Matches (19)
IPL (2)
ACC Premier Cup (2)
County DIV1 (5)
County DIV2 (4)
Women's QUAD (2)
WI 4-Day (4)
Numbers Game

The unsung heroes behind the stumps

Statistically, it's tough to convert a wicketkeeper's performances into numbers which can be compared and rated. S Rajesh tries his hand at rating glovemen

S Rajesh
S Rajesh
30-Sep-2005
Perhaps numbers never do reveal the full story, but they tell a large part of it. Every Friday, The Numbers Game will take a look at statistics from the present and the past, busting myths and revealing hidden truths:


Dave Richardson: more dismissals than byes © Getty Images
They're the unsung heroes of cricket. If umpires can complain that they come into the limelight only when they commit an error of judgment, then the same argument applies - perhaps to a slightly lesser extent - to wicketkeepers as well: while a brilliant catch will not go unappreciated, a crucial mistake will probably stick in the memory much longer. Kiran More's miss off Graham Gooch at Lord's in 1990 was the difference between him making 33 and 333; Ian Healy's lapses against Brian Lara and Inzamam-ul-Haq at Barbados and Karachi converted victories into defeats. These are just three instances - jog your mind further and a whole host of such instances will emerge.
Statistically, too, it's much tougher to convert a wicketkeeper's performances into numbers which can be compared and rated. The tally of catches taken or stumpings effected is the most obvious stat, but it glorifies those keepers who are fortunate to have quality bowlers who induce edges and beat the batsmen, and belittles the ones who, unfortunately, don't have such champion bowlers in the side. Syed Kirmani had 198 dismissals from 88 matches, while Jeff Dujon raked in 272 from 81 - do those numbers in any way suggest the relative wicketkeeping skills of Kirmani and Dujon?
The other way, then, is to look at the number of byes conceded per match. This method, again, is far from satisfactory, for some bowlers are much tougher to keep to than others. With a minimum qualification of 20 Tests, this table throws some unlikely names at the top: Paul Downton, England's wicketkeeper in the mid 1980s, and Khaled Mashud, Bangladesh's current gloveman, take the first two spots. And the wicketkeeper with the highest byes-per-match figure is Don Tallon, who was a legend behind the stumps for Australia in the 1940s and early 1950s. His stat of 13.90 byes per Test is quite clearly a reflection of just how difficult it was to keep on uncovered pitches.
Least byes per Test
Tests Byes conceded Byes per Test
Paul Downton 30 84 2.80
Khaled Mashud 35 107 3.06
Dave Richardson 42 139 3.31
Ian Smith 63 253 4.02
Adam Parore 67 293 4.37
Allan Knott 95 422 4.44
Ridley Jacobs 65 293 4.51
Andrew Flower 55 250 4.55
And there's something interesting about Dave Richardson's stats too - in 42 matches, he has 152 dismissals while having conceded only 139 byes, making him the only wicketkeeper with more than 20 Test caps to have more dismissals than byes conceded. Others who have played fewer than 20 matches (and more than 10) but have achieved the feat are Denis Lindsay, the South African (15 Tests, 56 dismissals, 20 byes) and Lee Germon, the former New Zealand captain (12, 29, 24), while England's Chris Read has 35 scalps and an equal number of byes in 11 matches.
A better system to rate wicketkeepers, though, would involve looking at chances missed as a percentage of dismissals made, and then bring in byes conceded and runs scored to offer a complete picture of a wicketkeeper's contribution to the team. Wisden Cricinfo began recording ball-by-ball data for all international matches from September 2001; hence such an analysis can only be done for the last four years.
The calculations have been done in the following manner - subtract the byes conceded from the total runs scored by a wicketkeeper, divide by the matches played to get a per-Test figure, and then add the percentage of chances taken to arrive at an overall rating. The results are interesting: Adam Gilchrist heads the list, which is no surprise, while Kumar Sangakkara comes in next, largely due to his exceptional batting record (66 runs per match). At No.3 position is Tatenda Taibu. No-one doubts his wicketkeeping ability, but his stats are helped by the fact that Zimbabwe bat twice in most games and field just once, allowing him more opportunities to boost his run-tally. (If, instead of runs scored, batting average is considered, then Taibu is in sixth place, behind Gilchrist, Sangakkara, Rashid Latif, Alec Stewart and Mark Boucher.)
Geraint Jones misses out on the top ten with a rating of 109.02 (803 runs, 197 byes, 78.72% catches taken). Of the 14 wicketkeepers who make the cut of ten Tests during this period, India's two representatives - Parthiv Patel and Dinesh Karthik - bring up the rear, and their numbers tell a sorry story: Parthiv spills, on an average, one out of every three chances that come his way (46 taken, 23 missed), while Karthik's undoing has been the high number of byes conceded - 139 in ten Tests - and the lack of runs scored - only 245.
Rating wicketkeepers since Sept 2001
Tests Runs scored/ byes conc % chances held (Runs-byes)/Tests+ %chances
Gilchrist 51 3299/ 333 86.50 144.65
Sangakkara 27 1798/ 226 81.82 140.04
Tatenda Taibu 22 1244/ 122 87.27 138.27
Rashid Latif 12 588/ 56 92.68 137.02
Alec Stewart 18 994/ 150 83.64 130.53
Mark Boucher 42 1568/ 259 89.14 120.31
Ridley Jacobs 33 1252/ 177 87.39 119.96
Brendon McCullum 17 893/ 99 73.02 119.72
Khaled Mashud 34 1173/ 97 87.95 119.60
Kamran Akmal 15 510/ 86 83.78 112.05
The above method isn't perfect either, and is open to debate - should runs scored be considered at all, and if yes, then should it be given a lesser weightage? (Write in, if you have a better suggestion.) By this method, though, it's certain that most of the earlier wicketkeepers would not fare well, simply because their batting averages are much lower than the current lot's. Gilchrist best illustrates the modern wicketkeeper, who must not only be skilled behind the stumps, but also competent enough to spend long periods in front of it with bat in hand. The table below shows just how wicketkeepers have improved their batting over the years - in the 1910s, they averaged 13; today, those digits have been reversed.
Wicketkeepers' contribution with the bat
Period Wk runs/ Total bat runs % contribution by wk Wk batting ave
1800s 3057/ 45,604 6.70% 17.47
1900-1909 1825/ 33,133 5.51 17.06
1910-1919 954/ 24,591 3.88 13.07
1920-1929 2373/ 48,620 4.88 19.61
1930-1939 5829/ 81,544 7.15 25.68
1940-1949 2633/ 43,745 6.02 23.94
1950-1959 9005/ 137,508 6.55 20.61
1960-1969 12,150/ 178,064 6.82 23.59
1970-1979 14,733/ 187,345 7.86 27.38
1980-1989 15,696/ 235,573 6.66 23.64
1990-1999 25,950/ 311,164 8.34 27.29
2000- 25,100/ 276,807 9.07 31.61
In fact, of the 20 wicketkeepers who average more than 30 in Test cricket (minimum qualification ten matches), eight of them played all or most of their cricket after 1990. Among the keepers of yesteryear, Les Ames's average of 43.40 in 44 Tests is the highest, while Clyde Walcott, who kept wicket in 15 Tests, and South Africa's Denis Lindsay averaged in the 40s too.
Wicketkeepers with highest batting averages
Matches Average Period
Andy Flower 55 53.71 1992-2002
Gilchrist 73 52.65 1999-
Les Ames 44 43.40 1929-39
Sangakkara 37 40.96 2000-
Walcott 15 40.36 1948-51
Denis Lindsay 15 40.00 1964-70
McCullum 17 35.72 2004-
Stewart 82 34.92 1991-2003

S Rajesh is assistant editor of Cricinfo. For some of the data, he was helped by Arun Gopalakrishnan, the operations manager in Cricinfo's Chennai office.