Test wicketkeepers: Everything you wanted to know about
Test wicketkeepers: inarguably the players with the toughest task in cricket. They have to be on their toes for hours on end, go down and up a million times a day, cannot afford to relax even for a minute, always have an eye on what is happening to the ball and nowadays are expected to bat for hours and/or score runs quickly. And some of these keepers are expected to captain their teams and/or open the batting. Why would anyone take up this tough task? This article is a homage to those tough men.
This is one long article and has taken nearly two weeks. Please take time to peruse all tables and come back with your comments. The effort I have put in would be nullified if readers do not give this the time it deserves. And please, no comments that all this information is available in StatsGuru. All the information is not available, would require many queries to get the information, the information would not be in the form of readable tables, there is no ratings work done and so on.
In order to get a reasonable number into the analysis set, I lowered my initial qualification criterion of 100 dismissals to 75. 48 keepers qualify and this seems to be a fair number. On an average this represents a lower limit of 20-30 Tests, which is a Test career between 3 and 5 years.
1. Total Dismissals (All tables current upto Test 2054 (Ind-Nzl First Test))
This is the base table and is ordered on the number of Dismissals. This is not just a longevity based achievement. To play over 100 matches, as a top-flight keeper, is something incredible and has to be recognized. An important fact to be noted is that the number of Tests shown are the Tests in which the players played as wicketkeepers. For two players, Sangakkara and Alec Stewart, this distinction is significant since they have played a number of Tests as batsmen. Andy Flower and Parore have also played a few Tests as batsmen. This table lists the keepers who effected over 200 dismissals.
The table is topped by the peerless South African keeper, Boucher, and then come three top keepers from Australia. Dujon follows next and then Knott and Stewart. The top-10 is rounded off by Bari and Dhoni.
Readers are likely to come in with comments that the misses by keepers are not included. This is true because the data is not available. Cricinfo might have the data for the past few years or so but not for over 75% of the Tests. And I will not make any guess-work. My analysis is based 100% on the available verifiable data.
2. Dismissals per Test
This is a performance measure. Dismissals per Test is a very important parameter to measure the keepers' contribution to the team. Only two keepers have effected more than 4 dismissals per Test, which works out an average of 30% of the team dismissals. Readers might argue that the bowlers create the opportunities for dismissals. However that is only partly true. The keepers have to gauge the bounce and stand in the correct position and posture. It is my belief that only one-in-three catches travel straight to the keepers. The other two have to be caught well. Anyhow 4+ dismissals per Test goes a long way in influencing the result in the team's favour. 3.06 is the overall average.
In the top-9 keepers, Langley, with a figure of 3.77 belongs to the 1950s and Marsh, with 3.70, belongs to the 1970s. The others all are current or recently retired. Why I wonder. More opportunities? Better techniques?
At the other end, we have Ames from way back, Khaled Mashud and Engineer. The last two have fewer than 2 dismissals per Test. Why should Engineer's numbers be so low, almost matching the numbers of top slip fielders. Possibly many of the spinners' wickets would have been effected through catches to close fielders, rather than keeper.
3. Dismissals as % of similar team dismissals
|Wicket Keeper||Cty||From||To||Tests||C-St||Team CSt||Dismissals %|
This is a nice-to-view table and is not used in any Ratings work. Alexander of West Indies had a hand in 44% of his team's similar dismissals. And look at Kamran Akmal: he has had a hand in over 44% of his team dismissals. That is something. Saleem Yousuf has a similar 40-plus % figure. Gilchrist makes it to the top-10, with an impressive figure of 37.2%. One-third seems to be the overall average.
At the other end, Engineer clocks in with a very low 21.1%, as expected.
The Byes per Test is a measure of the keeping quality. Somewhat indirectly, would it also have a correlation with the chances missed? Downton is somewhere in the stratosphere with an average of only 2.8 byes per Test. Richardson, the current CEO of ICC, also has a very low figure of 3.5, matched by Khaled Mashud. A few other not-so-well-known keepers have low byes/Test figures of around 5.0. 7.0 byes/match seems to be an above average performance level. 6.9 byes/match is the overall average.
The last three have all conceded more than 10 byes per Test. Ames probably has a high figure because of the way England bowled during the 1930s.
5. Stumping %
|Wicket Keeper||Cty||From||To||Tests||C-St||Stumpings||St %|
Number of stumpings and the % of total keeper-dismissals is another nice-to-have information. It really has no bearing on evaluation of a keeper performance: rather, reflects the way the bowling attacks were formed. Oldfield, keeping to Grimmett and O'Reilly, has got 40% of his dismissals as stumpings. No surprises there. Same with Ames. Lilley was a pre-WW1 keeper. Note Prasanna Jayawardene's high share of 23%, no doubt due to Muralitharan's presence and recently Herath. Two Indian keepers of old and Imtiaz come in afterwards.
At the other end are couple of West Indian keepers and the CEO of ICC. They probably did not keep to any quality spinner. Less than 2% means stumpings for these keepers were as rare as Haley's (or more aptly, here, Healy's) comet.
6. Total Runs
|Wicket Keeper||Cty||From||To||Tests||BPos||Runs scored|
Now we come to the batting measures. First the simple measure of Runs scored. Readers should remember that these are only the runs scored when the player played as a keeper. So Sangakkara, Stewart et al will have lower figures. Gilchrist leads with a tally of 5570 runs, but in only 96 Tests. Boucher has a slightly lower figure, but in 51 more Tests. Stewart and Flower, the batsmen-keepers, are next. Knott follows afterwards. Dhoni is likely to accumulate quite a few runs and may very well usurp Gilchrist from the top spot before his career is over. Sangakkara is unlikely to add any runs since he is unlikely to resume wicket-keeping duties in tests. Prior and McCullum could add a few.
The average Batting position is also shown. As expected most keepers have this figure well in excess of 6, other than Stewart, Flower and Sangakkara. This indicates that they batted in positions 7 and higher. One reason why I have not done the compilation of the runs added by the keepers with the lower order. Since they seem to have batted at no.7 and afterwards, most of the runs scored would have been with the late-order batsmen.
This is a great performance measure. I have gone on the less used Runs per Test measure than the Batting average since the latter figure is likely to be inflated seriously with many not outs since the keepers, traditionally, are late order batsmen. How many runs per Test, in addition to the dismissals, is a far more important measure of a wicketkeeper's quality than whether he remained not out. Andy Flower is the runaway leader in this regard, with 80 runs per Test, a figure in excess of many a specialist batsman. Sangakkara follows with 67. Then comes the explosive Gilchrist with 58 runs per Test, often in crunch late-order situations. Engineer, with an otherwise poor set of keeping numbers has a decent 56 runs per Test. The average is 38.7, indicating that any 40 plus figure should be considered above-average.
8. Opening Runs
|Wicket Keeper||Cty||From||To||Opening Runs|
These are the runs scored by keepers in the opening positions. Engineer is the only keeper with a tally in excess of 1500 runs. No one else has even scored 1000. Imtiaz is the second and Mongia who opened for India in a few Tests, follows next. We have to recognize this facet of Engineer's career, especially as his wicket-keeping returns are below-average.
9. Keepers as Captain
|Wicket Keeper||Cty||From||To||Tests played||Tests captained|
Dhoni is the leader in this measure, having captained India in well over half the Tests he has played. Alexander was the captain in most of the Tests. Then come Flower and Stewart.
10. Wicketkeeper Ratings
Finally I have decided to do a Test Wicketkeeper Ratings work. This is fraught with pitfalls and I must be ready for the bouquets and brickbats to come in alternately. But I realize that if I do not venture out, I (and the readers) would not gain anything and here we go. I am sure readers would have their comments. My only request is for readers to note that this is a first attempt and fine tuning is always possible. The measures which would be used in Test WK_Ratings are explained below. These are detailed in the order of importance.
1. Number of Dismissals per Test: This is a pure performance measure and carries the maximum weight. 10 points are given per dismissal per Test. The overall weight is around 50%.
2. Number of Runs per Test: The second most important measure. The batting contributions are second only to the basic dismissals per Test measure. The value to the team, especially as the runs are normally scored in the late order, is undeniable. The points are determined by multiplying the Runs per test measure by 0.3333. The overall weight is around 22%.
3. Number of Byes per Test: This is an important measure since this indicates wicket-keeping quality. The overall weight is around 13%.
4. Number of Dismissals: This is the only longevity-based measure. It is very essential to reward the keepers who have maintained their fitness and skill levels for many years and Tests. The allocation is one Rating point per 40 dismissal so that the maximum is 15 points. The overall weight is around 8% but there are keepers who go above 15%.
5. Number of Top order dismissals per Test: Again a specialized performance measure. There is a clear separation amongst wicketkeepers and this has been included since it means that much more to the team. The overall weight is around 3.5%.
6. Number of Tests captained: The additional responsibility of captaining has to be incorporated in the Ratings. However it must be noted that keepers have captained their teams in around 5% of the Tests played. The final weight is only around 0.7%. But the concerned keepers get allocated much higher level of points: Dhoni gets around 8%.
7. Number of Test runs in opening position: This is to recognize the keepers who, in 10 minutes flat, changed their gloves and pads and started an equally difficult task. However let me add that this measure, like the Captaincy measure, also has an overall low weight of around 0.4%. Fewer than 4% of the runs scored by keepers are in the opening position. But the concerned keepers get allocated much higher level of points: Engineer gets around 6%.
The final ratings results are as expected.
Adam Gilchrist is the undisputed no.1. I am not sure whether anyone could dispute this honour, especially since the scoring rate has not even been factored in. He has achieved this through an outstanding keeping performances and above-average batting performances. Mark Boucher is in the second position. His is more due to the keeping abilities. Andy Flower, a top-flight batsman, very good keeper and a born leader, is in third position. Let us forget the poor Tests he might have had: Kamran Akmal's performances have been outstanding and he fully deserves his no.4 position. Dhoni, assisted strongly by the captaincy duties, completes the top-5. This is a reflection of the overall value to the team. Subjective factors such as quality of keeping, keeping stance, method of catching etc do not (and should not) come in. The legendary Marsh and Healy are in the top-10 positions.
11. A BCG Chart of Test wicketkeepers
© Anantha Naryanan
Ah I am back to my favourite BCG charts. This is a perfect set of data elements to do a BCG chart and classify wicketkeepers into keeping-centric or batting-centric roles. The X-axis represents the more important measure: C-St per Test. The Y-axis represents the batting measure: Runs per Test. It can be seen that both are performance measures. Longevity does not come in. To make sure that we get in only players who have had a reasonably long career I have selected only keepers who cross the dual barrier of 100 dismissals and 1000 runs. 35 keepers qualify.
The top right quadrant contains the real top-fliers: those keepers who have done well both as keepers and batsmen. The quadrant is led by Gilchrist, followed by Kamran Akmal, Haddin, McCullum, Dhoni and Prior. All these are modern keepers and are here because they are all top quality batsmen.
The bottom-right quadrant is heavily populated and houses the top keepers who do not have great batting figures. This pack is led by Boucher and has Jones, Richardson amongst the modern players and Marsh, Dujon, Healy amongst the older players. Russell and Murray just about make it.
The top-left quadrant has the players who have batting figures comparable to any top batsmen but lag behind in keeping numbers. Andy Flower leads this group and other prominent players are Sangakkara, Stewart and Knott. The last is a surprise.
The bottom-left has a set of keepers who do not really make the cut. Average keeping returns and average batting returns. Kirmani, More, Oldfield, Evans, Mongia, Smith are in this lot.
Please note that the quadrant division is a subjective one and this graph is only a tool for visual separation. The Wicketkeeper Ratings table is far more in-depth one and incorporates all relevant measures.
12. 12 top innings played by Test wicketkeepers
Gilchrist - 149 (163) Pak 1469 1999 - Huge match-winning/saving stand with Langer Gilchrist - 144 (212) Bng 1797 2006 - Match saving/winning innings Gilchrist - 138 (108) Saf 1593 2002 - Match-winning inns in close match A Flower - 232*(444) Ind 1517 2000 - Possibly the best match-saving innings in India A Flower - 199*(470) Saf 1562 2001 - Another defensive classic (142 in first innings) Ian Smith - 173 (136) Ind 1139 1990 - Attacking match-saving inns (from 131/7 to 391) Rod Marsh - 110 (173) Eng 800 1977 - Priceless match-winning inns in close centenary Test K Akmal - 109 (154) Ind 1738 2005 - Match-saving away classic (from 243/6) J Dujon - 139 (158) Aus 997 1984 - Match-winning first inns (From 186/6) A Knott - 135 (210) Aus 806 1977 - Match-winning first inns (From 82/5) N Mongia - 152 (366) Aus 1335 1996 - Match-winner while opening the innings Imtiaz Ahd- 209 Nzl 414 1955 - All-time classic, coming in at 111/6 and taking score to 561.
This has been presented in no particular order and based my own, often subjective, perusal of scorecards. I have also used the Hallmark-100 ratings numbers as a guideline. It was indeed tough to leave out Gilchrist's 102 in 59 balls, Engineer's 121, Sangakkara's 230, Healy's 161 et al. But let the readers come out with their suggestions.
13. A combined table of Test and ODI dismissals
One day Boucher might be quite happy to have stopped at 999 international dismissals since that is likely to be talked about more than if he had reached 1000, a la 99.94. One is a longevity based achievement and the other is performance-based one. It is unlikely that either achievement would ever be surpassed. Gilchrist follows closely with 900 plus dismissals and this mark is also almost unreachable. Healy follows, after a couple of miles, with 628 dismissals. This clearly shows the gap between the top-two and the rest.
Gilchrist leads both formats in the key CST per match measure. There is no combined CSt per match measure since the scale of the numbers is quite different for different formats.
At the end of this long article, I am as tired as the keepers might have been, after a tough 7-hour day in the office, except that they were baking in the sun. May their tribe flourish.
To download/view the comprehensive Excel sheet containing all the tables related to Test wicketkeepers, please CLICK HERE.
New table added.
Wicket keeper Balls kept analysis
|Pace > 70%|
|Spin > 50%|
The results are as expected. The West Indian keepers dominate the first part of the table which shows the keepers whose share of pace balls is greater than 70%. And India and Sri Lanka dominate the second half of the table which shows keepers whose share of spin deliveries is greater than 50%.
Based on comments of Biswa, Boll and Raghav, I have modified and posted the Excel sheet with the addl worksheet on a /Inns basis rather than /Test basis. All measures have been done on a per innings basis. Pl download and view the same. There is more churning than what i expected. In summary,
Kamran Akmal, Flower go down and Dhoni, Marsh, Dujon and Healy move up. The revised order is
Gilchrist Boucher Dhoni Marsh Healy Dujon Kamran Akmal Richardson McCullum A Flower.
This, to me, seems a far more acceptable order in the top-10. All the top wicket-keepers are there at the top.
This article will be incomplete without a reference to Dennis Lindsay. He had, during 1966-67, against Australia, almost inarguably, the best series for an all-rounder (in the extended definition).
606 runs at 86.57, 24 dismissals and 6 byes (1.2 byes/Test). Unfortunately he played only 15 Tests. 61 dismissals and 1130 runs at 37+. So he had a truly wonderful series but no more than good otherwise. But a truly amazing byes/match value of 1.33. Many thanks to Gerry for poking me in the ribs about Lindsay.
Anantha Narayanan has written for ESPNcricinfo and CastrolCricket and worked with a number of companies on their cricket performance ratings-related systems