THE CORDON HOME

BLOGS ARCHIVES
SELECT BLOG
August 9, 2014

Peer analysis of Test teams: Part 1

Anantha Narayanan
The Australian team of the early 2000s is possibly the strongest Test team of all time  © Getty Images
Enlarge

I have tackled the peer analysis of players in various forms during the years. However, I have never done a peer analysis of players within a group, i.e. teams.

Since cricket is primarily a team game and the players are there to contribute their bit to help the team to achieve the desired results (at least let us think it is that way), the team peer analysis is overdue. This pair of articles will redress the lacuna.

My first thoughts were to do a single article covering all aspects of team peer analysis. Then I realised that the article would be too long and it would be difficult for readers to assimilate all the information. Hence I have split the analysis into two parts.

The first one will look into the team comparisons using players, batsmen and bowlers, as the basis. In the second, I will compare the team performances using the Team performance index (which is a contribution index developed jointly by Milind and me, and which was covered in detail in the article on Test series) and the results, tweaked with properly derived home-away weights.

To handle this analysis I have split the 137 years of Test cricket into nine periods, not necessary of equal duration, but logical and with a reasonably equal distribution of Tests.

The periods are 1877-1914 (Pre WW1), 1920-39 (WW1-WW2), 1946-59 (the post-war years), 1960-69, 1970-79, 1980-89, 1990-99, 2000-06 and finally the current period, 2007-14. This is as logical a split as I can possibly arrive at.

I am sure some readers will have good reasons for fixing 1952-64 or 1984-1997 or something similar as the periods and support such propositions with valid ideas. But let us all agree that this is a logical grouping and move on.

There are no cut-off levels, no minimum requirements, no restrictions of any other kind.

All the 2132 Tests, including the recent Ageas Bowl non-contest, are included.

The table below is a support table to help interpret the following ones.

It summarises the Tests played by each team in the stated time periods.

Each table entry indicates the total number of Tests played, the home Tests played and the away Tests played. For this purpose a neutral location is strictly taken as an away Test for both teams.

This is fair and does not invoke any assumptions. This is the reason why in some of the time periods, the home Test count is different to the away Test count.

During the first period, the three Tests played in England during 1912 between Australia and South Africa cause the 131-137 split. The next neutral Test was only played during the 1990-99 period.

Between 2000 and 2006 five neutral Tests were played, and during the past seven years, 15 more, mostly by Pakistan. Thus there are in total 24 neutral Tests.

We seem to have a problem during 2000-2005. The difference between the home and away Tests is odd, which is illogical.

This is because Test #1768, played between Australia and ICC XI, is accounted only once: as a home Test for Australia. This also explains why the totals in the last column are 4263 and 2155 and not 4264 and 2156.

Even this mundane table is an interesting one in that it is a kaleidoscope of Test cricket as it unfolded. The loss of a decade and half for the South Africans, the recent virtual disappearance of Zimbabwe, the wide disparity in home and away Tests for Pakistan during the last period, the fact that England play at home more than away barring the first period when they travelled to South Africa quite often, and so on.

Number of Tests played
1877-1914 1920-1939 1946-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2006 2007-2014All Tests
All-H-A All-H-A All-H-A All-H-A All-H-A All-H-A All-H-A All-H-A All-H-A All-H-A
Australia105-57-4867-35-3275-35-4067-30-3783-44-3997-54-43108-56-5284-43-4181-40-41767-394-373
England123-48-75120-58-62115-64-51100-53-4795-47-48104-57-47107-57-5092-49-4394-52-42950-485-465
South Africa40-26-1450-28-2247-25-2231-15-164-4-066-36-3078-38-4070-36-34386-208-178
West Indies22-8-1457-24-3349-20-2963-34-2982-30-5281-41-4082-39-4362-30-32498-226-272
New Zealand14-8-638-16-2243-19-2441-21-2059-28-3181-40-4156-29-2762-30-32394-191-203
India7-3-457-30-2752-36-1664-34-3081-42-3969-30-3972-32-4079-37-42481-244-237
Pakistan29-15-1430-13-1746-14-3280-43-3776-34-4266-28-3853-4-49380-151-229
Sri Lanka29-12-1767-30-3771-39-3264-34-30231-115-116
Zimbabwe39-22-1744-22-2210-7-393-51-42
Bangladesh44-21-2339-22-1783-43-40
Total268-131-137280-140-140418-209-209372-186-186396-198-198532-266-266694-346-348689-340-349614-292-3224263-2108-2155

Now let us see the Batting tables.

Batting: All matches
Bat-All 1877-1914 1920-1939 1946-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2006 2007-2014
Australia24.62(103%)36.50(116%)34.30(118%)34.45(109%)31.77(97%)33.94(105%)35.26(114%)43.71(135%)37.31(109%)
England25.31(109%)36.33(120%)31.38(107%)36.13(117%)32.28(98%)29.87(90%)30.29(95%)34.23(102%)36.92(108%)
South Africa20.21(81%)27.99(83%)27.34(90%)33.01(103%)40.36(124%)33.72(108%)37.73(114%)39.90(117%)
West Indies24.39(73%)36.16(125%)34.96(110%)36.49(114%)35.00(109%)29.85(94%)29.74(87%)30.34(86%)
New Zealand25.55(77%)21.25(69%)24.14(72%)27.93(84%)30.01(91%)29.68(93%)32.94(98%)29.88(85%)
India22.95(70%)27.13(89%)29.50(90%)31.91(97%)34.41(107%)35.24(113%)36.22(109%)38.13(112%)
Pakistan25.50(84%)28.67(88%)34.88(108%)35.64(111%)30.93(98%)35.35(106%)30.62(88%)
Sri Lanka25.89(78%)30.71(97%)34.36(103%)38.91(114%)
Zimbabwe26.62(84%)26.44(77%)24.90(72%)
Bangladesh20.78(60%)26.52(75%)
Total24.2032.6929.9532.2232.6632.5431.5633.6134.59

This table covers all Tests. The batting measure is simple and straightforward. It is really the Runs per Wicket value (RpW) with all runs and all wickets included. Extras are runs for teams and run-outs are dismissals by the bowling teams. This table compares the RpW for the concerned team with the total RpW value for all teams, excluding the concerned team. A percentage value above 100 indicates that the team has done very well. A percentage value below 100 indicates that the team has performed worse. Values higher than around 120% are highlighted in blue. Values below 75% are highlighted in red.

Only four teams during the 137-year long Test scene have a peer RpW ratio of greater than 120%. These are given below.

- Australia 2000-06 (135%). No surprise. Considered by many to be the greatest team ever.
- West Indies 1946-59 (125%). A surprise. Possibly the presence of the W's and the young giant Sobers helped. Also a low sub-30 batting RpW value for the rest, with bowlers ruling the roost.
- South Africa 1970-79 (124%). But only in four home Tests. So this can be ignored for all practical purposes. Probably more relevant is South Africa during 2007-14. South Africa scored at 117% on a high base of 34.6.
- England 1920-39 (120%). A top batting line-up, led by Hammond.

Now for the poor performers: The red lined entries.

- Bangladesh 2000-06 had the worst ratio - 60%. They were lambs to the slaughter. They improved slightly and finished the next (and current) period at 72%. Overall they have a ratio of around 65%.
- New Zealand, during the first two post-war periods, were very poor. Their ratios were 69% and 72%. Subsequently they have improved and are around the 90% mark now. It must be conceded that their home pitches were totally bowling-friendly ones.
- West Indies had a ratio of 73% when they started.
- Similarly India had a poor ratio of 70% during their first few years, albeit over seven Tests only.
- During the last period, Zimbabwe has been at around the 72% mark, but over ten Tests.

Now let us move on to the home and away performances of the teams.

Batting: Home matches
Bat-Home 1877-1914 1920-1939 1946-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2006 2007-2014
Australia26.85(116%)34.78(103%)34.40(112%)38.15(119%)32.10(93%)33.98(99%)37.23(113%)48.84(141%)40.95(111%)
England25.49(104%)39.70(127%)34.59(115%)33.72(102%)33.64(99%)29.74(84%)31.92(94%)37.50(105%)39.53(106%)
South Africa19.57(74%)28.52(80%)25.12(78%)35.86(109%)40.36(120%)33.14(98%)39.40(110%)37.34(99%)
West Indies32.81(96%)40.34(134%)36.96(113%)36.98(111%)40.03(120%)31.86(94%)33.20(91%)30.41(79%)
New Zealand26.27(76%)17.69(54%)23.74(69%) 27.30(79%)33.47(98%)32.50(96%)31.66(87%)34.82(92%)
India22.58(66%) 31.85(102%)31.25(93%)32.52(95%)36.55(108%)39.02(118%)36.12(100%)45.33(124%)
Pakistan24.85(78%)31.93(96%)46.90(142%)39.44(118%)33.08(98%)39.80(111%)47.59(127%)
Sri Lanka25.70(74%) 34.88(104%)39.52(111%)41.49(112%)
Zimbabwe29.10(86%)27.35(74%) 29.95(79%)
Bangladesh21.56(58%) 28.72(75%)
Total24.8034.0631.3033.1833.8534.1333.6536.1137.60

England 1920-29 (Wally Hammond), West Indies 1946-59, Australia 2000-06, Pakistan 1970-79 (Javed Miandad/Zaheer Abbas) and Pakistan in the few Tests played recently were the teams that exceeded 125% at home. Pakistan were way off on the bowling front, however.

New Zealand in the fifties and sixties, India in their initial matches, Sri Lanka in their first few matches, Zimbabwe and Bangladesh in 2000-06 went below 75%.

Batting: Away matches
Bat-Away 1877-1914 1920-1939 1946-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2006 2007-2014
Australia21.70(88%)38.92(131%)34.21(124%)31.22(100%)31.38(99%)33.89(111%)33.24(115%)39.08(128%)34.26(107%)
England25.21(117%)33.94(116%)28.12(97%)39.11(134%)31.05(98%)30.03(96%)28.44(95%)30.96(98%)33.94(106%)
South Africa21.40(90%)27.25(85%)29.74(104%)30.33(97%)34.47(118%)36.36(118%)42.56(136%)
West Indies20.13(61%)33.17(119%)33.72(110%)35.94(117%)32.50(106%)27.97(94%)26.99(84%)30.28(93%)
New Zealand24.78(78%)23.88(81%)24.46(75%)28.55(90%)27.43(87%)27.17(91%)34.29(110%)25.99(79%)
India23.24(73%) 22.87(77%)26.24(82%)31.26(99%)32.35(105%)32.61(111%)36.30(118%)33.25(104%)
Pakistan26.14(90%)26.59(84%)30.87(98%)31.84(103%)29.45(99%)32.18(103%)29.78(91%)
Sri Lanka26.01(83%)28.01(94%)29.26(93%)36.66(115%)
Zimbabwe23.77(79%)25.53(80%)13.80(42%)
Bangladesh20.07(62%)23.87(73%)
Total23.6231.4028.7131.2731.5331.0529.6331.4032.21

For these tables I will only highlight the exceptions. First the blue-lined teams.

Australia 1920-39 (Bradman in England!), Australia 2000-06 and South Africa 2007-2014 were the three teams that exceeded 125% in away matches.

Bangladesh 2000-06 and 2007-14 and New Zealand 1920-39 went below 75% away.

Bowling: All matches
Bow-All 1877-1914 1920-1939 1946-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2006 2007-2014
Australia25.35(108%)30.90(93%)25.18(81%)32.40(101%)31.40(95%)33.61(104%)27.97(87%)27.35(79%)31.40(89%)
England22.09(85%)30.74(90%)27.53(89%)30.68(94%)29.39(87%)36.09(114%)35.03(113%)33.37(99%)32.64(94%)
South Africa28.34(120%)37.06(116%)32.19(108%)30.55(94%)22.19(68%)26.95(84%)32.11(95%)29.53(83%)
West Indies34.74(107%)30.74(103%)32.38(101%)34.51(107%)25.71(75%)29.24(92%)36.75(111%)38.11(111%)
New Zealand44.79(139%)38.44(131%)32.28(100%)39.24(123%)30.50(93%)35.39(114%)33.66(100%)35.16(102%)
India39.14(120%)36.77(127%)33.75(105%)33.49(103%)36.51(114%)33.84(108%)35.02(105%)36.51(106%)
Pakistan30.10(101%)37.01(116%)34.58(107%)31.26(95%)29.13(91%)33.24(99%)33.62(97%)
Sri Lanka39.33(122%)35.40(113%)29.00(85%)36.55(106%)
Zimbabwe35.34(113%)42.84(129%)32.63(94%)
Bangladesh51.74(158%)49.50(146%)
Total24.2032.6929.9532.2232.6632.5431.5633.5934.59

I have used the same formula to derive the ratio for bowling. The reason for this will be seen later. This method also maintains consistency. This means that the high values indicate a weaker bowling side and the lower values indicate bowling strength. The colour tagging is reversed so that the blue continues to represent great bowling sides.

Let us now see the outliers. First the high-flying bowling sides. Let us not forget that these are figures for all matches: home and away combined.

- West Indies 1980-89 (75%). Why gild the lily? Unarguably the greatest bowling attack ever. Colin Croft could not find a regular place. That single statement explains everything.
- Australia 2000-06 (79%) was an outstanding bowling side. Glenn McGrath, Shane Warne, Brett Lee and Jason Gillespie formed potent bowling attacks everywhere.
- South Africa 1970-79 (68%). But only in four home Tests. So this can be ignored for all practical purposes. Probably more relevant is South Africa during 2007-14, bowling at 83%. Also South Africa during 1990-99, during which period they bowled at 84%.
- Australia 1946-59 (81%). Spearheaded by Ray Lindwall, and having Keith Miller, Bill Johnston and an emerging Richie Benaud, this was a wonderful attack everywhere.

There are only two really poor bowling sides.

- Bangladesh 2000-06 had the worst ratio of 158%. They were there waiting to be taken apart. They improved slightly and finished the next (and current) period at 146%. Overall, they have a very poor ratio of around 150%.
- New Zealand, during the first two post-war periods, were very poor. Their ratios were 139% and 131%. Subsequent improvements have lifted them to the present 100% mark.
- India, during the fifties at 127%, and Zimbabwe, during the first half of the 2000s at 129%, were quite sub-par.

Bowling: Home matches
Bow-Home 1877-1914 1920-1939 1946-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2006 2007-2014
Australia25.40(116%)29.18(90%)24.71(83%)34.17(111%)30.02(94%)31.90(104%)26.05(86%)27.36(85%)29.18(89%)
England19.67(77%) 29.76(91%)25.13(82%)27.18(82%)29.57(92%)35.11(117%)35.87(126%)31.62(101%)30.42(93%)
South Africa26.05(114%)35.67(117%)31.41(111%)30.87(99%)22.19(70%) 25.38(84%)28.20(88%)26.71(81%)
West Indies29.79(95%)35.53(128%)32.78(105%)35.77(117%)25.99(82%)27.71(93%)34.57(111%)34.00(106%)
New Zealand46.06(149%)34.94(123%)32.80(105%)37.08(120%)28.39(91%)34.17(118%)31.25(99%)33.62(105%)
India37.10(119%)36.01(130%)33.26(108%)29.17(91%)33.50(109%)26.48(89%)33.35(107%)33.25(104%)
Pakistan23.28(80%) 33.06(106%)35.40(113%)28.66(91%)27.41(92%)32.32(103%)52.26(163%)
Sri Lanka30.87(99%)32.25(110%)26.27(82%) 30.66(95%)
Zimbabwe33.19(113%)40.40(130%)29.95(93%)
Bangladesh45.26(147%)52.59(169%)
Total23.4731.4028.7131.2731.5331.0529.6031.4432.18

Australia 2000-06 and 1946-59 were outstanding at home. South Africa had three periods of glory, including the most recent one. West Indies, during the eighties, were invincible at home as a bowling unit. England and Pakistan, during their respective first periods, and Sri Lanka, during the first few years of the millennium, were very potent at home in the bowling department.

New Zealand, after the war, and Bangladesh, in their first 14 years, have been quite poor bowling teams, even at home.

A brief explanation on Pakistan's outrageous 163% in their last four home Tests. Unfortunately their bowlers were bush-whacked by Sri Lankan and South African batsmen to the tune of 450, 305 for 4, 644 for 7 and 606. However, Pakistan also responded well.

Bowling: Away matches
Bow-Away) 1877-1914 1920-1939 1946-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2006 2007-2014
Australia25.28(102%)33.15(97%)25.62(78%)30.93(92%)33.17(98%)36.28(107%)30.29(88%)27.33(73%) 33.80(90%)
England23.52(87%)31.57(86%)31.01(99%)35.00(107%)29.22(82%) 37.34(111%)34.09(102%)35.58(99%)35.75(96%)
South Africa32.94(137%)39.11(117%)33.05(106%)30.26(90%)28.91(85%)36.29(101%)33.07(88%)
West Indies39.29(117%)27.40(85%)32.12(96%)33.10(97%)25.56(69%) 30.83(91%)38.90(110%)43.12(118%)
New Zealand43.17(128%)41.02(134%)31.88(96%)41.71(126%)32.31(94%)36.73(110%)36.49(102%)36.87(100%)
India40.91(121%)37.75(124%)34.79(105%)39.06(118%)39.99(120%)40.68(124%)36.45(102%)39.77(108%)
Pakistan39.84(129%)40.78(125%)34.21(101%)34.61(102%)30.52(89%)33.96(94%)32.39(85%)
Sri Lanka46.52(139%)38.30(115%)32.79(91%)45.46(125%)
Zimbabwe38.64(116%)45.55(129%)45.96(124%)
Bangladesh59.76(171%)45.55(124%)
Total24.9134.0631.3033.1833.8534.1333.6535.8837.08

West Indies 1980-89, Australia 2000-06 and England 1970-79 were outstanding in away matches. West Indies, with a figure of 69% in away matches, should surely be classified as the best bowling side.

Quite a few teams have been powerless away. South Africa, New Zealand, Sri Lanka and Bangladesh, when they started, have all been toothless tigers away from home.

Differential Index: Without creating another table and adding to the profusion of tables, I created an interesting combination value. I subtracted the bowling ratio from the batting ratio for all matches and the resulting differential ratio is an indicator of the overall strength of the teams. A positive number indicates a strong side and a negative number a weaker side. First, let us look at teams that had the combined ratio greater than 30%.

Only one team, for all practical purposes, had both ratios in blue. This also indicates that it is not easy to do this. Australia 2000-06 had a combined ratio of +56% (141%-72%). This single factor indicates that this was the strongest team of all times. They had no weakness, other than the odd loss to India.
One other team, South Africa 1970-79, had blue in both leading to a combined ratio of +56% (124%-68%). But this has to be discounted because South Africa played four home matches during this period. But this gives us an idea of the potential greatness of this wonderful team unfortunately destined to lose the next 18 years through their own abhorrent policies.
Australia 1946-59 had a combined ratio of +37% (118%-81%) and could be considered to be the second-best team of all time. Their batting was very good and bowling was equally good.
West Indies 1980-89 had a combined ratio of +34% (109%-75%) and could also lay claim to being the second-best team of all time. Their batting was good, but not great. Their bowling was outstanding. My gut feeling is that they would finish with better results record than the two Australian teams.
South Africa 2007-14 had a combined ratio of +34% (117%-83%). Their batting was very good and bowling was equally good.

The best home differential is for Australia 2000-06 with 56% (141%-85%). The best away differential is for Australia 2000-06 with 55% (128%-73%). They have been equally devastating, home and away.

I have done a visual comparison for these numbers. If any reader locates any other team with the difference greater than 30% please bring it to our notice.

Only two teams qualify for the plastic spoon (wood is expensive nowadays). Bangladesh 2000-06 had an abysmal difference of -98% (60%-158%) and Bangladesh 2006-14 had a slightly better number of -71% (75%-146%). New Zealand 1920-39 and 1946-59 had low difference ratio of -62%.

The worst home differential is for Bangladesh 2000-06 with -89% (58%-147%) and the worst away differential is for Bangladesh 2000-06 with -109% (62%-171%). No further statements are needed.

There is no doubt that Australia 2000-06 is the best team ever. However let me reserve judgement on this until we have a look at the next part, which is more team-oriented and will cover performance and results. It is possible that West Indies 1980-89 may have a say there.

One interim conclusion, pending a final one after the second part, is that both Bangladesh 2000-14 and New Zealand 1920-59 have been poor teams. The only difference seems to be that New Zealand came out of this because they were overall very professional and the future stars at that time, Martin Crowe, Richard Hadlee and Glenn Turner, and Stephen Fleming, Ross Taylor and Daniel Vettori, later, were great role models and team players. The best player of Bangladesh is a prima donna with easy T20 league money as the lure, and is poorly handled by BCB. The future seems dicey. What are the odds on Shakib Al Hasan doing a Ryan ten Doeschate?

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   on (August 18, 2014, 4:24 GMT)

Anantha, I'd really like for you to do an analysis of the best pace bowlers of all time. I did a partial analysis using the following criteria: Test >40, Wickets >200, Average <26, Strike rate <58. It narrowed it down to 17 pace bowlers: Ambrose, Donald, Garner, Hadlee, Holding, Imran Khan, Lillee, Marshall, McGrath, Pollock, Roberts, Steyn, Trueman, Walsh, Wasim, Bob Willis, Younis. I did a ranking system from 1-17 for: WpI (wicket per innings), WpM (wicket per match), average, strike rate and economy for each bowler. I looked at the average ranking with and without economy as a factor. With economy factor the overall ranking I got was: 1. Marshall, 2. Hadlee, 3. Steyn, 3. Garner, 5. Trueman, 6. Donald, 7. McGrath, 8. Lillee, 9. Ambrose, 10. Imran Khan, 11. Younis, 12. Wasim, 13. Holding, 14. Pollock, 15. Roberts, 15. Walsh, 17. Willis. Without the economy factor, the overall ranking order changed slightly. What didn't change was who were in the top 6 or bottom 6. Looking forward!
[[
Will look at that but I have done some bowler analysis earlier. It was for all bowlers and Murali came on top. My feeling is that with 25% of Steyn's career ahead of us, maybe we should wait for some time.
Ananth
]]

Posted by Nayan-Jyoti on (August 12, 2014, 11:45 GMT)

@Ananth Sir, I think there is a small mistake in the article. After the last table which is for Bowling : Away Matches, look at the 3rd sentence of the 4th paragraph which is "Australia 2000-06 had a combined ratio of +56% (141%-72%)". But 141-72 equals 69, not 56. Actually you have written the differential index for Australia in 2000-06 correctly which is +56. But you miswrote "141%-72%". It should have been Batting in all matches minus Bowling in all matches which would give "135%-79%=+56". But you wrote Batting in home matches minus Bowling in away matches. Am I right or am I missing something. Sorry if I am wrong.
[[
Nayan,
Antony Purcell has already raised this matter and I have replied to him.
Australian net should read +56% (135%-79%). The final difference was correct but the component numbers were not. Thanks for being so alert.
Ananth
]]

Comments have now been closed for this article

ABOUT THE AUTHOR

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