THE DECADE ANALYSIS: Batting and Bowling in Test Cricket

Published 02/03/2016, 12:27 AM EST



Article continues below this ad

One of the leading theoretical physicists of CERN, retired Professor Dick works in his office. Despite the look of his office, he is still the master of his universe. His individual stacking order is world famous since he only needs a maximum of 5 seconds to find ANY paper in his office. Many colleagues used his “filing” method to shortcut their own search for publications. — Image by © Peter Ginter/Science Faction/Corbis

Like any scientist, I returned to my dungeon – the statistical one to be exact – focused and never so committed to unearthing what the average untrained eye hasn’t apprehended from a coup d’oeil at cricket’s statistics. Statistical mentors or idols, such as S Rajesh and Anantha Narayanan, have delved into various decade-by-decade cricket analyses.


Article continues below this ad

Notwithstanding, a few months ago, I promised a few friends that I’d delve into my own decade-by-decade analysis of both batting and bowling in Test cricket. This task on the surface sounded simplistic and unequivocal. I quickly learned that this task might take months – let’s hope, not years – if an intense and inductive approach was explored over a deductive or abstemious one. Nevertheless, what are the a priori or the a posteriori revelations unveiled so far from my analysis?


Figure 1. Decade-by-decade (1870s – 2010s) batting and bowling averages in Test cricket.

Batting and bowling averages have mirrored each other almost perfectly for over 15 decades (1870s-2010s). Bowling averages have been higher than batting averages for each decade because of the slight difference in how both are calculated. Yet, these averages began to deviate from the perfect mirroring in the 1980s due to one major rule change in cricket.

Figure 2. The 1970s to 1980s batting and bowling averages in Test cricket

Beginning in 1983 and eventually adopted in 1985 by all Test nations, extras, specifically no-balls and wides were charged to both bowlers and the bowling team. Prior to this, bowlers were only charged with runs scored off the bat with no-balls or wides charged to the bowling team only. This explains why when batting averages decreased from the 1970s (30.76) to the 1980s (30.45), we saw a slight increase in the bowling averages from 31.90 to 32.09.

Figure 3. The 2000s to 2010s batting and bowling averages in Test cricket

Similarly, but on the contrary, batting averages have increased from the 2000s (32.02) to 2010s (32.37). However, we’ve seen a decrease in the bowling averages from 34.10 to 33.85. This trend and/or impact of extras on bowling averages will have to be evaluated in another article.


Table 1. The batting and bowling averages for each decade and their rank

DecadeBatting AveDecade Batting RankBowling AveDecade Bowling Rank
1870s-1910s17.49 – 25.9111 – 1518.18 – 26.901 – 5

The early decades of Test cricket (1870s-1910s), i.e. before World War I (WWI), we saw the lowest batting and bowling averages in Test cricket. These first five (5) decades were the ‘growing pain’ years when the game of cricket and cricketers were still finding themselves. This puts Sydney Barnes godlike bowling average of 16.43 under severe scrutiny. However, what redeems him was his astronomical elevation above his peers (over 56% better) for the two decades, which his career overlapped.


For post-WWI, we’ve seen a more predictable trend with both batting and bowling averages staying slightly above the 30-mark, with the exception of the 1950s (both batting and bowling were below 30) and the 1990s (only batting averages fell below 30).

Table 2.  The decade-by-decade (top-5) best batting averages and (bottom-5) worst bowling averages

DecadeBatting AveDecade Batting RankBowling AveDecade Bowling Rank

Footnote: The higher the batting average is, the better it is. The lower the bowling average, the better it is.

The top-five highest batting and bowling decade averages are separated into two groups, five (5) decades apart: 1920s-1940s and 2000s-2010s. Unsurprisingly, the 1920s-1940s (aka ‘The Bradman Years’) are ranked 1, 4 and 5 in terms of batting averages or 11, 12 or 15 in terms of bowling averages. This puts the records of greats like George Headley, Herbert Sutcliffe, Denis Compton, Eddie Paynter, Dudley Nourse, Wally Hammond, Jack Hobbs, Jack Ryder and Sir Donald Bradman under the microscope (Len Hutton is excluded from this list for reasons I’ll share later). However, what about Bradman? Are there any points to save Bradman? I’ll look at these in another article in the series.

Table 3. Batsmen who averaged over 50 during the run-gluttonous 1920s – 1940s (min. 20 matches)

PlayerMatchesRunsHighest ScoreBatting Ave100s/50s
Don Bradman (Aus)52699633499.9429/13
George Headley (WI)212173270*63.9110/5
Herbert Sutcliffe (Eng)54455519460.7316/23
Denis Compton (Eng)36313220860.2313/12
Eddie Paynter (Eng)20154024359.234/7
Dudley Nourse (SA)26246923158.788/12
Wally Hammond (Eng)857249336*58.4522/24
Jack Hobbs (Eng)33294521156.6310/12
Len Hutton (Eng)41378836456.5311/17
Jack Ryder (Aus)201394201*51.623/9



The 2000s and 2010s (aka “The Ricky Ponting, Rahul Dravid, Jacques Kallis, Kumar Sangakkara, Shivnarine Chanderpaul, et al Years”) are ranked 2 and 3 in terms of batting averages and 13 and 14 in terms of bowling averages. This compels us to put the records of the modern greats like Ponting, Kallis, Dravid, Sangakkara and Chanderpaul through a commission of inquiry. Similarly like Hutton, I have excluded Sachin Tendulkar, Steve Waugh and Brian Lara from the list for soon-to-be-obvious reasons.

I will not get into what saves one player over the other. All I’ll say is that Kallis has the highest batting average during this time and is 2.19% better than the next best player (Sangakkara). Remember Bradman was 56% better than #2. Sangakkara takes the throne from Kallis by some margin in matches he plays as a specialist batsman only (not a wicketkeeper-batsman). His batting average becomes 14% better than the next best player, Kallis. Remarkable! Is Sangakkara the Bradman of this generation? Let me be careful, lest I’m stoned for blasphemy by the pundits.


The 1950s through to the 1990s were the days of great bowling post-WWI. They are ranked 6 through 10 in terms of both batting and bowling averages. Since post-WWI, the 1950s and 1990s – specifically – as mentioned above were the two decades that had extremely low batting and bowling averages. This elevates the status of players like Clyde Walcott, Gary Sobers, Everton Weekes and Neil Harvey who flourished with averages over 50 during the run-lethargic 1950s (min. 20 matches). Earlier I mentioned that I removed Hutton’s name from the list of questionable great players’ who played in the 1940s. This is because he averaged 56, both in the run-gluttonous years of the 1940s and the run-lethargic years of the 1950s. Simply, Hutton was a model of consistency.

Table 4. Batsmen who averaged over 50 during the run-lethargic 1950s (min. 20 matches)

PlayerMatchesRunsHighest ScoreBatting Ave100s/50s
Clyde Walcott (WI)33312922061.3513/11
Len Hutton (Eng)38318320556.838/16
Gary Sobers (WI)272213365*56.746/7
Everton Weekes (WI)39338320753.6910/17
Neil Harvey (Aus)56457320550.2515/19

Similarly, Tendulkar, Waugh and Lara were excluded from the list of questionable greats because they averaged over 50 both in the run-gluttonous years of 2000s-2010s and the run-lethargic years of the 1990s. I can hear a friend of mine uttering when he reads these points: “Form is temporary, but Class is permanent”.

Table 5. Batsmen who averaged over 50 during the run-lethargic 1990s (min. 20 matches)

PlayerMatchesRunsHighest ScoreBatting Ave100s/50s
Sachin Tendulkar (India)69562621758.0022/21
Steve Waugh (Aus)89621320053.1018/28
Brian Lara (WI)65557337551.6013/29
Graham Gooch (Eng)45417633351.5512/17



Now to my favourite people: cricket pundits. The following questions might be posed:

1. Were rule changes the reason for the explosion of batting and bowling averages in the 2000s and 2010s?

2. Were flatter pitches the reason for this explosion after 50 years of things relatively in check? Others will ask or postulate the popular point that poor bowlers are the reason. I’ll beseech you to also consider asking, is better technology in bat construction, or smaller grounds, and/or the influence of shorter formats, the reason(s) for higher batting and bowling averages in the 2000s-2010s?

Others will contend that the inclusion of weaker teams such as Zimbabwe or Bangladesh is the cause. We have always had weak teams in one decade or the other. Nevertheless, I’ll tell them that from my analysis, Zimbabwe and Bangladesh are NOT the cause. So, another theory that is all-encompassing needs to be proposed.

In closing, I’ll quote the great cricket statistician Narayanan’s statistical partner or assistant, Milind: “Sometimes a talent requires the recognition of being an awardee. Then there are cases when the award needs the prestige by associating itself with a talent that transcends awards.” This can be said of many players mentioned in this article, I truly hope one day that might be said in reference to me.

Until next time….

© Zaheer Clarke


Article continues below this ad

Writer’s Note: Two weeks ago, Zaheer E. Clarke and his blog, Zaheer’s “Facts, Lies and Statistics”, were awarded the 2015 Press Association of Jamaica Award for “Best News, Sports or Current Affairs Blog”. Eerily, this article – with Milind’s quote – was written in September 2014.

Editor’s Note: All statistics for this article were taken from ESPN Cricinfo’s Statsguru on December 18, 2015

Blog: Zaheer’s “Facts, Lies and Statistics”


Article continues below this ad





Zaheer E. Clarke

10 articles