Statistics April 13, 2013

Analysing Test matches across eras - part I

A comprehensive analysis of all Tests played, looking at first-innings advantage, result likelihood during different eras, margin of victories, and the best Tests ever played

Ian Botham was one of the protagonists in England's famous win against Australia at Headingley in 1981 © Getty Images

Milind and I are working on a mega-project on player contributions in Tests. Analysing Test results in depth was an essential part of that exercise where we unearthed some fascinating facts. This made me think that I could do an article analysing Test results. Initially I was apprehensive that it may not have enough material to fill my normal long article. Then things developed as I delved deeper and now I am left, as it often happens, with a two-part article. These two articles are partly anecdotal. The first one will deal with those matches that finished with a definite result. The second one will deal with draws, ties, follow-ons and the like.

This article is as much form as content. This is not the usual table-centric article. I have to present the analysis data in different forms. As I normally do, let me say that most of the data presented here will be available through Cricinfo's stats and some further work. Only thing is you might need quite a few queries and results will not be available in this clear, concise format.

First, let me talk about two special matches that ended with a definite result. Both these have been considered as wins for this exercise.

The first one was Match #1483. There was a lot of rain during the first four days at Centurion and South Africa had scored 248 for 8 in the 72 overs possible. South Africa could have batted on, and the result would have been a dull draw. However Nasser Hussain and Hansie Cronje agreed to forfeit their respective first and second innings and make a match of it. England had 90 overs to score 250 for a win. They did this in 75 overs and won the match by 2 wickets. Many players, commentators and writers have argued for and against this "arrangement". The added spice was the presence of Cronje as one of the protagonists and the subsequent match-fixing allegations. However, the presence of the impeccable English captain, Nasser Hussain, in the twosome should dispel any doubts on this match. I have only one wish. Instead of forfeiting the two innings, the two captains could have played an over each before declaration, making it a complete match. It would have avoided many programming headaches for me. More significantly, no one could have questioned the captains. Anyhow as far as I am concerned this is a match which ended in a 2-wicket win/loss.

The second is a serious "non-win" which became a win. Move the clock forward by six years, to Match #1814. England played poorly and were dismissed for 173 by Pakistan. Pakistan amassed a lead of over 300 runs and England took the field on the third day, in great distress. They reached 298 for 4, still 33 runs in arrears. Darrel Hair accused Pakistan of ball-tampering and awarded 5 penalty runs to England. Pakistan did not take the field and the match was awarded to England as a forfeited match. This was revised to a draw in 2008 and was reversed back to forfeit after another year. A sorry episode all through. Even now, I do not have any clarity in determining how to treat this match. A win no doubt, but not by an innings, not by runs, and not by wickets. For that reason I do not want to create a set of columns called "forfeited". So it remains an undefined win. It does not come under any of the three categories. It is an "un-earned" win. The two concerned teams have been allotted the result, though.

First, let me say that no team should ever forfeit a match, whatever be the provocation. But Pakistan had reason to feel aggrieved, not that it justifies Inzamam-ul-Haq's action. Let us not forget that the first forfeit almost happened 25 years earlier when Sunil Gavaskar, given out lbw, and more likely because of unnecessary and provocative on-field comments from the Australians, almost walked off the field at MCG during 1981. But for the timely action of the manager, Wg CmdrSalim Durrani, in pushing Dilip Vengsarkar onto the field and ensuring that Chetan Chauhan did not cross the line, unpleasant history would have been made then and there. We were 5 seconds away from the first forfeiture of a Test match.

A. Overall summary of results - by period
All Tests 2085324.6136565.5 69033.1 67532.4 72034.5
1877 - 1948 307307.8 21770.7 12540.7 9230.0 9029.3
1949 - 1979 560320.8 32858.6 18432.9 14425.7 23241.4
1980 - 1999 613316.7 36659.7 17328.2 19331.5 24740.3
2000 - 2013 605344.6 45475.0 20834.4 24640.7 15125.0

The first table sets the tone. We start sedately with a straightforward table of results across ages and then increase our pace.

In 135 years of Test cricket, 2085 matches have been played. There have been 1365 results, which is just over 65% of the matches. That is just under 2 out of 3. For the purpose of this article the two tied matches, have been categorized as draws. Teams batting first have won 690 matches, 33.1%, and the teams batting second, 675, around 32.4%. Over the long period, teams batting first have an infinitesimal edge of less than 1%. Take a look at the average of first-innings score of teams batting first, which stands at 324.6.

I have then analysed the results in four broad periods. My usual period work is done by creating 8 periods. Here, I felt that we could consider the two modern periods, one middle period and the first one, covering just over 70 years.

During the first period, upto 1948, the result percentage was higher than 70%, partly helped by timeless Tests. The average first-innings score was lower at 307.8, but the number of matches won by teams batting first was quite high, at over 40%. teams batting second won only 30% of the matches, a clear difference of 25%. Pitches deteriorated faster during this period, owing to them being uncovered.

The period between 1949 and 1979 is perceived to be the dull period when no risks were taken and draws were the first option for some teams. This is shown by the increase of average first-innings score to 320 and the huge drop of over 10% in result matches, to around 59%. Look at the draw percentage, which is over 41%. Teams batting first won more matches comfortably.

Now we arrive at the 1980s and 90s. Despite the presence of very formidable bowling attacks, this period does not even boast a 60% result rate. Probably it was difficult to shrug off the safety-first attitude. The average first-innings score dropped. For the first time more teams won batting second: 31.5% against 28.2%. Fairly high number of draws. So, contrary to the usual perception, the 1980-1999 was not that great a Test period. It is clear that the West Indian dominance was offset by mediocrity elsewhere.

Finally we reach the current era: a truly great Test playing period, despite the proliferation of ODIs and T20 matches. A 75% results in matches is truly phenomenal. That means a 4-Test series is likely to end in a 3-0 or a 2-1 result. Look at the 7% increase in the average first-innings score; it is now around 340-350. And a very significant increase in the wins by teams batting second, despite this increase. No longer can the toss-winning captain play the percentage-game of batting first. You are more likely to lose than not.

B. A look at first-innings scores

First Inns scores>=300   Tests:1070 Wins: 435 Draws: 488 Losses: 147 Perf: 63.5%
First Inns scores>=324   Tests: 958 Wins: 412 Draws: 435 Losses: 111 Perf: 65.7%
First Inns scores>=350   Tests: 832 Wins: 364 Draws: 385 Losses:  83 Perf: 66.9%
First Inns scores>=400   Tests: 607 Wins: 285 Draws: 283 Losses:  39 Perf: 70.3%

This is a special look at first-innings scores. As can be seen, the all-time average first-innings score is 324. I did a special analysis of all matches in which this mean level was reached or exceeded. There were 958 Tests and the teams batting first won 412 of these matches. The overall performance percentage was 65.7, which denotes a fair achievement level. Then I increased the cut-off values and found out that at around 394 runs, the 70% figure is reached. This represents a very good achievement. Rounding off this to 400, we get an excellent performance value of 70.3%. My take is that any target between 350 and 400 can be taken, with the stronger teams opting for the latter figure. The current acceptable level is certainly 350+. We have decided to set the target for all first innings as 400 in our contribution work. Just to get a clear overall picture, I have done this exercise for 300 and 350 runs also. The performance percentage is 63.5% and 66.9%, respectively. As the cut-off runs are reduced, the wins decrease but the number of draws increases significantly. One other interesting fact emerged as I pushed the bar further up. The magic figure of 50% wins is reached at a cut-off score of 594. Out of the 75 instances when this total was reached, 38 ended in wins and 37 in draws.

C. A summary of innings wins

Innings wins:363   Total=34938     Avge=96.2
  Wins by inns &  1-9    runs   23   6.3
  Wins by inns & 10-49   runs  101  27.8
  Wins by inns & 50-99   runs  102  28.1
  Wins by inns & 100-199 runs   96  26.4
  Wins by inns & 200+    runs   41  11.3

This is a summary of the 363 innings wins. The average shown is an interesting number. It is the average margin of the innings victories. In other words, the sum of win-runs divided by the number of wins. Also, the additional runs scored, forms the slack. The average win has been by an innings and 96 runs, which is a big win. This is explained by the analysis below. Only 6.3% of the wins have been by an innings and upto-10 runs. These are still not close contests. It was only a matter of avoiding innings defeats. It can clearly be seen that more than 66% of the innings wins have been by wide margins. This is reflected in the overall average also. Let us not forget that the winning team also had the 10 wickets in hand, in addition to the run margins.

D. A summary of wins by runs

Runs wins: 464     Total= 73968  Avge=159.4
  Wins by  1-9    runs     9    1.9
  Wins by 10-49   runs    57   12.3
  Wins by 50-99   runs    81   17.5
  Wins by 100-199 runs   175   37.7
  Wins by 200+    runs   142   30.6

This is a summary of the 464 wins by runs. The average is calculated similar to the innings victories. There is a lot of slack in these wins also. The average win has been by 159 runs, which is a reasonably big win. This is explained by the analysis below. Only 1.9% of the wins have been by upto10 runs. However, the big difference here is that all these matches were extremely close and could easily have gone the other way. In a way, even the next category, upto-50 runs, is similar to this. These are narrow wins. Beyond this, the wins become more comfortable. Look at the high %, 30.6, of the last category, 200+ runs, all these being huge wins. It can clearly be seen that more than 68% of the innings wins have been by wide margins. This is reflected in the overall average also.

E. A summary of wins by wickets

Wicket  wins: 537  Total= 3856    Avge=  7.2
  Wins by  1 wkt(s)   12   2.2
  Wins by  2 wkt(s)   17   3.2
  Wins by  3 wkt(s)   22   4.1
  Wins by  4 wkt(s)   32   6.0
  Wins by  5 wkt(s)   48   8.9
  Wins by  6 wkt(s)   47   8.8
  Wins by  7 wkt(s)   74  13.8
  Wins by  8 wkt(s)   97  18.1
  Wins by  9 wkt(s)   80  14.9
  Wins by 10 wkt(s)  108  20.1

First, let me say that these are the only wins in which there is no slack. Always, the required number of runs, i.e. the aggregate runs of the losing team plus the run(s) needed to overtake this aggregate, are scored, since the learned duo of Duckworth-Lewis have not made their appearance into the Test arena. This analysis of wins by wickets is an eye-opener. Most of the wins, 92% to be specific, are by 4 or more wickets, which are quite comfortable wins. These teams had a lot of resources to spare. Only wins by 3/2/1 wickets can be termed as narrow and that total is below 10%. The average win margin of 7.2 wickets reflects this. Also, note the increasing trend of the numbers.

F. Team performance summary
TeamTestsWinsDrawsLossesResult %Close winsClose losses
Australia 754353204199 60.3 410
Bangladesh 77 3 8 66 9.1 0 1
England 933331334268 53.4 5 4
India 472119205149 46.9 1 0
New Zealand 382 72154156 39.0 2 0
Pakistan 373115154104 51.5 2 1
South Africa377137114126 51.5 3 2
Sri Lanka 222 66 76 80 46.8 1 1
West Indies 490160169162 49.9 3 2
Zimbabwe 89 9 26 54 24.7 0 0
ICC World XI 1 0 0 1 0.0 0 0

No surprises here. Australia leads the table with a result percentage exceeding 60, the only team to do so. This is based on 1.0 for a win and 0.5 for a draw basis. But what do we have here? Australia has played 14 close matches (listed elsewhere: sub-10 runs and 1-wkt wins). They have lost 10 of those. That means they are the team, despite the abundance of talent and determination, to lose more than 2 of 3 such matches. A true paradox. And it is also a surprise that they get into these situations far more often than others.

England follows next with 53.4%. They seem to hold their nerves, as shown by their wonderful saves during the past 4 years. Maybe there is something there for the Australians to learn. Pakistan and South Africa, with their special bowling attacks capable of winning everywhere, are next with 51.5%. They do not get into these close situations often. These are the only four teams to have a performance percentage of greater than 50. West Indies fall short of the 50% mark, by a whisker. Look at the high percentage of drawn matches for India, a throw-back to the safety-first methods between 1950 and 1995. India and Sri Lanka are almost at the same level.

G. Team performance - Home/Away/Neutral
TeamTestsWinsDrawsLossesResult %TestsWinsDrawsLossesResult %Tests WinsDrawsLossesResult %
Australia 389219 76 9566.135612712710353.5 9 7 1 183.3
Bangladesh 39 1 5 33 9.0 38 2 3 33 9.2 0 0 0 0 0.0
England 47319117310958.745514015915648.2 5 0 2 320.0
India 242 82110 5156.6229 37 94 9836.7 1 0 1 050.0
New Zealand 186 44 81 6145.4194 28 71 9532.7 2 0 2 050.0
Pakistan 144 55 68 2161.8206 51 77 7843.4 23 9 9 558.7
South Africa203 87 52 6455.7169 50 59 6047.0 5 0 3 230.0
Sri Lanka 113 47 40 2659.3105 19 34 5234.3 4 0 2 225.0
West Indies 223 80 91 5256.3264 80 7810745.1 3 0 0 3 0.0
Zimbabwe 47 7 16 2431.9 42 2 10 3016.7 0 0 0 0 0.0
ICC World XI 0 0 0 0 0.0 1 0 0 1 0.0 0 0 0 0 0.0

Australia has a rather even record everywhere, confirming that they are not just lions at home. They have a result percentage of 66% at home, 53.5% away and 83% in neutral locations: all figures the best in each classification. This makes their recent 0-4 loss an aberration. Pakistan is the only other team with a home result percentage of greater than 61. All other major teams are grouped between 55% and 60%. England is quite good, playing away, with a result percentage of just below 50. South Africa, West Indies and Pakistan follow next. Sri Lanka, India and New Zealand have only average performances away from home. Only Pakistan have played reasonable number of matches in neutral locations and have only a fair result percentage and outside UAE, they have not done well.

H. Location results summary
LocationTestsResultsDrawsResult %
Australia 389314 7680.7%
Bangladesh 47 37 1078.7%
England 47830417463.6%
India 24213311054.9%
New Zealand 186105 8156.5%
Pakistan 144 76 6852.8%
South Africa 203151 5274.4%
Sri Lanka 114 74 4064.9%
West Indies 223132 9159.2%
Zimbabwe 47 31 1666.0%
U.A.E. 12 8 466.7%

This is an analysis by match ground. UAE, a non-Test playing country, has hosted 12 matches. The key measure here is the percentage of matches which ended in a result. Australia leads this table with just over 80% results. That means that a 5-Test series in Australia is likely to end with a single draw. In addition to the type of cricket played by Australia, the pitches with true bounce which provide equal help to batsmen and bowlers alike, would be the main reason. If and when the World Test Championship (WTC) is played, it should be conducted in Australia, Indian money notwithstanding. South Africa follows suit, reasonably close behind with nearly 75% results. Fine, let us have some of the WTC matches there too. Sri Lanka and England are fairly high, at around 65%. West Indies falls just short of 60%. Surprisingly New Zealand manages to produce a result only 56% of the time. India and Pakistan do not even reach 55%, no doubt weighed down by the awful 50s/60s.

Here is a potpourri of interesting result-matches gathered using special criteria. In all these summary score lines, the winning team is shown in upper-case letters.

1. Big Innings wins
    0266 1938 By I&579 runs ENG-903/7 aus-201ao fo aus-123ao
    1590 2002 By I&360 runs AUS-652/7 saf-159ao fo saf-133ao
    0463 1959 By I&336 runs WIN-614/5 ind-124ao fo ind-154ao
    0279 1946 By I&332 runs AUS-645ao eng-141ao fo eng-172ao
    1600 2002 By I&324 runs PAK-643ao nzl- 73ao fo nzl-246ao
    1289 1995 By I&322 runs WIN-660/5 nzl-216ao fo nzl-122ao
    1630 2002 By I&310 runs bng-139ao WIN-536ao    bng- 87ao
    2033 2012 By I&301 runs NZL-495/7 zim- 51ao fo zim-143ao

The Oval disaster for Australia leads the chart. Imagine facing 903 without Don Bradman, who twisted his ankle while bowling. Zimbabwe arrived after a long delay to get slaughtered by New Zealand during 2012.

2. Big Run wins
    0176 1928 By   675 runs ENG-521ao aus-122ao    ENG-342/8 aus- 66ao
    0237 1934 By   562 runs AUS-701ao eng-321ao    AUS-327ao eng-145ao
    0114 1911 By   530 runs AUS-328ao saf-205ao    AUS-578ao saf-171ao
    1726 2004 By   491 runs AUS-381ao pak-179ao    AUS-361/5 pak- 72ao
    1905 2009 By   465 runs SLK-384ao bng-208ao    SLK-447/6 bng-158ao
    0779 1976 By   425 runs WIN-211ao eng- 71ao    WIN-411/5 eng-126ao
    0300 1948 By   409 runs AUS-350ao eng-215ao    AUS-460/7 eng-186ao
    0870 1980 By   408 runs WIN-328ao aus-203ao    WIN-448ao aus-165ao

The first match featured here was Bradman's debut Test. Maybe this Brisbane massacre, followed by him being dropped in the next Test at SCG, must have been partly responsible for Bradman's later run-hunger/thirst.

3. Close Wins - by 1 wkt
    0074 1902 By     1 wkt  aus-324ao ENG-183ao    aus-121ao ENG-263/9
    0088 1906 By     1 wkt  eng-184ao SAF- 91ao    eng-190ao SAF-287/9
    0097 1908 By     1 wkt  aus-266ao ENG-382ao    aus-397ao ENG-282/9
    0149 1923 By     1 wkt  saf-113ao ENG-183ao    saf-242ao ENG-173/9
    0345 1952 By     1 wkt  win-272ao AUS-216ao    win-203ao AUS-260/9
    0873 1980 By     1 wkt  win-140ao NZL-249ao    win-212ao NZL-104/9
    1268 1994 By     1 wkt  aus-337ao PAK-256ao    aus-232ao PAK-315/9
    1453 1999 By     1 wkt  aus-490ao WIN-329ao    aus-146ao WIN-311/9
    1497 2000 By     1 wkt  pak-269ao WIN-273ao    pak-219ao WIN-216/9
    1658 2003 By     1 wkt  bng-281ao PAK-175ao    bng-154ao PAK-262/9
    1812 2006 By     1 wkt  saf-361ao SLK-321ao    saf-311ao SLK-352/9
    1972 2010 By     1 wkt  aus-428ao IND-405ao    aus-192ao IND-216/9

I wanted this section to have no more than 10 Tests in each category but had to accept this classification featuring 12 Tests. The criterion is fixed. The amazing fact is that Australia featured in 6 of these Tests and lost 5. England featured in 3 and won all. One of Australian losses was the Laxman-heist at Mohali during 2010.

4. Close Wins - by fewer than 10 runs
    1505 2000 By     7 runs SAF-253ao slk-308ao    SAF-231ao slk-169ao
    2021 2011 By     7 runs NZL-150ao aus-136ao    NZL-226ao aus-233ao
    0009 1882 By     7 runs AUS- 63ao eng-101ao    AUS-122ao eng- 77ao
    0019 1885 By     6 runs AUS-181ao eng-133ao    AUS-165ao eng-207ao
    1243 1994 By     5 runs SAF-169ao aus-292ao    SAF-239ao aus-111ao
    0943 1982 By     3 runs ENG-284ao aus-287ao    ENG-294ao aus-288ao
    0073 1902 By     3 runs AUS-299ao eng-262ao    AUS- 86ao eng-120ao
    1758 2005 By     2 runs ENG-407ao aus-308ao    ENG-182ao aus-279ao
    1210 1993 By     1 run  WIN-252ao aus-213ao    WIN-146ao aus-184ao

These are the other close matches. Single-digit-run wins should be considered the closest of wins. Australia have featured in 8 of these 9 matches and have only a 3-5 record. They seem to lose their nerve or is there another reason?

5. Lost after scoring 500 runs in first inns
    0042 1894 By    10 runs aus-586ao ENG-325ao fo ENG-437ao aus-166ao
    1673 2003 By     4 wkts aus-556ao IND-523ao    aus-196ao IND-233/6
    1819 2006 By     6 wkts eng-551/6 AUS-513ao    eng-129ao AUS-168/4
    0635 1968 By     7 wkts win-526/7 ENG-404ao    win- 92/2 ENG-215/3
    0365 1953 By     6 wkts aus-520ao SAF-435ao    aus-209ao SAF-297/4
    0180 1929 By     5 wkts eng-519ao AUS-491ao    eng-257ao AUS-287/5

Scoring 500 and losing the match. Very painful. Australia again. They have featured in 5 matches and have only a 2-3 record. One was after making England follow-on and the other was the famous Indian win at Adelaide during 2003. The English win over West Indies was because of the declaration by Garry Sobers, often widely perceived as silly.

6. Lost losing fewer than 15 wkts
    1483 2000 By 2 wkts saf-248/8                      ENG-251/8  8 wkts lost
    0635 1968 By 7 wkts win-526/7 ENG-404ao  win- 92/2 ENG-215/3  9 wkts lost
    1814 2006 Forfeited ENG-173ao pak-504ao  ENG-298/4           10 wkts lost
    0313 1949 By 3 wkts saf-379ao ENG-395ao  saf-187/3 ENG-174/7 13 wkts lost
    1556 2001 By 6 wkts aus-447ao ENG-309ao  aus-176/4 ENG-315/4 14 wkts lost

The first is the match we have discussed earlier. Had to be present here for completion of presentation. The West Indies - England match has also been referred to briefly earlier. West Indies made two declarations, at 7 and 2 wickets, and lost the match. Quixotic, to say the least. The third match also has been discussed extensively earlier. I cannot get a handle on the fourth match. The South African bowling was very ordinary: no bowler even reaching a career aggregate of 60 wickets. Yet they declared leaving England, with Len Hutton, Cyril Washbrook, and Denis Compton, to get a sub-180 target. A strange decision by Dudley Nourse, indeed. But in his favour, it must be said that there was only 96 minutes of play possible. The amazing fact is that South Africa bowled 24 8-ball overs (32 6-ball overs) and captured 7 wickets. So let me equivocate - it is not so strange a decision as I had mentioned earlier. The fourth is the famous dead-rubber loss by Australia, orchestrated by Mark Butcher.

7. Won after scoring 100 runs in first inns
    0025 1887 By    13 runs ENG- 45ao aus-119ao    ENG-184ao aus- 97ao
    0009 1882 By     7 runs AUS- 63ao eng-101ao    AUS-122ao eng- 77ao
    0043 1895 By    94 runs ENG- 75ao aus-123ao    ENG-475ao aus-333ao
    0094 1907 By    53 runs ENG- 76ao saf-110ao    ENG-162ao saf- 75ao
    0059 1899 By   210 runs ENG- 92ao saf-177ao    ENG-330ao saf- 35ao
    2034 2012 By    71 runs PAK- 99ao eng-141ao    PAK-365ao eng-252ao

All but one of these matches were before 1908. Uncovered and deteriorating pitches meant that any runs on board were very valuable. The last match is of current vintage. England's loss was because of their first-innings failure to build a substantial lead after dismissing Pakistan for 99. Then Saeed Ajmal took over. The red herring which made everyone think that they were candidates for a 4-0 drubbing later in the year in India.

8. Won in least overs
    0047 1896 By   288 runs ENG-185ao saf- 93ao    ENG-226ao saf- 30ao (41.2 ov)
    0032 1889 By I&202 runs ENG-292ao saf- 47ao fo saf- 43ao           (50.3 ov)
    0082 1904 By   218 runs AUS-247ao eng- 61ao    AUS-133ao eng-101ao (54.1 ov)
    0216 1932 By I& 72 runs saf- 36ao AUS-153ao    saf- 45ao           (54.3 ov)
    0030 1888 By I& 21 runs ENG-172ao aus- 81ao fo aus- 70ao           (55.5 ov)
    1617 2002 By I&198 runs pak- 59ao AUS-310ao    pak- 53ao           (56.4 ov)
    0353 1952 By I&207 runs ENG-347/9 ind- 58ao fo ind- 82ao           (58.1 ov)

One match finished in fewer than 250 balls. That is a strike rate of a wicket every 12 balls. The umpires must have had sore shoulders. The most recent instance was Pakistan's Sharjah collapse in less than 60 overs for two sub-60 totals.

9. Won losing fewer than 3 wickets
    0154 1924 By I& 18 runs saf-273ao ENG-531/2    saf-240ao   2 wkts lost
    0456 1958 By I& 71 runs nzl- 67ao ENG-267/2    nzl-129ao   2 wkts lost
    0741 1974 By I& 78 runs ind-165ao ENG-459/2    ind-216ao   2 wkts lost
    1640 2003 By I& 60 runs bng-173ao SAF-470/2    bng-237ao   2 wkts lost
    2049 2012 By I& 12 runs eng-385ao SAF-637/2    eng-240ao   2 wkts lost

These are all matches with winning teams declaring 2 wickets down. For the first innings, there are a few matches won losing 3 wickets. The last match is fascinating. The same bowlers who orchestrated 4-0 and 2-1(away) wins against a strong Indian batting line-up managed to capture 2 wickets in 189 overs, in the interim period.

10. Won scoring fewer than 200 runs in match
    0009 1882 By     7 runs AUS- 63ao eng-101ao    AUS-122ao eng- 77ao
    0028 1888 By    61 runs AUS-116ao eng- 53ao    AUS- 60ao eng- 62ao
    0030 1888 By I& 21 runs ENG-172ao aus- 81ao fo aus- 70ao
    0034 1890 By     2 wkts aus- 92ao ENG-100ao    aus-102ao ENG- 95/8
    0128 1912 By    10 wkts saf- 95ao ENG-176ao    saf- 93ao ENG- 14/0
    0216 1932 By I& 72 runs saf- 36ao AUS-153ao    saf- 45ao
    0238 1935 By     4 wkts win-102ao ENG- 81/7    win- 51/6 ENG- 75/6
    0275 1946 By I&103 runs nzl- 42ao AUS-199/8    nzl- 54ao

This is a fascinating collection, mostly filled with pre-WW1 matches. Imagine, the winning team did not even need 200 runs in the two innings combined, to win, thrice with an innings to spare. I am fascinated by the 1888 match when Australia followed on, merely 91 runs behind. Australia's convenient declaration at 199/8 during 1946 got them in.

11. Won trailing by more than 250 runs
    1814 2006 Forfeited   ENG-173ao pak-504ao    ENG-298/4           Deficit 331
    1194 1992 By  16 runs AUS-256ao slk-547/8    AUS-471ao slk-164ao Deficit 291
    1535 2001 By 171 runs aus-445ao IND-171ao fo IND-657/7 aus-212ao Deficit 274
    0042 1894 By  10 runs aus-586ao ENG-325ao fo ENG-437ao aus-166ao Deficit 261
    0320 1950 By   5 wkts saf-311ao AUS- 75ao    saf- 99ao AUS-336/5 Deficit 236
    0905 1981 By  18 runs aus-401/9 ENG-174ao fo ENG-356ao aus-111ao Deficit 227
    1945 2010 By  36 runs AUS-127ao pak-333ao    AUS-381ao pak-139ao Deficit 206

In these matches the winning team trailed by more than 250 runs in the first innings. Three of these matches were won after following on. The fourth is the famous Botham-Willis match. The sixth is the equally famous Laxman-Harbhajan match. The last of these matches happened three years back at SCG. Possibly Michael Hussey's best ever Test innings gave Australian bowlers 175 runs to defend against Pakistan which they did in style. Nathan Hauritz's moment of glory.

12. 1/2 wkt wins with 50+ runs partnership for 8th/9th wicket   9w  10w   
   0096 1907 By  2 wkts eng-273ao AUS-300ao  eng-300ao AUS-275/8   56
   1012 1985 By  2 wkts pak-274ao NZL-220ao  pak-223ao NZL-278/8   50
   1097 1988 By  2 wkts pak-309ao WIN-306ao  pak-262ao WIN-268/8   61
   1268 1994 By  1 wkts aus-337ao PAK-256ao  aus-232ao PAK-315/9   22   57
   1453 1999 By  1 wkts aus-490ao WIN-329ao  aus-146ao WIN-311/9   54    9
   1658 2003 By  1 wkts bng-281ao PAK-175ao  bng-154ao PAK-262/9   52    9
   1972 2010 By  1 wkts aus-428ao IND-405ao  aus-192ao IND-216/9   81   11

These are fascinating matches in which the chasing teams achieved narrow wins by 1 or 2 wickets. However, these wins were unlikely ones in that there was a 50-plus run partnership for either the 9th or the last wicket. Each is a classic. Possibly the most exciting win was Pakistan's win over Australia, which is the only instance of an unbroken last wicket partnership of over 50 runs. The other classics are all here: the Brian Lara masterpiece at Bridgetown, Laxman special at Mohali and the Inzamam-inspired win against Bangladesh.

Which was the greatest of wins? The 1-run or 1-wkt wins are very close and might have provided great drama but the cause might not have been hopeless. But for sheer drama, courage and coming from behind, I have to narrow this search to the three wins after following on. Of these three, the one most talked about is the Calcutta win engineered by Laxman-Harbhajan-Dravid. But this match does not strike the right chords for me. India, at 232 for 4, were still 42 runs in arrears, but not a desperate situation. And Sourav Ganguly's decision to bat on the fifth day gave me the impression that he was not looking for a win. The win happened happenstance, that too by a wide margin. Now let us examine the other two matches. In the 1894 match, England had taken the lead with 6 wickets in hand. At SCG on the fifth day, any total above 150 was defendable, with an excellent bowling attack of Tom Richardson, Johnny Briggs, Bobby Peel and Bill Lockwood. However there is no arguing that this was a very close match.

The most exciting Test ever. The choice is individual: one could pick between the "The Ashes" Test, the first tie, the West Indian win by a single run against Australia, or those heart-stopping wins at Bridgetown, Karachi, Multan and Mohali. No one would go wrong. Result in doubt until the last ball, great individual performances and both teams "winning": what more can be asked for?

What about the 1981 Headingley classic? When Graham Dilley walked in, England were 135 for 7, still 92 in arrears. When Dilley was out, England were only 25 ahead. When Chris Old got out, England were only 92 ahead. Finally when Bob Willis was dismissed, the target for Australia was only 129. It needed one of the greatest fast bowling performances ever to give England the unlikeliest of wins. All things considered, I have no hesitation in nominating this Botham-Willis effort as the greatest of all wins. It is possible that non-match factors such as series status, opponent's fearsome reputation, own team's poor showing etc. might tilt the scale in favour of the 2001 win. But the 1981 match cannot be touched if we look at only the match.

As I have already explained this part itself has gone quite long. In the second part, I will examine the fascinating area of Test cricket, in the form of draws (in different hues), follow-ons, tied matches et al. I trust that at the end of this pair of articles you would have known whatever there is needed to know about Test results.

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