The all-rounder is a special breed. Every team craves one, and sometimes dreams of two. In trouble while batting, the all-rounder steps up with a quality knock. Pray for a wicket and he responds with a vital breakthrough. Needless to say, most all-rounders are also exceptional fielders who can turn the game on its head with a brilliant catch or run out.

But then, who do you truly classify as an all-rounder? What is the yardstick one chooses to determine who the better (or best) all-rounder is? Test cricket has featured some extraordinary all-rounders. Perhaps, the greatest of them all, is Garry Sobers. An all-rounder in every sense of the term, Sobers was good enough to be picked as a pure batsman. He could bowl pace, spin – both left-arm orthodox and Chinaman, and was a fielder par excellence.

The 1970s and 1980s saw the emergence of four of the finest all-rounders – Ian Botham, Imran Khan, Richard Hadlee and Kapil Dev. In the 1990s and 2000s, the epitome of consistency, Jacques Kallis, made South Africa a formidable opponent on any surface. Over the last few years, Test cricket has been seeking the next great all-rounder. Do Ben Stokes and R Ashwin fall in this category?

What makes one a truly great all-rounder? Is it consistency with both bat and ball? Or is it the impact that they have on a match and series? Maybe it is the advantage and balance they provide to the team. Perhaps, only numbers can provide the answer.

How to classify all-rounders?

The all-rounder class is not a homogeneous one. In fact, it is an extremely heterogeneous. There are a few all-rounders who are terrific with the bat but are not capable of causing enough damage with the ball. Few others tend to be a major strike force with the ball but demonstrate little consistency with the bat.

Hence, only players who have achieved the double of 1000 runs and 100 wickets in Tests are considered. These 62 all-rounders are then classified in five categories broadly based on the average difference (Avg Diff) which is the difference between their batting and bowling averages.

AR1, Avg Diff of 20 and above: Only two all-rounders fall in this elite list – the peerless Garry Sobers and the modern great, Jacques Kallis. While Sobers has an Avg Diff of 23.74, Kallis is not far behind with 22.71. The players in this category have a very high batting average (55+ for both) and a very useful bowling average (< 35).

AR2, Avg Diff between 5 and 15: Some of the greatest all-rounders, including Imran Khan, Ian Botham and Keith Miller belong to this category. It may be early days yet but the presence of R Ashwin and Ravindra Jadeja does reflect India’s dominance in Tests. An interesting name to note is Tony Greig; the England all-rounder had his career cruelly cut short after World Series Cricket but his numbers are among the best ever (Avg Diff 8.23). Overall, this category which features nine players, is one where the players are slightly consistent with the bat (only two have a 40+ average) and very good with the ball (six of them with an average below 27).

AR3, Avg Diff between 5 to -5: Among the 16 players in this category, the stand-out names are Richard Hadlee, Kapil Dev, Wasim Akram, and Andrew Flintoff. The characteristic of this category is an acceptable batting average (only one of them with a 35+ average and eight of them with an average in the 20s) and a solid bowling average (five of them with averages below 24 and 15 of the 16 with a bowling average below 35).

AR4, Avg Diff between -5 to -15: This category features 24 players and most of them are world-class bowlers who were no pushovers with the bat. Shane Warne, Mitchell Johnson, Jason Gillespie and Dale Steyn figure in the top half of the list. Muttiah Muralitharan, Anil Kumble and Harbhajan Singh figure in the second half. Among the 16 players, only one batsman (Carl Hooper) has a batting average > 30. Further, six of the 24 players have a bowling average < 27.

AR5, Avg Diff more than -15: – The final category features 11 players who were not terrible with the bat but had slightly high bowling averages. Javagal Srinath, Ashley Giles, and Zaheer Khan feature in this category. While James Anderson and Peter Siddle are the only players with a bowling average < 30, Nicky Boje is the only player with a batting average > 25.

All-round comparison

In the last few years, R Ashwin and Ravindra Jadeja have given India a serious advantage whenever they have played at home. While Ashwin has been quite consistent with the bat in away games too, Jadeja has more or less been a constant threat capable of an explosive knock on the odd occasion, such as Lord’s in 2014. However, it is their outstanding bowling that has pulled India on top in Tests. Given that their Avg Diff puts them both in the second category (AR2), it might be worthwhile comparing them with the finest Test all-rounders across multiple factors. These include peer comparison on the batting and bowling fronts, analysis of series-level dominance, home-away performance split and the career journey (measuring the moving average difference).

Peer pressure and more

One of the best indicators of an all-rounder’s ability is how he stacks up against his peers. For this exercise, the player’s batting average is compared with the average of the top five batsmen who played during the same period. The same exercise is repeated on the bowling front.

The ratios, R1 (player batting average to the average of the top batsmen) and R2 (bowling average of top players to the player’s bowling average). Also, the proportion of wickets (Wkt %) taken by the bowler relative to the total number of wickets taken by the team (bowler wickets only) is considered.

* For Garry Sobers and Jacques Kallis, the cut off for runs is 4000. The cut off is 1000 runs for R Ashwin and Ravindra Jadeja. For all other batsmen, the cut off is 2000 runs.

Sobers, considered the finest all-rounder in cricket, is nearly 15% better than the average top batsman (R1 of 1.15). On the bowling front, the ratio R2 is 0.63 suggesting that he was a more than useful bowler but not someone who could be classified as a strike bowler. Kallis is the only other player to have an R1 exceeding 1 (1.03); his R2 (0.64) is also quite similar to Sobers’.

Keith Miller, the flamboyant Australian all-rounder, has an R1 of 0.65 and an R2 of 0.87. This suggests that as a bowler, Miller was among the very best but was slightly less dominant as a batsman. Vinoo Mankad, renowned for his heroic feats against England, has remarkably similar numbers to Kapil Dev, India’s top all-rounder for almost 15 years.

While Ian Botham, England’s greatest all-rounder, has an R1 of 0.64 and R2 of 0.73, Imran Khan, has a much better R2 (0.96) suggesting that he was a stand-out bowler. Richard Hadlee, New Zealand’s highest wicket-taker, has a low R1 (lack of consistency with the bat) but an R2 close to 1 indicating just how great a bowler he was. Ashwin and Jadeja have fairly similar numbers with the bat (low R1) and high R2 (0.93 and 0.99 respectively). These numbers suggest that the two modern all-round stars have been extremely useful batsmen but fall short of being regarded as very good. On the bowling front, however, their numbers are among the very best.

Miller, Kallis and Sobers have the Wkt % well below 20%. This clearly indicates that they played in fairly powerful teams with strong bowling attacks. Mankad, Botham and Kapil Dev have a much higher Wkt % score (25% and more). Only two players – Richard Hadlee and R Ashwin – have their Wkt % in the 30s. While Hadlee picked up an astounding 36% of the team wickets in the matches he played, Ashwin is not too far behind at 33%.

Home vs Away

Most batsmen and bowlers have better stats in home conditions. In a few cases, the contrast in performance is stark. The difference between home and away games is measured using two ratios AH1 (ratio of away to home batting averages) and AH2 (ratio of away to home bowling averages).

Sobers, who has a terrific overall average of 57.78, has a slightly low AH1 (0.75) and an AH2 close to 1. This suggests that he was more dominant at home with the bat but did equally well with the ball home and away. Miller has a higher AH1 (0.84) but a much higher AH2 indicating that he was a much better bowler in home conditions.

All Indian players on the list have a very high AH2 with Jadeja and Mankad topping with 1.85 and 1.59. This clearly points to the fact that Indian bowlers have struggled to create the same impact in foreign conditions. On the other hand, Botham and Kallis have very even distributions for both AH1 and AH2 while Hadlee has the lowest AH2 (better average in away games).

Dominating a series

All-rounders are remembered for dominating a series and turning the team’s fortunes around. Think back to Flintoff in the 2005 Ashes and Botham in the 1981 Ashes. Both had exceptional series with bat and ball to lift England from a seemingly hopeless situation. To analyse the series-level dominance, the graph shows series of three or more matches and proportion of these series in which these players have:

  1. Picked up 10 or more wickets
  2. Scored 200 or more runs
  3. Achieved both a and b

Sobers achieved a in 62% and b in 76% of the series he played in. He achieved c (200+ runs and 10+ wickets) in the same series 10 times in 21 series (48%). Miller had corresponding scores of 91%, 64% and 64% respectively. Hadlee and Ashwin have picked up 10+ wickets in a series 96% and 92% of the series they have played in. While Sobers has the highest score in category b, Kallis (68%) and Botham (65%) are second and third. Botham achieved the double of 200+ runs and 10+ wickets in 10 out of 20 series (50%) which is second only to Miller’s record. Miller, however, played only 11 series overall.

Phasing it

Given the range of skills top all-rounders possess, one can surmise that their career would have different phases when one skill dominates the other and that there is perhaps only a small proportion of matches (or series) when both they are batting and bowling at their best.

Three interesting cases illustrate this best. The players considered for this analysis are Sobers, Khan and Botham. Each player’s career is divided into four phases with each phase having approximately a comparable number of matches. For example, Imran Khan’s career (88 Tests) is divided into four phases which have 25, 20, 25 and 18 matches respectively.

Sobers had a stunning run with the bat in the first 3 phases averaging 63, 58 and 65 respectively. His batting average fell to 46 in the last phase. On the bowling front, he started very poorly averaging 49. This was a phase when he played more or less as a batsman who could bowl occasionally. In the 2nd and 3rd phases, his average was close to 30. The average difference numbers (batting average – bowling average) in the second and third phases were 27.5 and 35.4, which are nearly 16% and 49% better than his career mark.

Botham is an intriguing case. He started superbly averaging close to 37 with the bat and 19 with the ball in the first phase representing an average difference of 17.55 – more than three times his career mark. However, the batting average fell and the bowling rose in the next two phases before a major drop in form on both fronts in the final phase where the average difference crashed to -21.83. The range of average differences across the four phases was the lowest for Sobers (26.1), second for Imran (29.65) and the highest for Botham (39.38).

Ashwin and Jadeja, along with Ben Stokes, have had excellent starts to their career. However, as we have seen with most great all-rounders, it is near impossible to maintain the same level of consistency with bat and ball throughout a long career. How will the careers of these modern talents pan out? Will they also be a major force in away games? The next three-four years are likely to provide the answer to whether they will belong in the pantheon of all-round greats.

Note: Statistics as of August 28, 2017.