Five-wicket hauls in Tests: a look across and deep - part one
Part one of a detailed statistical analysis of five-wicket hauls in Tests
Michael Holding: 14 wickets on a flat track at The Oval in 1976 • Getty Images
There is a tendency to ignore the bowlers in Test cricket. I myself am guilty of this and do not allocate equal time and effort for these forgotten species. This time I have decided to make amends by doing the article on fifers in Test cricket immediately after I finished the one on Test hundreds.
First, the term used. Let me reproduce the Wikipedia entry below.
Five-wicket haul (also Five-for, five-fer, fifer, or shortened to 5WI or FWI)
Five or more wickets taken by a bowler in an innings, considered a very good performance. The term fifer is an abbreviation of the usual form of writing bowling statistics, e.g. a bowler who takes 5 wickets and concedes 117 runs is said to have figures of "5 for 117" or "5-117". Sometimes called a "Michelle", after actress Michelle Pfeiffer.
I like the term "Fifer". However since that also refers to the foot-soldier who plays the "Fife", the Scottish flute, I am somewhat reluctant. "Pfeiffer" would be injudicious. I am not too comfortable with "Five-for", being slightly contrived and seemingly incomplete. So I will stick with "fifer", a single non-hyphenated (!!!) word and my favourite. Much better than "DLF maximum" or "Karbonn Kamaal Katch".
Some maxims have to be repeated in EVERY article since quite a few readers have a one-track mind and see what only they want to see. This is not a Bowling Ratings article. The ordering is based on an indicated measure and is visible to the reader clearly. Do not draw any unintended inferences and come out with comments based on those. There is no personal discretion involved other than setting up the parameters. In view of the size of the articles and number of tables, I have kept my narratives to a minimum.
Test Bowling is a fascinating subject. It is far more nuanced that Batting when it comes to analysis.
- The number of wickets in an innings is strictly limited to 10.
- Bowling successes are very clearly defined and measurable in terms of wickets (who and when) and accuracy.
- Bowling is three-dimensional: balls, runs and wickets. These three dimension-related values are available for all bowling spells. (Batting is also three-dimensional: runs, time, balls. Unfortunately only runs information is available for all matches.)
- Batsmen win and save matches. Bowlers, almost always, win matches. They rarely draw matches, a la Atherton, Hanif et al. But you will be surprised: wait for the next article !!! A great ODI team can be founded on top-class batting and average bowling, not a great Test team.
- 5 batsmen can score hundreds in an innings, and have done so. Only two bowlers can capture 5 wickets each in an innings.
All these nuances lead to a more exciting analysis of fifers.
It took me nearly a week to think of all possibilities, write the program, prepare the tables and then weave the article around the tables. I did so much work on the keyboard that my legs (yes, you read it correctly) started aching. This turned out to be the longest article I had ever done, barring none. So I decided to release this in two parts. This will also enable me to do some specialized requests and add those tables. At the end of the article, I have indicated the types of analysis which have been included in Part 2. Even now, the current article has been exceeded in size by only one article, the one published last, on Special hundreds.
A note on the tables. I have standardized the presentation to have the first 14 columns common. These are self-explanatory. I have shown Home/Away (H/A), Bowling Type (S for spinners), innings bowled in and Result (W for Win, = for draw and * for loss).
First the basic table. I did not do this for the hundreds. However it is necessary to start with this table in the bowling analysis since many readers may not be familiar with all these performances.
1. 9+ wicket bowling performances in Tests
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R
I have limited this to bowling spells in which the bowler captured 9 or more wickets. Only twice have bowlers captured all 10 wickets. Jim Laker's feat came 79 years and 427 Tests after Alfred Shaw bowled the first ball to Charles Bannerman. Anil Kumble's feat came a further 1015 Tests and 43 years after Laker dismissed Len Maddocks. I wonder how many years would pass before this happens again: let me say, around 2050.
Laker had another 9-wicket haul, in the same match. Muttiah Muralitharan is the only other bowler to capture 9-wkts in an innings twice. Quite surprisingly, the three spinners, Muralitharan, Abdul Qadir and Subhash Gupte, captured 9 wickets on the first day. Another wonderful spinner, Hugh Tayfield's 9 for 113 was adjudged to be the best ever bowling performance in the Wisden-100 analysis. More of this performance later. Kapil Dev, Gupte and Jack Noreiga all captured 9-wickets in an innings, in vain. Surely let us all agree that no one, I repeat no one, in the next 1000 years, if Test cricket survives that far, would capture all 20 wickets in a match.
Now for something I think is very important, performance away from home.
2. Wonderful performances, away from home
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R
In view of the importance of this classification, I have lowered the cut-off to 8 wicket captures at the risk of going beyond my self-imposed limit of 25 table entries. The table is ordered by the bowling performance.
George Lohmann, on those uncovered pitches of yonder, crossed 8 wickets mark no fewer than four times. Quite a few achieved this twice. Barnes, Fraser, Kapil Dev, Massie (in the same match) and Muralitharan. The 9-wicket captures of Hadlee, Muralitharan and Sarfraz Nawaz are probably the pick of the lot, all resulting in winning matches. Sarfraz, to boot, in the last innings. The last time this was done, was by an off-spinner on a baptism debut of fire in India.
Now for some special selections. The bowlers who captured the top-six batsmen.
3. Bowling spells in which top six wickets are captured - 1
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R BA-T Avge
This table is ordered by the average of the batting averages of the six batsmen dismissed. O'Reilly dismissed Walters, Sutcliffe, Wyatt, Hammond, Hendren and Leyland, two of these on either side of 60.0. An imposing collection indeed. Gupte accounted for Holt, Hunte, Sobers, Kanhai, OG Smith and Butcher. Muralitharan dismissed Trescothick, Strauss, Cook, Pietersen, Collingwood and Flintoff.
One cannot keep these two greats out. Muralitharan and McGrath are the only bowlers to do this twice in their career. Now for another view of the same group.
4. Bowling spells in which top six wickets are captured - 2
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R Runs
This table has been ordered by the aggregate of runs scored by the top six batsmen dismissed by the bowler. This is an indication of the mayhem which was caused by the bowler.
Gillespie's decimation of the West Indian top order, including Brian Lara, reads like this: 6, 0, 4, 0, 4, 0. Looks like a telephone number or a T20 over. See how far ahead Gillespie is of McGrath, whose numbers are 9, 1, 17, 27, 1, 0. Spare a thought for O'Reilly, who was first in the previous classification and is now last. The top six English batsmen scored 52, 63, 0, 4, 132, 153.
Now the bowlers who out-performed their compatriots hundreds of times, okay by more than 12.5 times.
5. Bowling out-performers: many times the rest of the team
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R BAvg TAvg Ratio
Jermaine Lawson's 6 for 3 had an average of 0.5. His fellow bowlers captured 4 for 73 and the out-performance ratio is a whopping 36.5. Clarke's equally amazing 6 for 9 had an out-performer ratio of 32.0 and Ernie Toshack's unbelievable spell of 5 for 2 against India, ended with a ratio of 27.0. These three are bizarre performances.
Colin Croft's is a genuine case of out-performance. 8 for 29 against 2 for 136, resulting in a ratio of 18.8. Mankad, the peerless Indian all-rounder captured 8 for 55 against 2 for 200. Shane Warne's 8 for 71 against 2 for 226 is an all-time classic. One would have expected Muralitharan present in this table. However he appears quite a few times in earlier tables but not in this one.
Please note that this table should be looked in conjunction with the 17 bowlers in Table 1. Those 14 bowlers who capture 10 and 9 wickets almost always become out-performers.
Next is an important variation of the top order wicket captures.
6. Based on difference between batting average and runs scored
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R T7W Diff Avg
This is based on the dismissals of top-7 batsmen. The bowlers who captured at least 4 wickets are considered. For each such bowler, I have compiled the sum of the difference between the batting average and the runs scored by the batsman. This has been averaged and we get the notional runs saved. This table lists the bowlers whose average runs saved value is greater than 35/40 depending on whether the bowler captured 5/4 wickets.
Jerome Taylor's once-in-a-lifetime effort of 5 for 15 is on top. He dismissed Strauss (9), Cook (0), Pietersen (1), Collingwood (1) and Prior (0). The total batting average of these five batsmen was 227.5 and the saved runs average worked out to 43.3.
McGrath captured the wickets of Trescothick (4), Strauss (2), Vaughan (3), Bell (6) and Flintoff (0). The total of batting averages for these five comes to 208, leading to a runs saved value of 38.6. Shoaib Akhtar dismissed Ponting (7), M.Waugh( (0), S.Waugh (0) and Gilchrist (5). The total batting average was 192.5, leading to a runs saved value of 45.3. Shoaib Akhtar's and McGrath's performances were also away.
Martin's was during the 15 for 5 debacle of India and Watson's was on that manic November Thursday at Newlands.
Let us now look at bowling performances in bat-fests. The match RpW value here applies to the top-7 batsmen only.
7. Bowling performances in matches with high RpW values: > 50.0
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R MatRpW
These are heart-breakers. However most of these performances have been in drawn matches, as the qualification criteria suggests. The stand-out performance is Michael Holding's 8 for 92 and 6 for 57 on an Oval shirt-front pitch, possibly the greatest match bowling performance ever. He, almost certainly more than Viv Richards, was responsible for the fine West Indian win. Harbhajan Singh has held his own on the flat wickets twice, the only bowler to do so, other than Holding. Virender Sehwag is an unlikely name in this table.
Now for a unique table. I would not spoil the fun. Pl see the table.
8. They captured 5 and only 5 wickets: but nos 7 to 11
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R
These bowlers captured fifers, no doubt. But they also captured the LAST five wickets. And, to boot, these were the ONLY 5 wickets captured by the first five them. Don't think it is easy to do that. Some other bowler could spoil the fun. One batsman could remain not out. Everything has to work. This leaves us with just 5 bowlers, almost all of recent vintage. It is ironic that Gillespie appears at the top of the top-6 wickets list and also here.
If I did not have the ONLY 5 wickets criteria, there are quite a few, 20 in all, who fit in. However these other 13 bowlers have had the satisfaction of capturing one or more top order wickets. Muralitharan appears thrice here indicating the way he dominated the late order batting.
Now for those who toiled for hours on end. These are fifers in innings of 600+ runs.
9. Bowling on and on and on ... in 600+ innings
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R
This table is ordered by balls bowled. Scott bowled a third of the team overs. Lucky he got a couple of wickets in the end. Peculiar match. A timeless Test, which was drawn, by agreement. West Indies fall behind by 577 runs and England bat again. Then Headley's famous 223 saves the match. 9 days, and no result. A follow-on and they might very well have won by an innings. I know Shri might have something to say: but strange captaincy by Hon.FSG Calthorpe, the lone "gentleman" in the team. Over 9 days, he scored 13, bowled 4 overs and batted when he should have bowled.
Commendable are McKenzie and Kapil Dev who captured 7 wickets amongst the batting mayhem although Kapil went for nearly 6 runs per over, thanks to four Pakistani centuries. Also noteworthy is Farnes' capturing 6 for 96 out of a 600+ total.
Now for some nice alternate tables. First is the one where the bowlers have been very economical.
10. 5-wkt bowling performances with RpO less than 1.0
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R RpO
These are matches in which the number of overs bowled are greater than the number of runs conceded. This is ordered by the number of overs bowled. The table is led by Gibbs who had a RpO value of 0.71 while bowling 53 overs and capturing 8 wickets. Is it is possible today ? Look at Marshall's performance, the stand-out one amongst this lot. Out of an Australian total of 401, he captures 5 for 29, at an RpO of 0.94, while his compatriots capture 5 for 338, at an RpO figure of 2.1. In fact he just misses out on the out-performer table, with a ratio of 11.7. Lohmann's 8 for 7 has appeared in various tables. Only point of question would be the dicey quality of South African batting and the minefields he bowled on.
Now for the final table in this first part article. The two extreme sets of fifers.
11. The two extremes of 5-wkt bowling performances
MtId Year For Vs Score HA Bowler BT I <--Analysis--> R
This is the one table which contains the two ends of the bowling spectrum. Fifers for 200 runs and above and fifers for 10 runs and below. Most of these bowlers have already appeared in the earlier tables and this is just a different classification. Spare a thought for poor Krejza. On debut he toils hard with a 8-for-million performance and then is forgotten.
Barring table 8, which points to a slightly negative aspect of bowlers, in which Muralitharan appears three times, he has appeared 10 times in the other 11 tables. This may not be conclusive but is a pointer to the range and depth of his bowling achievements. Lohmann, no surprise, appears 8 times. Signs of the times he bowled in. Two bowlers, contrasting in their teams' strengths, McGrath and Kapil, appear 7 times each. Wasim and Waqar appear very few times. That is a sign of the way they shared the spoils.
Let me give a preview of what is covered in Part 2. To these will be added analysis based any readers' good ideas. The Readers' selection will appear there since the analysis is still incomplete.
1. Fifers instrumental in dismissing teams for sub-100 scores in first innings.
2. Fifers instrumental in dismissing teams for low scores in second innings, while defending similar low scores.
3. Match-winning fifers dismissing teams for low scores in third innings, with the team in substantial arrears.
4. Fifers in fourth innings, responsible for winning matches by low-run margins.
and surprisingly, 5. Fifers in fourth innings, responsible for drawing matches narrowly.
6. Fifers in lost matches, with suitable cut-offs.
7. Two fifers in matches by bowlers
8. Two bowlers running through sides with a fifer each and
9. Types of dismissals - all bowled/lbw.
Anantha Narayanan has written for ESPNcricinfo and CastrolCricket and worked with a number of companies on their cricket performance ratings-related systems