July 19, 2012

Test Bowling: location summary, by innings and vs country

Anantha Narayanan
Malcolm Marshall had an above-par performance against all teams  © Getty Images

This completes the sequence of four articles in which I analyzed ODI/Test Batting/Bowling performances at home or away, against different teams and in the first or second innings. While the graphs and tables provide immediate and easy viewing, the core of the articles is the exhaustive set of Excel tables which could be downloaded by readers and different types of analysis done by themselves. That is what has been done so far and hundreds of informed and insightful comments have come out so far. Normally I do analysis-centric articles which take on and expound a theme. Once in a while I do different types of articles in which I go deep in one area of the game and provide data tables around it. This is one such article.

This information is certainly available through StatsGuru of Cricinfo. However, what will not be available are the composite multidimensional tables which are provided here. You would have to put in multiple queries and saving the tables in an accessible format is another problem.

In order to avoid the usual questions and comments which relate to specific players, let me explain how these series of articles have been structured. I have covered the top/selected 10-15 players in a graph to visually present the variations. Then I present data tables, in the body of the articles, which normally cover the top 30 players or so. However the most important of the tables are the ones which have been uploaded and are available for downloading for permanent storage and perusal. Normally these cover the complete set of players, say 160 or so, who meet the cut-off criteria. So, before coming out with comments that "Willis or Bedi or Croft is not mentioned", please download the tables and check. Superficial reading of the articles is not enough.

The vs Country grouping is simple. I have 10 countries: Australia, Bangladesh, England, India, New Zealand, Pakistan, South Africa, Sri Lanka, West Indies & Zimbabwe. And the analysis is very extensive in that it is by country played against: at home, away and across career. These being Test matches, I have also analyzed the career averages by first and second innings.

1. The criterion is 100 Test wickets for the career analysis and other analyses. I know that Bond and Mailey will miss out. However I do not want to lower the target further since the vs Country numbers would be too low. I have also raised the bar for the Home/Away analysis so that we are able to look at the real performers.
2. Bowling Average is a complete measure and explains what all needs to be communicated regarding a bowler. In addition, in my graphs, I have taken the liberty of not including pre-WW1 bowlers other than Barnes. The reasons are obvious. Averages of below 20 were quite frequent and this would have distorted the presentation.
3. There are problems with the single Australia-ICC Test match. It could be said that the ICC players played against Australia away. Fine. But what about Australia ? Which country did they play against ? And I am not certainly going to allocate part of the match runs/wickets only. So this match has been completely excluded from the analysis. So do not come out with a complaint if you see Muralitharan with 795 wickets.
4. There is no neutral location. Too few matches (probably a maximum of 20) have been played in the neutral locations for me to classify these. These are treated as "Away" for both teams, probably a very fair assignment.

First the graphs. I would only offer limited comments since I expect the readers to come out with their own comments. I might anyhow miss some obvious comment. Should not really matter. The ordering is different for different modes of presentation since we can get different insights. In general, the graphs are ordered by the concerned Bowling Average values and the tables are ordered by the appropriate Wickets captured values.

Bowler analysis - Summary by location / innings / quality of batsmen dismissed

Summary of bowling averages and wickets distrbution
© Anantha Narayanan

Look at Hadlee, McGrath and Warne. Almost the same bowling averages at home and away. Laker, Muralitharan and Imran Khan have performed significantly better at home. Davidson has performed much better away, as do Garner and Ambrose, to a lesser extent.

Most bowler have performed better in the first innings. Marshall, Laker, Warne and Donald have performed much better in the second innings. Look at Imran Khan's almost similar first and second innings performances.

Nearly 50% of the wickets captured by McGrath and Donald are top-order wickets. Ambrose, Davidson, Miller are close behind. Warne has the highest share of late order wickets, followed by O'Reilly and Muralitharan.

Bowler analysis - All matches - by opposing country

Summary of bowling averages against each team
© Anantha Narayanan

This graph requires some explanation. These are ordered by the Bowling Average values. The player's performance against the 10 teams are plotted. Blue circles indicate Bowling Average values of below 25.0 and Red circles indicate Bowling Average values above 25.0. The number of wickets and bowling averages are displayed under each country.

Possibly the best performances across all countries are by Davidson, Trueman and McGrath. Only for this composite graph, I have included a new table indicating the top three bowlers against each country. I have done a selection of mine, keeping in view both number of wickets and bowling average.

AustraliaHadlee130 @ 21.6Ambrose128 @ 21.2Laker79 @ 18.3
EnglandAmbrose164 @ 18.8Garner92 @ 17.9Marshall127 @ 19.2
West IndiesVaas55 @ 16.6McGrath110 @ 19.4Muralitharan82 @ 19.6
IndiaTrueman54 @ 14.8Donald57 @ 17.3McGrath51 @ 18.6
South AfricaBarnes83 @ 9.9Grimmett77 @ 15.6Muralitharan104 @ 22.2
PakistanCroft50 @ 19.6Warne90 @ 20.2Marshall50 @ 20.7
Sri LankaImran Khan46 @ 14.6Wasim Akram63 @ 21.3Waqar Younis56 @ 22.7
New ZealandWasim Akram60 @ 17.0Willis60 @ 18.9Bedi57 @ 19.1
BangladeshMuralitharan89 @ 13.4Vettori51 @ 16.1  
ZimbabweMuralitharan87 @ 16.9Waqar Younis62 @ 19.9 

Some of these figures have to be taken with a lot of salt. For instance, Trueman's figures against India or Barnes' figures against South Africa. The Indian batting during the 1950s, especially away, was quite poor and any bowler worth his salt would have averaged below 20. The performances to appreciate and applaud are Hadlee and Ambrose against Australia, the Caribbean trio against England, McGrath against West Indies and Donald and McGrath against India. Mention has to be made of Vaas against West Indies. Surprisingly Warne has done phenomenally well against Pakistan and quite poorly against India, despite the fact that both teams have very good players of spin.

Bowler analysis - Home matches - by opposing country

Summary of bowling averages against each team in home Tests
© Anantha Narayanan

Surprisingly the only bowlers to have above-par performances against all countries are Trueman and Marshall. Laker has a blip against West Indies. As far as teams are concerned, Australia and Pakistan have travelled well. The triple-red-circles against India are with very few wickets. I have also shown the worst three bowlers at the lower end of the graph. Sobers, Flintoff and Vettori have done worst at home. Flintoff is a surprise. To average above 35 in helpful conditions at home: maybe some opinions about his bowling prowess should be changed. This, despite a great 2005 Ashes outing.

Bowler analysis - Away matches - by opposing country

Summary of bowling averages against each team in away Tests
© Anantha Narayanan

McGrath has been the best travelling bowler. Ambrose has been equally good, although his career has been dominated by tours against Australia and England. Kumble and Harbhajan have been the worst performers away from home. Ntini also has not set any away grounds alight.

Now for the tables. Most of these are self-explanatory.

Test bowler summary: by location, innings and quality of Bowlers dismissed

BowlerCtryCareerCareerHome Away 1 Inns 2 Inns TopOrdMidOrdLateOrd
  WktsAvgeWkts~Avge       WktsWktsWkts
Warne S.KAus70825.42366~25.5551.7%342~25.2748.3%349~28.0549.3%359~22.8650.7%225220263
Kumble AInd61929.65383~24.961.9%236~37.3638.1%339~32.1754.8%280~26.6145.2%237181201
McGrath G.DAus56321.64316~21.9756.1%247~21.2343.9%329~21.9558.4%234~21.2241.6%282139142
Walsh C.AWin51924.44252~23.1548.6%267~25.6651.4%279~28.4853.8%240~19.7546.2%228128163
Kapil Dev NInd43429.65225~26.8251.8%209~32.6948.2%299~31.268.9%135~26.2131.1%21497123
Hadlee R.JNzl43122.30215~22.2349.9%216~22.3750.1%289~22.7767.1%142~21.3532.9%189113129
Pollock S.MSaf42123.12247~21.3558.7%174~25.6341.3%255~23.0560.6%166~23.2239.4%186111124
Wasim AkramPak41423.62173~23.1341.8%241~23.9758.2%242~25.7158.5%172~20.6941.5%163106145
Ntini MSaf39028.83261~24.4366.9%129~37.7133.1%245~29.0662.8%145~28.4337.2%19010892
Botham I.TEng38328.40226~27.9959.0%157~2941.0%262~28.2468.4%121~28.7531.6%155110118
Waqar YounisPak37323.56180~20.6448.3%193~26.2851.7%225~25.2660.3%148~20.9839.7%163101109
Imran KhanPak36222.81181~19.3450.0%181~26.2950.0%227~22.9162.7%135~22.6437.3%16791104
Vettori D.LNzl35934.16179~34.749.9%180~33.6350.1%225~32.6762.7%134~36.6837.3%131106121
Lillee D.KAus35523.92234~23.7965.9%121~24.1834.1%208~22.8258.6%147~25.4841.4%1669297
Vaas WPUJCSlk35529.58187~26.3352.7%168~33.247.3%223~30.9262.8%132~27.3137.2%1858981
Donald A.ASaf33022.25192~21.3558.2%138~23.5141.8%200~24.0260.6%130~19.5439.4%1648086
Lee BAus31030.82191~29.1161.6%119~33.5538.4%174~29.8456.1%136~32.0743.9%1478083
Gibbs L.RWin30929.09151~26.7748.9%158~31.3151.1%160~33.4951.8%149~24.3648.2%98100111
Trueman F.SEng30721.58231~20.0775.2%76~26.1624.8%196~21.3463.8%111~22.0136.2%1397197
Zaheer KhanInd28831.78102~35.3935.4%186~29.8164.6%188~32.1565.3%100~31.0934.7%1488356
Kallis J.HSaf27632.45163~30.6559.1%113~35.0540.9%140~38.3650.7%136~26.3749.3%1049379
Steyn D.WSaf27223.19168~22.1761.8%104~24.8438.2%160~22.558.8%112~24.1741.2%1186588

This table includes all matches, including the one-off ICC Test. The Home/Away and First/Second innings columns are self-explanatory. The last three columns contain the top order, middle order and late order wickets captured by the bowler. The batting position is important since, irrespective of the batting average, top order wickets are important for both teams. In the uploaded table I also have the average Batting average of all wickets captured by the bowler. The higher the average Batting average, the higher the average quality of batsmen dismissed.

McGrath's consistency across locations and innings and his 50% top order wickets tally are amazing achievements. Hadlee matches McGrath in all but top order wickets which is still a respectable 44%. Ambrose is in between with a top order figure of 47%. Ignoring the freak 60% figures for two bowlers, Motz and Pathan, at the end of the table, the highest top-order capture is for Vaas, with 52%. The lowest amongst the top bowlers is for Warne with 32%. MacGill is still lower at 28%.

Vaas is also the lowest in the late-order wickets tally, with 23%. Zaheer Khan clocks in with 19% in the sub-300 wickets band of bowlers. Warne is the highest with 37%.

I will let the readers come out with their comments on the following three tables. The selection criterion is 100 wickets and the ordering is by wickets captured. Let me also say categorically that, as far as I am concerned, a Test wicket is a wicket and does not go down in value because it is of a lesser team's batsman. Just as we have accepted the runs, we should accept wickets. I will always bring in the example of Tendulkar's 100 against Bangladesh in 2010. It is one of his best five efforts. But for him India would have lost. So let us not bring in the bogey of cheaper wickets. In that case we have to discount many more wickets since many teams, at different times in history, have been poor and of lesser quality. And let me conclude by saying that if Warne did not get more Bangladeshi/Zimbabwe wickets, it was because Australia did not play enough matches there, for their own own valid reasons.

I will only talk about the first data row, which is a weighted average of the bowling average of the 160+ bowlers against the specific country. This is determined by the formula: Sum(Bowler wickets x Bowling average) / Sum(Bowler wickets).

Test bowler summary: All matches vs other teams

All matches 28.0330.6019.3827.5128.2624.2529.4128.1829.0228.8822.74
MuralitharanSlk795~2354~3689~13112~20105~3382~2280~25104~22 82~2087~17
Warne S.KAus702~26 11~27195~2343~47103~2490~20130~2459~2665~306~23
Kumble AInd619~30111~3015~1792~31 50~2681~3284~3274~3174~3038~23
McGrath G.DAus560~22 5~25157~2151~1957~2580~2257~2737~22110~196~15
Walsh C.AWin519~24135~29 145~2565~2043~2263~2351~208~35 9~15
Kapil Dev NInd434~3079~25 85~37 25~3599~308~3745~2789~254~34
Hadlee R.JNzl431~22130~21 97~2565~23 51~28 37~1351~22 
Pollock S.MSaf421~2340~379~1591~2452~2043~2245~21 48~2270~2323~15
Wasim AkramPak414~2450~26 57~3145~2960~17 13~3063~2179~2147~21
HarbhajanInd406~3290~296~4843~39 43~3325~5260~2852~3956~2331~25
AmbroseWin405~21128~21 164~1915~3813~2142~2821~1914~14 8~12
Ntini MSaf390~2958~3535~1670~3436~2946~2541~24 35~3063~286~46
Botham I.TEng383~28148~28  59~2664~2340~32 11~2861~35 
MarshallWin376~2187~23 127~1976~2236~2250~21    
Waqar YounisPak373~2430~3418~1050~278~4970~20 24~2956~2355~2362~20
Imran KhanPak362~2364~25 47~2594~2431~28  46~1580~21 
Vettori D.LNzl358~3465~3651~1645~3740~48 20~4821~7351~2433~2632~27
Lillee D.KAus355~24  167~2121~2238~1971~30 3~3655~28 
Vaas WPUJCSlk355~3038~3219~2649~3130~4542~2447~3727~34 55~1748~28
Donald A.ASaf330~2253~31 86~2357~1721~2127~22 29~1943~2114~16
WillisEng325~25128~26  62~2360~1934~24 3~2338~36 
Gibbs L.RWin309~29103~31 100~2963~2311~5732~24    
Lee BAus308~31 8~4762~4153~3244~215~4750~3516~1864~236~37
Trueman F.SEng307~2279~25  53~1540~1922~2027~23 86~23 
UnderwoodEng297~26105~26  62~2748~1236~24 8~1238~44 
McDermottAus291~29  84~2634~2948~3018~3521~2927~2759~29 
Zaheer KhanInd288~3261~3631~2439~27 35~2617~4733~3428~3923~3021~27
Kallis J.HSaf275~3248~3817~1446~3518~4324~3723~38 26~3352~3021~15
Steyn D.WSaf272~2345~2622~1731~3453~1945~1919~31 22~2735~19 
AndersonEng267~3041~399~25 45~3027~2432~1848~3818~3636~2811~20

The across-bowlers-countries bowling average is around 28. The all-inclusive average is around 30. But remember that these are the top 160 bowlers. Australia has been the toughest team to bowl to, with a bowling average of over 30, even for these top bowlers. Pakistan is the next toughest at 29.4 and surprisingly Sri Lanka follows with 29.0. West Indies follows next with 28.8 and only then comes India. It is no surprise that Bangladesh is at the other end with an average below 20. Zimbabwe fares much better with 22.7.

Test bowler summary: Home matches vs other teams

Home matches 25.2526.8316.5625.8825.4922.8025.7628.2725.2925.2818.71
MuralitharanSlk493~2047~2660~1064~2165~2552~2230~2669~20 45~1761~12
Kumble AInd350~2562~24 56~24 39~2257~2839~3244~2229~2724~19
Warne S.KAus313~27  66~269~6354~2745~2269~2422~3248~27 
McGrath G.DAus286~23 5~2570~2318~1427~3347~2128~3131~2060~18 
HarbhajanInd258~2881~24 29~34 22~4225~3842~2627~3120~1712~20
Ntini MSaf249~2440~3123~1327~3218~3039~2132~18 26~2239~265~18
Pollock S.MSaf235~218~472~3256~2339~1717~2027~21 26~2050~2110~12
Lillee D.KAus231~24  71~2221~2216~1568~27  55~25 
Walsh C.AWin229~2463~22 58~2722~2411~2530~2429~207~31 9~15
Trueman F.SEng229~2050~24  53~1521~2322~2027~23 56~19 
Botham I.TEng226~2879~27  29~2737~2138~31 8~3135~32 
Kapil Dev NInd219~2628~27 42~35 10~2355~22 29~2354~261~41
AmbroseWin203~2150~23 76~1615~388~2027~298~1011~15 8~12
Hadlee R.JNzl201~2353~25 27~2434~24 41~24 10~1436~20 
McDermottAus193~26  54~2234~2632~3111~2714~1813~3035~33 
Lee BAus184~29 6~3233~3645~2726~24 23~4316~1829~196~37
Vaas WPUJCSlk180~2621~2513~2740~2119~476~308~6216~25 39~1218~33
Donald A.ASaf177~2224~34 41~2240~1811~1820~19 17~1923~171~83
WillisEng176~2456~21  30~2432~1625~23  33~34 
AndersonEng173~2712~459~25 35~3019~1923~1430~377~2927~2411~20
Abdul QadirPak168~2733~27 61~2021~4616~26  9~3128~27 
Bedser A.VEng167~2257~24  44~137~3110~1638~24 11~34 
Imran KhanPak163~1919~17  67~2214~30  31~1332~15 
Vettori D.LNzl159~3729~3217~2319~469~61 19~4214~7821~2025~266~36
Steyn D.WSaf159~2227~268~2523~3427~1836~174~22 14~2220~19 
MarshallWin157~2042~22 33~2140~2027~1815~19    
Waqar YounisPak156~2110~26 5~236~4036~12 4~3131~2623~2141~19
Kallis J.HSaf155~3130~3412~1511~5711~4514~3015~32 19~3125~2918~12
Wasim AkramPak154~2214~30 4~5618~3110~16 6~1928~2344~1630~18

For the home matches of the bowlers, the across-bowlers-countries bowling average falls, as expected, to around 25. This time there are changes. South Africa has been the toughest team to bowl to, for the bowlers bowling at home, with a bowling average of just over 28, even for these top bowlers. Australia follows next with 26.8 and again surprisingly England follows with 25.8. It is again no surprise that Bangladesh is at the other end, travelling very poorly with an average around 16.5. Zimbabwe fares better with 18.7.

Test bowler summary: Away matches vs other teams

Away matches 27.6931.2019.2126.4726.6524.7530.2426.7229.0529.0225.52
Warne S.KAus389~25 11~27129~2234~4349~2145~1961~2437~2117~406~23
MuralitharanSlk302~287~10729~1948~1940~4530~2050~2535~26 37~2326~28
Walsh C.AWin290~2572~34 87~2443~1932~2133~2222~191~60  
McGrath G.DAus274~21  87~1933~2130~1833~2229~246~3650~216~15
Kumble AInd269~3649~3815~1736~41 11~4024~4245~3230~4545~3114~29
Wasim AkramPak260~2536~24 53~2927~2850~17 7~3935~2035~2717~26
Hadlee R.JNzl230~2277~18 70~2531~22 10~45 27~1215~27 
MarshallWin219~2245~23 94~1936~259~3235~21    
Waqar YounisPak217~2620~3818~1045~272~7634~27 20~2825~1932~2521~22
Kapil Dev NInd215~3351~25 43~39 15~4244~408~3716~3335~233~31
AmbroseWin202~2178~20 88~21 5~2315~2513~243~9  
Imran KhanPak199~2645~29 47~2527~2817~27  15~1848~25 
Vettori D.LNzl199~3136~3834~1226~3031~45 1~1787~6430~268~2726~25
Zaheer KhanInd191~3025~3531~2431~28 24~2311~4123~3318~4015~3813~22
Pollock S.MSaf186~2632~347~1135~2513~2726~2318~23 22~2520~2913~18
Gibbs L.RWin183~2959~33 62~2639~238~4515~26    
Vaas WPUJCSlk175~3317~426~249~7811~4136~2339~3211~47 16~2830~25
D KaneriaPak164~3434~4225~1114~5931~4016~31 15~2615~2714~40 
Holding M.AWin163~2463~24 63~2130~227~48     
Botham I.TEng157~3069~28  30~2627~272~45 3~2226~40 
Donald A.ASaf153~2329~28 45~2417~1610~257~32 12~1920~2513~11
UnderwoodEng152~2750~31  54~2724~1411~38 8~125~63

Now for the travelling bowlers. The across-bowlers-countries bowling average is around 28, almost the same as playing at home. Even for these top bowlers, Australia has been the toughest team to bowl to, away, with a bowling average of over 31. Pakistan is the next toughest at 30.2 and surprisingly Sri Lanka follows with 29.0. West Indies follows next with 29.0 and only then come South Africa and India. Bangladesh is at the other end with an average below 20. Zimbabwe fares much better, at home, with 25.5.

To download/view the Excel sheet containing the following tables, please CLICK/RIGHT-CLICK HERE. The serious students of the game are going to have a link to this Excel file on their desktop and refer to it a few times a day.

Bowlers location summary and innings summaries.
Bowlers analysis vs Team - for all matches
Bowlers analysis vs Team - for home matches
Bowlers analysis vs Team - for away matches

No specific conclusions. I decided against coming out with any selection of bowlers. It will be a red herring.

Finally a request. Often we go off on a tangent and pursue areas which are completely off-track. In my anxiety to avoid rejecting comments (I might have rejected fewer than 100 comments over 4 years), I allow lot of freedom to readers. With greater freedom comes greater responsibility. Bring in discussion points only if they have some relevance to the topic of the article. Bring in First Class records only to support something specific. Do not bring in Lara or Tendulkar or Richards into a discussion of bowlers' performances other than for discussing specific areas. Do not highlight failures of great players just for the sake of doing it. Do not start with an agenda and take it to the nth level, just for the heck of it. If everyone wants to have the last word, there would not be a last word at all. YouTube videos are wonderful. I myself have had the pleasure of watching many a video of which I was not aware of. But use these only to enlighten readers and offer additional insights. Then the experience is enjoyable.

Henceforth I will neither publish a comment nor answer any queries related to it if I feel it is way out and is not going to steer the discussions in a positive manner. Alternately, if appropriate, I will publish after editing, making sure that the spirit behind the comment is retained.


Anantha Narayanan has written for ESPNcricinfo and CastrolCricket and worked with a number of companies on their cricket performance ratings-related systems

RSS Feeds: Anantha Narayanan

Keywords: Stats

© ESPN Sports Media Ltd.

Posted by Rajveer on (August 16, 2012, 9:42 GMT)

Great analysis. My question is regarding data collection. I am in the process of developing a similar effectiveness index for a bowler and I wanted to the factor in the top and middle order wickets. So I am curious to know how have you aggregated the data on batting order. Is there a query option in statguru that enables me to do that? [[ I have my own proprietary database and use only that for analysis. I do not use Statsguru other than stray queries. Ananth: ]]

Posted by Shafiq on (August 2, 2012, 8:45 GMT)

Great article, i wanted to appreciate you after seeing many of sub-standard comments.

Posted by Hemant on (August 2, 2012, 3:21 GMT)

Can't wait to see him bowl in australian coindtions.He may suprise me and average under 100 per wicket but I doubt it, after all, even Murali struggled on australian decks averaging about 70 odd from memory. [[ Who??? Ananth: ]]

Posted by Julie on (July 31, 2012, 17:02 GMT)

Sam, He did actually inevnt the carrom ball, Iverson's was almost the exact same but it actually spun, so it couldn't have been the carrom ball. He also never called it the carrom ball, he gave it no name at all. Mendis named it, thusly inevnting the carrom ball the same way saqlain had inevnted the doosra that had been bowled in the 60s by an Australian reserve batsman.

Posted by srini on (July 31, 2012, 15:44 GMT)

My father mentioned he saw GRV's 97 live (and I am burning with jealousy even though I wasn't born then) but that got me thinking about batsman vs bowler as in terrific batting against a fearsome bowling individually. GRV's 97 vs Roberts 7-fer, Sachin's 136 vs Saqlain's 5-fer or something of that sort. Maybe u can set a higher standard. Sorry if it seems very vague or if you've done it before.

Posted by Eri on (July 31, 2012, 5:52 GMT)

Height is one of the most imortant faoctrs. Malinga is short but you have forgotten the fact that he is a slinger and height is an independent factor when it comes to Malinga's actionYou got to have both, strength and height otherwise you will do nothing on pace oriented bouncy wickets. Curtley Ambrose lost miles of pace at the end of his career but he was still unplayable. It's coz of the bounce generated by height. I'm sick of seeing 5'8 skinny bowlers form SL. Nuwan Kulasekara has been the worst bowler in the world during the last 18 months statistically. You got to be one of the following1. Strong with a slinging action in this case height doesn't matter being short is alright2. Tall to extract bounce3. If you got neither, got to be able to move the ball bothways as stock deliveries (VAAS) (+1)

Posted by Ranga on (July 31, 2012, 3:45 GMT)

@ Sudhir: I remember Ananth churning out an excellent analysis on series efforts of allrounders. Which I think would be an excellent precursor to the article on allrounders, which would come after, as Ananth mentioned, bowling and batting articles. In case you missed it, please look into the archives. That was one article I enjoyed as it quantified how allrounders clicked in both batting and bowling in the same series. Not so surprising entries but quite a few surprising absences, made it an intriguing piece.

@ Ananth: You have done an extensive series on batting and bowling if I am not wrong, twice each, in the recent 15 months. Have we not covered every aspect - Like the quality of attack faced, nature of pitch, match conditions, etc? In addition to those, you also did an article on centuries and 5-fers. Don't too many similar articles too very soon sound repititive? I agree that they are all on different parameters, but I find all those articles having some common link. [[ You are talking of 15 months. I am looking at possible repetitions within 6 months. What do I do. Two-thirds of my articles are related to Tests and that now means about 15 a year. Previously it was more. There is batting, bowling, teams. So there are bound to be repetitions. Short of making a frequency of a month for each article, this is bound to happen. Frankly I think it is a miracle that I seem to get new ideas, many contributed by readers, to come out with new themes: albeit related but at least covering new ground. A career half or third analysis is something new. However many readers, irrespective of the article, take a single theme, be it McGrath vs Ambrose, Warne vs Murali, Marshall vs the rest, Richards vs Lara vs SRT etc and run it down to earth. The problem is whatever I do you see the same players. Ananth: ]]

Posted by sudhir on (July 30, 2012, 8:56 GMT)

Hi Ananth. Regarding the future you have mentioned in the previous comment..what will you do for all will u seperate their best batting "X" years and best bowling "X" years?? because both best years may not coincide for an all rounder let us get the batting/bowling going and then work on the all-rounders.

Posted by Ram on (July 29, 2012, 9:35 GMT)

Batsmen/Bowlers have lean/mean patches. Longevity is one factor for evaluating them, but I agree with the comment that players are better compared against each other on the basis of their best "x" years, in succession, and also "x" best years taken throughout their career (x=10 is a good sample size that can coverup injuries, loss of form etc, and a slightly lesser value of x for fast bowlers who have relatively smaller careers).

Most players take some time to make it on the big stage and towards the end also their stats often take a beating as age begins to take its toll. By taking off these periods ( first three years and last two-three years) we would be comparing their "peak" years when they were at the peak of their abilities. The best batsmen may be expected to average 65+ and the best bowlers come under 20.

Alternatively, best 50 tests and best 100 tests ( in succession and otherwise) may be alternate ways to compare great players. [[ Sarosh's idea is an excellent one and, properly conceived and implemented, can become a landmark analysis. However there is a lot of tough work ahead. I would determine for each player their best decade in batting and bowling exactly to the day. This is not very difficult to do especially as I have the 200-Test data segment available. It is just a single program. However I am going to give a third of the weight for the peer comparison and I want this peer period to be exactly matching, that means special for each player and different for batting and bowling. The only way I can do this is by creating for each of the 2000+ tests the exact 10-year-ahead batting/bowling data. Now I do not even have a container for this data and have to create one. So I am looking at this to come out during first week of September. Ananth: ]]

Posted by Dinesh on (July 28, 2012, 9:18 GMT)

Ananth : An off topic Question: How do you decide its time for a new Article to be posted? When the comments sections slows down or whenever a new Article is up and ready? #TongueInCheek [[ Dinesh A very valid question and I wonder why no one has asked this so far. I had started with 4 articles a month but stabilized at 3 per month over the past three years. However I found that I was finding it difficult to manage this workload (and other work) because of my multiple hand-wrist-shoulder related problems. The trend in an article is a slow start, move up and settle down and then a lull. This normally takes 10 days since normally my articles are long and multi-topic ones as compared to other articles which normally cover a single topic. When I was doing 3 per month, by the time this period of lull settled in, the next article would kick in and I almost had no respite or rest. Now with my settling at 2 articles a month I have these 5 days of rest for the tired limbs, as will be there until Wednesday when my next article is scheduled. Of course it happens once in a blue moon that I might send an article and another one would be ready for publication as happened with Ric Finlay's article 10 days back. I then ask Cricinfo to publish the other one and wait for most of the comments to be received. Anyhow since I have done over 80% of the articles in "It Figures" so far, this does not happen often. Thanks for the question. Ananth: ]]

Comments have now been closed for this article


Anantha Narayanan
Anantha spent the first half of his four-decade working career with corporates like IBM, Shaw Wallace, NCR, Sime Darby and the Spinneys group in IT-related positions. In the second half, he has worked on cricket simulation, ratings, data mining, analysis and writing, amongst other things. He was the creator of the Wisden 100 lists, released in 2001. He has written for ESPNcricinfo and CastrolCricket, and worked extensively with Maruti Motors, Idea Cellular and Castrol on their performance ratings-related systems. He is an armchair connoisseur of most sports. His other passion is tennis, and he thinks Roger Federer is the greatest sportsman to have walked on earth.

All articles by this writer