain sources.
I hope that the reconstructions can be accepted in this spirit; that they are not exact, but offer a useful guide to the progress of certain important innings and matches.
There are more sources out there, unexamined: perhaps others can take up the challenge of tracking more of them down. For instance, Gerald Brodribb gives a ball-by-ball list of Gilbert Jessop’s famous 104 at the Oval, but does not name his (newspaper) source. Having looked at dozens of potential sources, I have yet to find this.
So don’t fret over potential errors: improve on it if you can!
A question on Ask Steven had me looking for bowlers who took wickets with their last two balls in Tests. The only ones that I found were Gerry Hazlitt (Australia) in 1912 and Godfrey Lawrence (South Africa) in 1962. Current players were not considered.
Hazlitt, who only took 23 wickets in Tests, took 5 for 1 off his last 17 balls at the Oval in 1912.
Lawrence took just 28 wickets in Tests. His last two were the 8th and 9th of New Zealand’s second innings at Port Elizabeth in 1961/62. Peter Pollock then took a wicket with the first ball of the next over to finish the innings and complete a rare 'team hat-trick': three wickets in three balls by two different bowlers.
That said, I thought I would re-visit these ‘team hat-tricks’, which I reported on in 2012. The only ones I know of are listed below, followed by a few other cases where three wickets fell in the space of three balls, but no hat-trick occurred. They are rare indeed and getting rarer: it appears that it has never happened in Australia, or India. The most extraordinary thing is that Godfrey Lawrence was involved in two, and in consecutive Tests. This is one of the strangest coincidences that I have seen in cricket statistics.
Three wickets in three balls, by two different bowlers
I have been successfully convalescing after major surgery, and while doing so I have been able to keep up with updates to my Online Database. The data update is now complete for Tests in the 19th Century. Once I finish 1900 to 1914 the database will be complete up to 1960. The next step would be to refine and update some of the Test data from 1920 to 1939.
Slowest 50s in ODIs (balls faced)
50-over matches only
UPDATE: 132 balls KR Stackpole (61 off 145) Australia v England, Edgbaston 28-Aug-1972
Most ‘Dismissals’ by no balls since 2000 – an updated list (bowlers).
The most runs scored after a no ball 'dismissal', recorded since 2001, is 279 by Kumar Sangakkara (287) in Colombo in 2006. He was bowled by a Dale Steyn no ball on 8. A catch was dropped in the same over. The next wicket fell 603 runs later.
Chris Gayle (333) was caught off a no ball when on 287, at Galle in 2010.
Earlier this year at Galle, Kusal Mendis was caught first ball off a no ball, and went on to make 194.
At Wellington in 2016, Adam Voges, on 7, was bowled by a no ball from Doug Bracewell. Replays showed that the umpire erred, and that the ball was legal. Voges went on to make 239.
The leading beneficiaries among batsmen are Rahul Dravid on 7, with Michael Clarke, Virender Sehwag and Alistair Cook on 5. Although he has benefited only twice, Sangakkara has added the most runs after his ‘dismissals’ with 326, followed by Kevin Pietersen on 323 (three innings) and Michael Clarke on 295.
I recently came across a note, in a linear score, of Len Hutton being caught off a no ball (by Bill O'Reilly) during his 364 at The Oval in 1938. Hutton was on 153 at the time.
This is not really the same thing as modern instances, because under the back-foot rule, the batsman had time to change his shot after the no ball call. Hutton was, in effect, taking a 'free hit' which was caught in the deep. He scored one run.
I have embarked on an effort to expand the Davis Test Match Database Online, to cover Tests from the beginning in 1877, up to 1914 (1920-1960 is already online).
A covering page for the early Test series is here. The match scoresheets have a new feature, in that the fielding locations of nearly all catches are specified, using ‘shorthand’ notation of “sl” (slip), “wk” (wicketkeeper), “mn” (mid-on) etc. Each series cover page has a full key to the notation used.
Ball-by-ball records of Tests have been posted, where possible. In the case of most very early Tests, these are not derived from scorebooks, but are reconstructions based on the very detailed reports in some newspapers of the time, which often mention the events of every over if not quite every ball. As such, there are inevitably some approximations and anomalies, and readers should bear in mind that complete precision is not possible in such cases. Balls faced data from these reconstructions should be regarded as indicative only. I hope that they can be regarded as useful, nevertheless.
In some Tests of the 1890s and later 1880s, I have been able to reconstruct only parts of the matches, but I will post these fragments anyway, since they usually relate to interesting parts of Tests.
I will endeavour to update the Database when I can; however, impending health problems may reduce the rate of updating of the Database, and the blog, for the foreseeable future.
For something completely different, I have started writing reminiscences from some of my travels over the years. The stories are spiced up with photos from those travels. I have posted a story, of a journey from China to Pakistan along the old Silk Road and the Karakoram Highway in 1989, in two parts. Part 1 is here and Part 2 is here. An account of the joys of ‘Hard Seat Class’ on Chinese trains is here.
The first ball faced by Bob Willis in Test cricket was a hat-trick ball (Perth 1971, bowled by John Gleeson). I was surprised to find that this is not particularly uncommon, with a few dozen cases.
The LBW Miser
It’s somewhat notorious but also true: in the six Tests of the 1970-71 Ashes series, not a single Australian batsman was given out leg before wicket. The umpires involved were Lou Rowan and Tom Brooks (five Tests each) and Max O’Connell (two Tests).
These men did see their way clear to give five Englishmen out lbw (itself a very small number) in that series. I have been able to determine which umpires gave the decisions. Three of them were given by Rowan, one by O’Connell, and just the one (in five Tests!) by Brooks. The rather unfortunate batsman was John Hampshire, given out at Adelaide, just before a declaration when England was chasing quick runs.
I do recall my father talking about Tom Brooks. Dad was a first grade umpire in Sydney the 1970s when Brooks was the dominant figure in umpiring there. Brooks, Dad told me, was adamant that the conditions for lbw were extremely hard to satisfy, and that lbw decisions should be rare. I don’t think Dad actually agreed with this, and modern-day DRS data shows conclusively that Brooks’ opinion was incorrect.
Rowan was a police Sergeant. He was, I am told, also a man of unswerving opinions, with great confidence in his own judgement. I suppose that helps if you want to be a top-flight umpire.
With only one lbw decision against them, the Englishmen could hardly accuse Brooks of bias, I suppose. They were troubled, though, by the umpires intervening to apply restrictions on intimidatory bowling, in a way that they saw as being unfair to star fast bowler John Snow.
I happened to be at the SCG on the day that Snow hit Terry Jenner on the head with a bouncer; it is just about my earliest ‘live’ memory of a major event on the cricket field. Jenner retired hurt, although he was able to bat again later. Rowan stepped in and warned Snow about bowling bouncers, arguably unfairly under the guidelines of the time. It was arguable enough for England captain Ray Illingworth to get into a shouting match with Rowan. Shortly afterward, Snow was accosted by a drunken spectator, and England stormed off in protest. With the Ashes at stake, cooler heads prevailed and the match continued. England won the match and the Ashes.
It was a rugged series. Here is a photo of Graham McKenzie’s last ball in Test cricket. Hit on the face by a good length ball from Snow and retired hurt. That was “Blow Number One” as the photo says. Blow Number Two was delivered by the selectors, who obligingly told McKenzie, that very evening while he was convalescing, that he was dropped from the team.
I don’t know if it was a fashion at the time, but there were also only five batsmen given out lbw in the six Tests of the subsequent 1971 season in England. Only two of the five were English. Something seems to have happened between then and the 1972 Ashes series, in which 27 batsmen were given out lbw in five Tests.
Did Brooks relax his views? In the 1974/75 Ashes, he stood in all six Tests. He gave nine batsmen out lbw, six of them Australian, including two ducks for Wally Edwards, who would later become Cricket Australia chairman. The other umpire, Robin Bailhache, gave six lbws, two of them Australian batsmen.
I have notes on 46 instances in Tests of a batsman being run out ‘accidentally’ at the non-strikers end, when a shot ricocheted off the bowler. No batsman has been out this way twice. However, there are two batsmen who, as striker, twice saw their partners run out this way, VVS Laxman and SV Carlisle. In Laxman's case, Harbhajan Singh was run out at Mohali in 2003, and Dravid was run out at Kanpur in 2009. In the Dravid incident, the ricochet was a dropped catch by Herath off Laxman.
In Carlisle's case, the batsmen run out were Ebrahim at Bulawayo in 2001, and Taylor at Harare in 2005.
It is possible, though unlikely, that there are others.
Hugh Tayfield took a catch off the second ball of his debut Test, and Ian Chappell took one off the third ball of his.
A substitute named Chris Sabburg took a catch off his second ball on the field in a Test a few years ago. Sabburg has not yet played first-class cricket, although he has appeared in the BBL.
been at least one umpire who was required to give a batsman out caught behind
from his first ball in Test cricket (HP Sharma in India in 1974-75, I think).
This was probably exceeded by Hanif in 1958, but there is no data available. Alec Bannerman failed to score off 568 out of 620 balls faced at the SCG in 1891/92. He scored only 91 runs.
The list of non-Test players who scored most first-class runs is dominated by England county cricketers (not all of them English), led by Alan Jones. So I wondered who hit the most f-c runs among those who never played in England. The database came up with Sajid Ali who hit 15,368 f-c runs. Ali played no Tests; however, he did play ODIs for Pakistan. The player with most f-c runs without ever appearing for his country or playing in England appears to be Amol Muzumdar with 11,167 runs in 171 matches.
In the Ranchi Test against Australia, Cheteshwar Pujara became the first Indian batsman to officially face more than 500 balls in a Test innings (202 off 525 balls, batting for 162 overs). As a first, this is actually quite odd, since more than 60 batsmen from other countries have played innings longer than 500 balls.
However, it is likely that Pujara has at least one Indian predecessor. At Port of Spain in 1952, Midhav Apte batted for 200 overs for 163. With half the strike, that would come to 600 balls.
I don’t have a balls faced figure for Apte, but the odds of someone batting for 200 overs but facing less than 500 (or 525) balls are extremely small. The standard deviation for balls faced for an individual batsman over a span of 200 overs is about 25 balls. There is a greater than 99% chance that Apte faced more balls than Pujara.
There is a tenacious myth in Australian cricket: the ‘87 hoodoo’, that holds that a score of 87 is somehow unlucky. Tenacious yet tenuous. The myth has long been debunked – in fact it is the safest score in the 80s for Australian batsmen – but the story still crops up regularly during idle moments in Test matches.
The origins of the myth are somewhat obscure. The most accepted line is that Keith Miller originated it; the story is here. Miller says he formed the idea on seeing Bradman dismissed for 87 in 1929. Personally I have been a bit sceptical. Miller was prone to spinning tall tales, and the claim that he would originate such an idea at age ten sounds fanciful. Note how he glides over the difficulty of Bradman not actually being dismissed for 87 (“the scoreboard was slow”: if you say so, Keith. In fact, Bradman was 85 at lunch and went from 85 to 89 with a four, so it seems unlikely that the scoreboard would ever have read 87).
Anyway, I have come across an item that sheds light on the origins of the hoodoo. From the Sporting Globe in 1950, it is an anecdote from a club match (albeit one involving Test players) and it sets out the superstition as explained by Test spinner Ian Johnson. Johnson said that his 87 anxiety came from the previous season’s Australian tour of South Africa, where, Johnson said, wickets seemed to fall frequently on a score of 87. Curiously, the report applies the hoodoo more to team scores than individual scores.
There are some interesting aspects to the report. For one thing, the reporter hasn’t heard of the hoodoo and has to have it explained to him. This suggests that the hoodoo was a new thing in 1950. For another, Johnson is quite clear about it originating on that tour, and offers the interesting detail about the team avoiding hotel rooms numbered 87. Gideon Haigh, in The Summer Game, has something to add here. Early in the tour, reports Haigh, Bill Johnston was seriously injured in a car accident in Natal: he had been staying in Room 87 in the team’s hotel.
The next morning, on the first day of first-class cricket on the tour, star batsman Neil Harvey was out on a (team) score of 87. [I checked and found that there was nothing ‘unlucky’ about 87 on that 1949-50 tour. Just four wickets fell on 87 in the entire tour, none of them in Tests and only two involving Australian batsmen: by contrast, eight wickets fell on a score of 84, but no one seemed to be worried by that.]
The combination of Johnston’s room number and Harvey’s dismissal may have been just enough to give impetus to the superstition. Since Miller was on the tour, one could imagine him being the instigator, and perhaps reviving an earlier superstition of his. One problem with this, however, is that Miller was not in South Africa at that time: he was only selected as a result Johnston’s injury (a selection saga that is a whole other story) and it took him weeks to travel to South Africa. It certainly seems that the myth took hold without him.
However, Miller, Johnson, and Lyndsay Hassett all played for South Melbourne club (the same club mentioned in the Sporting Globe report) in the late 1930s and just after the war, before Miller moved to New South Wales. An article in The Age in 2007 mentions a South Melbourne connection to the hoodoo, as does the Arunabha Sengupta article linked to earlier. So in spite of my scepticism, signs still might point towards Miller; one hypothesis would be that the myth was just a South Melbourne club ‘thing’ (started by Miller, perhaps in the 1930s), until its elevation to Tests was sparked by the ‘87’ events in Durban (instigated by Johnson and possibly Hassett).
I have found nothing else on the hoodoo in the NLA Trove database from 1945 onwards. Others might like to have a look.
Ken Piesse tells me that Miller told him that the whole thing was “sheer bunkum”; I’m not sure if Miller was talking about the hoodoo itself, or the origin story.
In the past I have heard claims that Ken Mackay originated the belief. Mackay was known for his superstitions, but the 1950 report rules Slasher out. Mackay did write an article on cricket superstitions, published in 1964 in Jack Pollard’s Six and Out; this article discusses the 87 hoodoo in some detail, proving that the idea was widespread in Australia at that time. (Even so, newspaper accounts of Brian Booth’s 87 against South Africa at the SCG in 1964 do not mention any hoodoo.) However, Mackay said he did not know the hoodoo’s origins, and he does not mention Miller in his article.
*Ogdontaeptaphobia is a word I made up using the Greek words for eighty seven.
Postscript: Miller, Johnson, Hassett, Harvey, and Johnston were all from Victoria.
I noticed, from an Aslam Siddiqui post, that Cricinfo was missing a ball-by-ball text of an ODI, Zimbabwe v Afghanistan in 2014. As it happened, Cricbuzz covered this match and a couple of others that Cricinfo missed in 2012. Using those gives complete ball-by-ball for the last 650+ ODIs, since 2011.
However, prior to that, data is not complete. Although Cricinfo started doing ball-by-ball in 1999, more that 15% of ODIs are missing from 1999 to 2011. The great majority of these are what might be called 'minnow matches'; the gaps are bigger in the earlier years.
My collection of ODI scores covers about 50% of ODIs from 1985 to 1999, and a small number of earlier ones. So far, I have re-scored about 300 out the 550 or so obtained from 1985-99. (I do one each day, last thing before I got to bed: most can be done 20-30 minutes.)
There was a curious incident in that Zim/Afg game in 2014: an instance of '6 wides'. No other information is given. However, the batsmen changed ends, so I presume that they ran one and there were four overthrows.
There are no other cases of six wides in my database of ODIs and T20 internationals. There is just one in a Test, which involved a helmet penalty and the batsmen did not change ends.
UPDATE on the history of streaking: Sreeram has now posted an article on the subject with more information. Published on the Cricket Country website.
In an ODI at the MCG in January 1995, Darren Gough opened the bowling for England, but injured himself in the delivery stride of his first ball and did not deliver a ball in the game. Angus Fraser bowled the over and Gough is not listed as a bowler. Gough was taken to hospital with a stress fracture in his foot.
On the subject of unfortunate first overs, at Adelaide in 1969 opening bowler Charlie Griffith conceded 19 runs off his first over. It was an eight-ball over that also contained four no balls. The runs off the bat were all hit by Keith Stackpole. The over was the second of the innings, Sobers having bowled the first. The most runs off the first over of an innings, where known, is 18 hit by Bob Simpson off Wes Hall at the MCG in 1961.
In the final innings of the Bombay Test of 1969, India managed to drop (or miss) five catches off the New Zealand batsmen; in spite of this, all ten wickets fell to catches, and New Zealand, chasing only 188, lost the Test by 60 runs.
In the Galle Test between Sri Lanka and Bangladesh, Kusal Mendis was caught off a no ball, first ball, and went on to make 194. It is quite unusual for anyone to be 'out' to a no ball when on 0. I know of 14 cases since 1999, but only one batsman went on to make more than 32. That was Hasan Raja (68) at Sharjah in 2002.
Only two previous batsmen were 'out' to a no ball first ball and neither reached double figures.
Data is limited to 1999-2016.
For something completely different…
The origins of ‘streaking’ at sporting events go back to about 1974. According to Wikipedia, instances at US colleges dated back to the 1960s; it became a major fad in colleges in 1973, and began to be seen at major sporting events the following year. Wikipedia puts the first streak at a major sports event at April 1974, but Sreeram has found reports of a streaker on the field at a Test match in March of that year. It was the Auckland Test on 22 March 1974; there was a streaker on the first day, followed by another on the second day. The culprits disappeared into the crowd and are unidentified. The New Zealand Herald mentions them in only a backhanded way, commenting that the final day was “for once streakerless”.
Searching the Canberra Times for the word ‘streaker’ comes up with no real hits before March 1974, but it does mention the first incident in Auckland, in its report from the first day. The same paper appears to have no reports of streakers from the 1973-74 Australian season. The Times also lacks hits until March 1974, and even then it only mentions incidents unrelated to sporting events. The first streak at Lord’s was in 1975.
Nowadays the fad has largely disappeared at cricket grounds, thanks to intense security at sports fields, and (in Australia) massive fines for setting foot on the playing surface.
something else quite rare about that first Auckland day. Doug Walters scored
104 not out on a day when 18 wickets fell, seven of them ducks. Australia was
bowled out for 221 with New Zealand 85 for 8 at stumps. This is the most
wickets to fall on a day containing a complete individual century in the last
100 years. Apart from Walters, the batsmen batting that day averaged 10.1.
In my files, I
have come across a hand-written note by Colin Clowes
on those hat-tricks by Matthews, with some detail that might not have been
The second came in Matthews’ 7th over of the second innings with the score on 70 for 5: W, W, W, 0, 4, 4. The wickets were Taylor, Schwarz and Ward, with the runs scored by Beaumont. Kelleway, bowling at the other end, had taken a wicket in the previous over, and took another wicket (the 9th of the innings) in the next over after the hat-trick (a maiden). Matthews’ 8th and last over was 2, 3, 0, 4, 0, 0, at which point he was taken off. The hat-trick appears to have occurred in the 24th over of the innings. South Africa was out for 95, in 95 minutes, in the 29th over of the innings, Kelleway taking the final wickets. In playing time, the hat-tricks were about 85 minutes apart.
Five wickets fell
in three overs; the exact number of balls from first to last is not recorded.
I’m not sure what possessed the captain to elevate Ward in the second innings
batting order, so as to face another hat-trick ball.
Here is an article by me, just published in the Cricket Monthly, on the “end-over” jitters, that is the effect of having to bat in the final overs of the day (very little as it turns out). There is a curious comment under it by a cricket captain who basically says “I don’t care what the stats say, it feels good so I will keep doing it.” It’s rather difficult to reason with that.
At some stage I will also post the article on my website.
I have completed another stage of a survey of Test match catches, identifying the field locations of as many catches as possible. The years covered in this part of the survey are 1877 to 1970 (670 Tests), with locations identified for more than ten thousand catches. Wicketkeeper and bowlers, of course, are easy, but not so the others. Nevertheless, locations have been found for about 96% of catches in this period. There are only two series for which I have almost no data: MCC in West Indies in 1929-30 and New Zealand in Pakistan in 1969-70.
One complication has been the evolving names for fielding locations, and it is difficult to be certain about some old terminology. For example, ‘cover slip’. According “The Language of Cricket” (Eddowes), this is an old name for third man, but some old reports mention fielders in both positions in describing field settings, suggesting that they were different things.
The term ‘midwicket’ was not encountered until 1931, and did not become used widely for a few years after that. It seems that previously ‘short leg’ was used instead, with other terms for what we would call short leg now. The nomenclature going around the legside field was mid-on, then short leg (midwicket), then square leg, and then long leg. ‘Fine leg’ was mostly a later term.
The term ‘cover’ seems to have originated as a covering or backup fielder for the point fielder. Before 1945, ‘cover point’ was the almost universal term. ‘Cover’ or ‘the covers’ as standalone terms came later.
During the rest day of the Bridgetown Test of 1977, Pakistan players Zaheer Abbas and Wasim Bari were rescued from drowning, by life guard Aldolphus Griffith, while attempting to swim back to their hotel from a raft that had drifted out to sea. Zaheer was not actually playing in the Test, but Bari was. The following day, Bari, batting at #11, scored a match-saving 60 not out, adding 133 for the last wicket with Wasim Raja.
In the second Test of 1953-54 (South Africa v New Zealand), John Reid broke his bat playing a shot and was out caught at short fine leg.
Abdul Azeem, who played first-class cricket in India in the 1980s, had a complete career of 114 innings but made no ducks at all. He did, however, make a couple of ducks in List A cricket.
At Bombay in 1956, Neil Harvey should have been caught on 99, but the fielder failed to move to the ball. Harvey then took another half an hour to reach his century. The number of balls he faced is not recorded.
A piece of umpiring trivia. The first time that Australian umpires were permitted to take the field without coats was the Adelaide Test of 1967-68 (v India). Reason: extremely hot weather. They still had to wear ties.
The Odd Fields of the Early Days
Here is a little bit of data that suggests that the game was played rather differently in the very early days. In the reports in The Times for Tests in 1888 and 1893, there is a listing of the field placings deployed at the beginnings of some innings. There are lists for 14 separate innings/bowlers; all of them apply to the first one or two overs of an innings. While hardly exhaustive, there is enough data to tabulate to give a taste of how fields were set in those days.
Use of field placings: 14 examples 1888-1893. Field place names have been converted into modern parlance where I can be confident of the translation.
The number 14 generally means that the position was used in every case. There are 15 ‘cover’ fieldsmen (always referred to as ‘cover-point’ in those days) because there two cover fielders listed in one innings, and one in all the others.
The most striking thing, though, is the massive concentration of fielders from mid-off to mid-on. Some innings featured two mid-offs, a long-off, a mid-on and a long-on. (In a small number of cases, positions were referred to as ‘short’ mid-off or mid-on). If I have interpreted the accounts correctly, fielders between what we call midwicket to fine leg were, by contrast, extremely sparse. Remember that these are opening bowlers in their first over.
I have illustrated two fields given for the opening overs of Old Trafford 1893. There is a remarkable contrast between field settings for the opening bowlers Arthur Mold and Johnny Briggs.
The Mold field is the only one of the 14 that resembles a modern field setting for an opening bowler; even so, it would be considered somewhat defensive for a modern Test match. It is the only field with a third slip (called ‘cover slip’, while second slip was ‘extra slip’). As for the Briggs field, I can’t say I have personally seen anything quite like it. Long-off, straight hit, and long-on for the opening over, plus mid-off and mid-on? That is what it says in The Times.
There was method in those field settings that suggests that they were set because the range of shots of batsmen was more restricted. I am still preparing data on this, but a lot of batsmen were out caught between mid-off/mid-on in those days.
Note some caveats: the exact positions, or ranges, of ‘point’ and ‘third man’, as used on those days, are possibly open to interpretation. ‘Midwicket’ did not exist as a named position: ‘short leg’ is sometimes used in reports instead, but does not mean a close-in fielder. ‘Gully’ did not exist as a named position; there is a possibility that some point fielders fielded there.
The following players had a six as the first scoring shot of their careers in Tests
EW Freeman Aus v Ind, Brisbane ('Gabba') 1967/68
CA Best WI v Eng, Kingston, Jamaica 1986
KM Dabengwa Zim v NZ, Bulawayo (Queen's) 2005
DM Richards WI v Ban, Arnos Vale, St. Vincent 2009
Jahurul Islam Ban v Eng, Mirpur 2009/10
Shafiul Islam Ban v Ind, Chittagong 2009/10
Al-Amin Hossain Ban v SL, Dhaka (Mirpur) 2013/14
MD Craig WI v NZ, Kingston, Jamaica 2014
DM de Silva SL v Aus, Pallekele 2016
Kamrul Islam Ban v Eng, Dhaka (Mirpur) 2016/17
Craig is the only one to do so first ball; Freemen did so second ball, by hitting the ball out of the ground. The Bangladeshis, apart from Shafiul, did not do this in their first Test innings; all made ducks before hitting their first runs.
The frequency of recent cases shows how debased the hitting aspect of cricket has become, due to smaller grounds and bigger bats.
At Trinidad in 1948, umpire Henderson had to be escorted off the field by police at the end of the third day, after an unpopular decision to give Frank Worrell out caught behind on 97. Journalists in the press box thought the decision a fair one.
The Times reported in 1949 that on the second day of the Test at the Oval, Godfrey Evans hit a five, all run and without overthrows, off GF Creswell. I also came across a report of Graeme Hole hitting such a five, to the long boundary at the Adelaide Oval in 1951, on the day that 22 wickets fell (all the fielders were clustered around the wicket). This brings to 13 the number of known cases, five of them at The Oval and three at Adelaide.
At Lord’s in 1950, Clyde Walcott kept wickets in England’s first innings but opened the bowling in the second innings, with Robert Christiani filling in as glove man.
The batsman who hit the winning run at The Oval in 1936 was Charlie Barnett. This might seem rather trivial - and it is - but it means that I now have a complete set off all batsmen who have hit the winning runs in Tests, and all the bowlers involved.
For the last couple of years, that 1936 match was the last holdout and difficult to research, but I found the information in the Portsmouth Evening News, a newspaper that is now available online, through subscription to the British Newspaper Archive.
Taslim Arif (210*) scored runs off all 11 bowlers at Faisalabad in 1980. In the same innings, Javed Miandad (106*) faced all 11 bowlers, but scored runs off only ten: he did not score from the three balls he faced from Allan Border.
Consecutive runs entirely in boundaries: Test and ODI.
At Bulawayo in 2004, VS Solanki began with 9 fours = 36 runs. (He scored 56 out of his first 58 in boundaries). This is the most known in ODIs since 1999.
In Tests, the most consecutive runs scored entirely in boundaries, where known, is 52 by Shakib Al Hasan (100) against New Zealand at Hamilton in 2009/10. He went from 4 to 56 with two sixes and ten fours.
Jayasuriya played a Test innings of 32 with 8 fours at Colombo 1997 (against India). At Bridgetown in 1978, Bruce Yardley (74) started with 7 fours and a six in his first 34 runs. But this was not even a ground record. The only greater figure I have noted was at the same ground. In the equivalent Test of 1955, The Jamaica Daily Gleaner reported that the then little-known teenager Garfield Sobers started with nine fours in his first 36 runs. He was out for 43 with ten fours.
Just some bits and pieces. Here are some stats on run outs in Tests, presented in a way you may have not seen reported before. They cover Tests in this century up to 2014. The excess of non-strikers being out appears to be related to strike-farming with tail-end batsmen.
Opposing Captains in Most Tests
Batsmen Dismissed Twice in the Same Session
Some notes on the Question by an Ask Steven commenter "Who holds the most records?"
Source: Cricinfo Test records batting section
Qualification: highest position on a high-performance list that has more than one name.
Number of appearances:
DG Bradman 15
BC Lara 7
SR Tendulkar 5
Take out the somewhat artificial ‘milestone’ records (fastest to ‘x’ number of runs etc) and Bradman drops to 10, Lara 6, and Tendulkar 3.
Murali has about a dozen in the bowling, but some of these are really subsets of bigger records (e.g. "most batsmen out caught") and half of them are artificial 'milestone' records ('fastest to x number of wickets')
Bobby Abel spent the first 429 overs on the field in the SCG Test of 1891/92. Several others have exceeded 400. Glenn Turner managed about 420 overs in 1971/72, not at Georgetown but at Kingston. The most for a team batting first is 418 overs by Bert Sutcliffe at Delhi in 1955/56.
Frank Worrell was on the field for the first 385 overs of the five-day Leeds Test of 1957. This may have been matched or exceeded by Glenn Turner at Georgetown in 1972; the precise number is uncertain.
Bob Simpson was on the field for 550 out of 553 overs in the Manchester Test of 1964.
The most overs by a player who spent the entire match on the field is 413 by MS Atapattu at Galle in 2001.
Alastair Cook spent the first 1490 minutes on the field at Abu Dhabi in 2015.
These figures presume that the player was not substituted as a fielder at any stage.
Most centuries in a calendar year: Don Bradman made 22 centuries in first-class cricket in 1938, and Dennis Compton the same in 1947. I don't think these have been surpassed in the era of multiple formats. Martin Crowe holds the records for most total runs (f-c + List A), 5200 in 1987, but he only made 18 centuries, and Jimmy Cook made 17 in 1990.
The most List A centuries in a year is 10 by Saurav Ganguly in 2000, but he made no f-c centuries at all in that year.
With England chasing 234 in an ODI at the SCG in 1987, Allan Lamb reached 59 without hitting a single boundary, but was then faced with the task of 18 runs off the last over. He hit 2,4,6,2,4 off Bruce Reid to win the match with a ball to spare.
I believe that Renshaw is the 20th Australian to bat unbeaten through the first day of a Test, based on a minimum of 450 balls in the day. There have been 36 previous instances, including Justin Langer five times. Renshaw is the youngest, displacing Graeme Wood who was aged 22.
Some cases of players not being present at the start of a Test:
· Everton Weekes was selected to play for West Indies (Kingston 1948), after he was previously told that he had been dropped, but word got to him so late that he couldn't make it to the ground on time, and he actually saw play in progress from the air as he flew in. Weekes scored 141, the first of a still unsurpassed sequence of five consecutive Test centuries.
· Sandeep Patil did not arrive for the Nagpur Test of 1983 until late in the first day. Link: an article by A Mukherhee on the extraordinary circumstances.
· Just before the start of the Leeds Test of 1935, Maurice Leyland pulled out with a back injury. Someone was sent to fetch Arthur ‘Ticker’ Mitchell of Yorkshire; he was found pottering in his garden. Normally an opener, Mitchell batted down the order on the first day.
· At Sheffield in 1902, some odd selector shenanigans led to S.F. Barnes being belatedly informed, by telegram on the first day, that he was to show up and play. He arrived late, but bowled first change and took 6 for 49. Barnes, who for much of his career operated outside the county system (although he was playing for Lancashire at the time), had been a success on the 1901/02 tour of Australia, but this was his first Test in England. He took wickets with the second and third balls he bowled in a Test in England.
· UPDATE: At Johannesburg in 1994-95 Aamir Nazir was called as a replacement but had to fly in from Pakistan and arrived at the ground 36 minutes after the match had started. The South African captain permitted a substitute while Pakistan fielded during this period. Nazir broke down and was unable to finish an over twice on this first day. It is the only case I have on record (up to 2015) of a bowler breaking down and not finishing an over twice in one match.
There is an increasing availability of old newspapers online, which extends the detail available for old Test matches. One subscription service, the British Newspaper Archive, is particularly helpful for some Tests in England. I used it to get more detail on one of the more intriguing pre-War innings, a score of 56 by Clifford Roach at The Oval in 1933. There is no original surviving score from this match.
Roach scored his runs in 45 minutes, and reached 50 in 33 minutes, making it competitive with the fastest innings of its day (or any day). What the newly-available papers were able to confirm was that Roach reached 50 in the ninth over of the innings. The number of balls he faced is still uncertain, but a reconstruction suggests that the strike favoured Roach, and the 50 came off about 32 balls. This is very similar to a number from a similar reconstruction of John Brown’s 50 in 28 minutes in 1895.
Reaching 50 in the ninth over is extraordinary in any era. Even in these times of Superbats, which dominate this category, Roach’s effort rates very highly…
Earliest to Reach Individual 50 in an innings (total balls bowled)
At Karachi in 1985/86, Mudassar Nazar may have reached 50 in as few as nine overs against Sri Lanka.
*Update 6th January
Ironically, it took Roach two hours elapsed time to reach 50, thanks to a lunch break and some rain. Roach was 24 in five overs at lunch, and reached 50 in the fourth over on resumption. There was one other rare incident: a ball that went for seven leg byes in the third over, when Roach was facing. Had it been called runs off the bat, Roach’s fifty would have come an over earlier. (There is only one other instance of seven leg byes known, in 1989.) Incidentally, I also determined that a pair of ducks in this match, by HC Griffith, was not a king pair; he was out first ball in the first innings, but second ball in the second.
On a similar subject, here are some notes I have made on the claim by Farouk Engineer that he scored a century off 46 (or 48) balls at Chennai in 1966/67. Engineer was 94 not out at lunch on the first day, and claims to have hit a six off the first ball thereafter to reach his century…
I don’t have any scorebook for this Test. However, 46 balls (I have also read 48 balls) is effectively impossible. For one thing, in reality Engineer took 23 minutes and (probably) eight overs after lunch to reach his century against the spin of Sobers and Gibbs. After Gibbs had taken a wicket in each of his first two overs after lunch, there was a maiden by Gibbs to Engineer, and Engineer reached his century with a single to midwicket in Gibbs’ next over. He reached 100 in 143 minutes with 17 fours and was out, for 109, twelve minutes later.
With Hall and Griffith opening, there were only 28 overs bowled before lunch, so scoring 94 off that was a quite remarkable achievement, and 44 overs between lunch and tea with Sobers and Gibbs bowling spin.
The Hindu newspaper records 44 scoring shots in his 109 (18x4, 2x3, 7x2, 17x1), with no sixes. That paper has a detailed account, but mentions no imbalance in the strike, and it would have taken an extreme imbalance to produce a century in less than 50 balls in that time.
An interesting feature of this and other innings at the time is the disparity in over rates depending on the bowling type. In this match, Hall and Griffith bowled only 12 overs in the first hour, but when Sobers and Gibbs were bowling spin, the over rate peaked at 23 overs per hour. The rates in each hour on the first day were 12, 16, 21, 23, 11 (new ball after 75 overs) and 6 in the last half-hour. The Hall/Griffith over rates look slow even by modern standards. I am of the opinion that the overall slowdown in modern over rates is largely due to spinners taking longer to bowl their overs. Constant changing of field settings, and long conferences with captains, are factors.
I don’t yet have a big collection of ODI scores from the 1980s, but I noticed an interesting item in one of that I do have, an England v Australia one-dayer at the WACA in 1986. Record sources list this match as containing a 26-run over, scored by Ian Botham off Simon Davis (4,4,2,4,6,6).
However, the official score is quite clear: the over also contained a wide (4,4,2,4,wd,6,6), making 27 runs. This makes it the most expensive over known up to that time in ODIs (previously, it was equal leader). It was not exceeded until Sanath Jayasuriya hit 30 off an over in Singapore October 1995, and was not exceeded on a major international ground until November 1999, when 28 runs were scored, by Tendulkar (mostly), at Hyderabad.
Since then, tallies like this have become regular occurrences, thanks to the twin evils of boundary ropes and monster bats.
One can distill the progressive record in this category as follows
26 Rod Marsh off BL Cairns, Adelaide 1980/81
27 Ian Botham off SP Davis, Perth 1986/87
30 Sanath Jayasuriya off Aamer Sohail, Singapore 1995/96
36 Herschelle Gibbs off DLS van Bunge, St Kitts, 2007
Off Test-ranked bowling, the records are 32 by Shahid Afridi off CM Bandara in 2007, and 35 by NLTC Perera off RJ Peterson in 2013
The record prior to 1980 is not clear (perhaps readers can help here). The Cricinfo records do not list any overs less than 26 runs. A 1998 book, One-Day International Cricket Lists, also lists overs from 23 to 25 runs (from research by Ross Dundas), but none of those listed occurred before 1980. The most expensive over in the very first ODI was 17 runs off an (8-ball) over by Basil D’Oliveira.
To some extent, this must remain a ‘where known’ record.
UPDATE: Steve Pittard reports a 22-run over at Old Trafford in 1978, bowled by Richard Hadlee to Ian Botham. It was the last over of the 55-over innings, with a sequence 4,4,4,2,2[nb],6. Since the Dundas research found no overs of 23 or above in this period, this should stand as the record at the time.
For entries January 2013 to March 2015 click here
For entries November 2010 to December 2012 click here
For entries Apr 09 to November 10 click here
For entries Apr 08 to March 09 click here
For entries May 07 to March 08 click here
For entries November 06 to March 07 click here
For entries April 06 to October 06 click here
For entries January 06 to March 06 click here
For entries Nov 04 to June 05 click here