Test cricket September 24, 2010

# Test teams: an analysis of results across ages

A graphical analysis of Test results for teams across the ages
59

This is a simple analysis of the results of teams across ages. I have split the 133 year period into the following 8 ages.

```PreWW1: 1877-1914
PreWW2: 1921-1939
1950s:  1946-1959
1960s:  1960-1969
1970s:  1970-1979
1980s:  1980-1989
1990s:  1990-1999
2000s:  2000-2010
```

A simple formula is used. Readers might find this a little simplistic but I am working with limited parameters to do justice to such a macroscopic analysis. My idea is to bring to light the teams which performed well during each period and then see how each team performed over the years since they made their debut in international cricket. Many of these insights might be obvious to some of the readers but this article is a single place compendium of team performances across the years. And the normal complaints of comparing players/teams across the ages do not arise in this analysis.

A win carries 2 points. A draw/tie will carry 1 point. The total points will be compiled during the concerned period. This is evaluated against the maximum points available for the team and a Performance % arrived at. There is also a need to recognize away performances. This is especially needed to break deadlocks. Take two teams which have played 10 matches each. Both win 5 matches and draw the remaining 5 matches. Both teams will have a 75% performance index. If team A won 3 away and 2 at home and the other team 2 away and 3 at home, Team A should be considered to have done slightly better. Hence I have provided 25% additional weight for away performances, that too, only for wins and draws. The actual weight given is less consequential than the fact that the away performances are recognized.

This is a simple analysis based on results. The relative team strengths or the series position or the win margins are not considered. That is a totally different type of analysis of Team Ratings.

Reg the Graphs. The first graph is the one covering all 1971 tests. This is across 133 years. This graph can be used to lead on to the other graphs. The Period graphs have been drawn in the order of teams' performances. I have also included the summary table for the period in the right as part of the graph for easier viewing and identification. At the end of the 8 period graphs, the graphs for the teams are drawn.

Summary of Test results across ages

Australia leads the all-time table comfortably with a Performance value of 66.0%. England come next with 59.3%. It may be a surprise that Pakistan edges out West Indies for the third position. This has been a result of the recent fall from grace of the West Indian team. Again it is a surprise that Sri Lanka edges out India for the fifth place although it must be admitted that India has had a 50 year head start to put in some awful years earlier. This has also been made possible by Sri Lanka's strong showing during the 99 tests played during the 2000s. New Zealand is the only leading team to have an overall sub-50% index value.

Summary of Test results in the 2000s © Anantha Narayanan

The dominant team during 2000s has been the Australians with a Performance Index (PIdx) value of 84.6%. Even their recent wobble has only got them down a bit. They are still the team to beat. South Africa are next with 66.6% and then India with 63.9%. It is debatable whether India can maintain this ascendant graph over the next few years with the huge void which is going to be created. England, with its periodic high-level performances are in next and Sri Lanka, buoyed by their strong home record, complete the top-5. Pakistan comes in next despite their continuing problems and their inability to play at home. The next 4 teams each have significant daylight between themselves and the team ahead of them. One reason why the 2000s has seen a wider dispersion of the numbers are the increased number of decisive results (only 23.8% draws) and presence of two weak teams.

Summary of Test results in the 1990s

The dominant team during 1990s has again been the Australians with a PIdx value of 69.7%. They have not had the extent of domination they had during the last decade. South Africa are next clocking in at a very close 68.2%. Pakistan, with the lethal bowling attack and great batsmen, are next placed with a PIDx value of 65.0%. The West Indies, not be confused with today's hapless and dispirited team, were fourth placed at 56.8%. India completed the top-5 barely crossing 50%. Sri Lanka, England and New Zealand were closely bunched around the 45% mark and even Zimbabwe clocked in at a respectable 34%.

The interesting point in this period was the close bunching of the teams. The difference between the first and ninth team was a low 35% as compared to the 2000s where this difference is a whopping 74%. The other surprising feature is the low number of matches played by teams other than the Ashes rivals. There have also been a greater number of draws (35.7%).

Summary of Test results in the 1980s

It would not be a surprise to read the 1980s charts. The dominant team, by a mile, was the great West Indian team, with their quintet of outstanding pace bowlers and feared batting attack. They clock in at 81.2%. Next comes the Imran Khan controlled Pakistan with 61.9%. Now comes the Hadlee-inspired New Zealand with 58.0%. Australia is the only other team to have a 50+%. India's lack of match-winning players kept them in the lower half. England comes in next and finally the new entrants, Sri Lanka. This period witnessed 46.1% draws.

Summary of Test results in the 1970s

The 1970s was an interesting period. Packer and World Series happened. England, probably less affected by WSC than Australia and West Indies were the leading, if not dominant, team with PIdx value of 63.3%. They are followed by four teams with 50+ %, led by West Indies. India, no doubt bolstered by Gavaskar and the spinners, did not do too badly. New Zealand had only a 33.5% index value. A tweak had to be done for this decade. South Africa played 4 tests and won all these. The 100% index value is an anomaly and should be removed from the analysis. This has been done. No major impact, though.

Summary of Test results in the 1960s

The 1960s was very much a defensive era as evidenced by the single digit column of wins for four of the six teams. The three leading teams, West Indies, Australia and England were separated only at the decimal point level, that too only because West Indies had slightly better away results. The close bunching of teams during these two periods, 1960s and 1970s, is a reflection of the parity which existed between the teams. It is also caused by fewer decisive results 42.6% and 47.8%).

Summary of Test results in the 1950s

The post-war period of 40s/50s was probably much better than the later dreary period. Bradman was there to start with. His legacy was continued by strong players. Australia had an outstanding PIdx value of 78.3%. England also had a very good team and were second with 61.7%, very closely followed by the W-driven West Indies. Surprisingly, Pakistan the new entrants were the next team having a better than 50% record. This is probably the best entrance decade for any of the later entrants. One must also allow for the fact that the pitches were conducive to the great strength of Pakistan, their seam bowling. The draw % was around 35%.

Summary of Test results pre World War 2

The in-between Wars period was a two team period with Australia comfortably ahead of England. That England, despite Bradman, were only 7% behind Australia indicates the effective manner in which their strategies, starting with body-line, worked. The newcomers, India, New Zealand and West Indies propped up the table. There was a spurt in the draw % compared to the previous era, 37.1%.

Summary of Test results pre World War 1

There were only three teams before WW1. England were the comfortable leaders during this period, no doubt aided by their bowling attack, led by Barnes and Lohmann. Not to forget Hobbs and Sutcliffe. Only 17.9% of the matches were draws, no doubt contributed by the types of pitches.

Now for the team performance graphs, presented in a different format. I have used line graphs instead of the bar graphs since it is easier to follow the changes. Also the graphs are shown in a chronological sequence. There is no graph for Bangladesh which has had one decade nor for Zimbabwe which has had two decades. It is not possible to derive anything sensible without three decades.

Summary of Test results for Australia

Australia has maintained very steady performance levels throughout the 133 years. they are the only team never to have fallen below 50% in any of the periods. What is important is that Australia have topped in 4 out of the 8 periods, the PreWw2, 1940s-50s, 1990s and 2000s period.

Summary of Test results for England

Barring the 1980s and 1990s, England have always maintained a 60+ % level. That is a consistency which is comparable to that of Australia. They have led the table in two of the eight periods, the 1970s and the Pre-WW1 periods.

Summary of Test results for West Indies

West Indies led the table during two periods, the 1960s and 1980s but have since fallen off drastically, especially during the past decade. Their 80+% can be compared only to the Australians of the 2000s. Compared to the awful 2000s even the average 1990s looks good.

Summary of Test results for India

India have had a poor start, understandable, and had a poor 1980s and barely acceptable 1990s. They recovered in the current decade although the huge chasm is in front of them. The day without the three gladiators at 3/4/5 is looming ahead. The bowling is another major concern. Where are the bowlers to take 20 wickets on good pitches?

Summary of Test results for South Africa

Barring a slight dip in the current decade, South Africa have improved their figures every decade. Possibly the only team to do so. There is a caveat so far as South Africa are concerned. This has already been referred to in the period graph. They played 4 Tests during the 70s and won all. Since I did not want their graph to have an abrupt dip or spurt, I have allotted a notional % for this period. 1980s, of course, is excluded.

Summary of Test results for Pakistan

Pakistan started very well, dropped off, picked up very well again but again fell of during the current decade. Overall they have been quite good. Very understandable in view of the circumstances. We must feel for the talented Pakistanis. As they seem to come out of one problem, another one crops up. Maybe it is time for Imran Khan to come forward and run Pakistan cricket the way he ran his team.

Summary of Test results for New Zealand

New Zealand have been like the proverbial yo-yo. Down, up, down, up and so on.They had a golden 1980s when the kings were scattered around the tropical islands near Florida..

Summary of Test results for Sri Lanka

Sri Lanka have had only three decades and have been steadily improving a la South Africa. However the impressive thing is that their excellent performance of 61+% has been over the last 100 tests, more than a half of their tally.

To view the "Results Summary - By Periods" tables, please click here.

To view the "Results Summary - By Teams" tables, please click here.

An important announcement to the readers. I have created an open mailid to which the comments and suggestions, not meant for publication, can be submitted. The mail id is ananth.itfigures@gmail.com. Since the readers would have to use a mail route I give the readers my assurance that the mail id is safe and will never be used by me for anything other than communicating with the reader specifically. This will not be part of any group mail nor will mails be cc'd.

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

• fanedlive on October 14, 2010, 16:22 GMT

Hi.... This is my 1st post...although i have read your few articles before, but i found this one very interesting & can't wait to see the similar analysis of ODI, where it would be more interesting to see the result.

A request - pl. also point out batsman & bowlers over the same periods who have done well & how they fared home & away.

• fanedlive on October 4, 2010, 1:23 GMT

Really excellent analysis! Shame the charts are not colour-coded consistently by team, that would have been the cherry on the cake. (I love the fact that more charts are appearing now in this column; perhaps it is time to think about standard colour-codes for the main international teams when they appear on such charts! Maybe by ODI kits - green for Pakistan, black for NZ, red for Zim...)

People uneasy that e.g. 1975-1985 isn't being represented as a time period have a point - using "years ending in zero" as cut-offs is in one sense totally logical but in cricketing terms also quite arbitrary. But to draw up tables, you have to cut off somewhere, and cutting off by decade (and by inter/pre-war period in the periods with fewer tests played) at least removes subjectivity. For charting, it would be possible to produce line graphs from 1960 onwards, using 10yr moving averages for the teams, so e.g. 75-85 appears and smoother development visualised. I doubt it'd show any new trends though.

• fanedlive on October 2, 2010, 0:04 GMT

Hi, I have always read your blogs & found it very insightful & intriguing. Especially this one is very well done & excellent dose of info. on teams across ages. I wud really love to see the ODI analysis of similar nature maybe with span of 5 yrs instead of a decade as the ODI cricket is only 30 years old compared to its "grand version”. I agree with Amit for an analysis by series & I’m sure the numbers would be quite astonishing… for ex. in 1990s & 2000s some of England’s wins have come in dead rubbers(Ashes prominently) & still these are taken into account for PIdx%. I'm aware that TEST WIN is a test win no matter how it is achieved… but fact that some teams have chosen to rest their players after securing a series win puts question on their strength in a ‘dead rubber’ Last bit, do not be deterred by some posts that have been posted. It actually shows great character that you publish them. There are a lot more who like you blogs & are always waiting for another one from you :-) [[ Nishant I can assure you that the Test Series analysis which I have scheduled after the Test All-rounders analysis will be worth the wait. It has been a great eye-opener for me because of the myriads of colours a Test Series brings as against a single Test. Ananth: ]]

• fanedlive on October 1, 2010, 18:47 GMT

Alex, Yes, kalicharran, Lloyd, Grenidge and Richards did do quite well against Indian spinner, though Warne and Mcgrath as a pair would be a more potent than the slightly overated (IMHO!) Indian spinners. You missed that Bradman lost the Bodyline series as well in 34 (and a little fortunate to walk away with a drawn series in 38, thanks to Bradmans brilliance). BTW: Bradman overall record series record (W-L-D) is 8-2-1. Adam Gilchrist's is: 26-2-2.

• fanedlive on September 30, 2010, 6:08 GMT

Amit - a fair evaluation of a bygone era is always tough. However, I like to think that since Viv, Lloyd & Kallicharan did very well vs Chandra-Bedi-Prasanna in India and in WI, they would have fared well vs Warne anywhere. I think the WI peaked in bowling _and_ batting '80-'81 and it might have been the best team of them all.

Also, I don't think Bradman ever lost a series barring his first vs Eng in '28. So, it's not as if only the '48 team was special. Aussies became mortal only after he retired.

• fanedlive on September 30, 2010, 5:34 GMT

Amit - I feel the word "team" applies across a series. Then, Aus '21, Eng '11, and Aus '48 qualify with Aus '48 probably the all-time best: the sole weakness of Aus '48 is in the spin bowling department. They lack a 90mph bowler but that is not a major weakness. Their Barnes was a troubled individual but prolific when he put the pads on. Also, their fielding must have been top-notch.

As for the "poor" figures: Harvey declined in his later half and Morris was a terrific talent, rated by Bradman himself to be the best left-handed opener he ever saw. These two (circa '48), Bradman, Barnes, Miller, and Lindwall would probably have fit in any Aussie side of 90's and 00's.

Also, Viv's WI managed 1-1 vs Imran's Pak in Pak in late 80's courtesy Ambrose & Marshall. I think the WI ('80-'81 or '84-'85) were beatable on certain pitches but would have done better than anyone (maybe even Aus '48) in a 7-test series featuring unbiased playing conditions.

• fanedlive on September 30, 2010, 3:13 GMT

Ananth, it seems unfair to compare the 1948 Aus Invincible to the WI team in 80 that didn't lose a series for 15 years and Aus of 2000, that won about 80% of all matches they play and most of them by a massive margin over a 10 year period. I am curious how long did their 1948 domination last? Also, Statistically, apart from Bradman and to a lesser extent Miller, the remaining nine are inferior (or at best par) with the Aus of 2000. As a kid, I kept hearing of great old-timer (from Hassett, Ponsford and Harveys), but none averaged better than modern great (And todays easy batting condition cant be the justification, as the MacGraths, Warnes, Gillespies averaged no worse than 1948 bowlers). My point being: Are we inherently biased toward recognizing greatness of the past? Alex, I am not quite so sure about WIs skills play spin (by their standard). They didn't play Pak between 1980-1986. They had enough trouble with Hirwani, Qadir and Holland. Warne would be handful for them, IMO. [[ Amit I had done a Team Strength analysis earlier. It is possibly time to re-visit that. I have also made a similar response to a reader comment on the all-time XI. Incidentally Barnes is 63 (albeit 1000 runs). Harvey, McCabe and Ponsford 48. Hassett, Morris, Woodfull and Brown are all 46+. Not exactly poor figures. Ananth: ]]

• fanedlive on September 29, 2010, 16:13 GMT

Hi Anantha

Find your articles and number crunching really insightful.

I was wondering (since cricinfo doing an all time 11 special) whether your could do an all time 11 based solely on numbers. Your team selection would depend on a perfect team make-up i.e. 2 openers 3 middle order 2 all rounders 1 keeper and 3 bowlers (1 spinner). The best in each category (using your previous work) should be chosen and subjective positions such as keeper chosen solely on batting stats.

Please look into this as im sure we all looking forward to this.

Thanks [[ First Your name and mailid would be very useful. The problem with number-based selection is the criteria. It will open a lot of unnecessary debates. What I can do is to form elevens across ages and determine the team strength. Or apply the team strength analysis to the elevens selected by the selectors. That will give an insight into who got it really right. Ananth: ]]

• fanedlive on September 29, 2010, 14:36 GMT

Ok. Thanks for the explanation. Understood.

• fanedlive on September 29, 2010, 6:22 GMT

Amit & Ananth: Benaud was not a playing member of the '48 side in tests. In mid-80's, Lloyd's team was often compared to Armstrong's '21 (Aus), '11 England, and Bradman's '48. Among the old-timers, the consensus was that it should be rated #2 behind Bradman's '48.

- 1 through 11, Bradman's team was stacked with genuine talent and, besides a great Miller, had a decent all-rounder in Lindwall whereas Lloyd's had only Gomes (whose all-round skills roughly cancel out Lindwall's).

- Also, the #6 position in Lloyd's team was always suspect after Kallicharan's sad departure - the unit of Richardson, Lloyd & Gomes played barely 1 year together: so, I think WI was at its strongest in '80-'81 and '84-'85. The WI were never the same after Lloyd retired.

I think the '80-'81 or '84-'85 WI would have done well vs any Aussie team of 90's or 00's on any track. The reason is that the lesser WI teams in late 80's went 1-1 vs Imran's tough Pak teams _in_ Pak on spinning tracks.

• No featured comments at the moment.