Shane Warne bowls Andrew Strauss with lunch approaching

Not all of cricket's great deliveries matter equally

© Getty Images

Stats feature

Golden wickets

Zeroing in on the most important dismissals in Test history

Anantha Narayanan  |  

One of the editors of the Cricket Monthly recently made an enquiry about an article in which I had valued wickets by the quality of batsmen dismissed. He asked, "Can you then tell me what is the best wicket ever?" I said, "Let us go further. I will give you the top dozen wickets, selected objectively." And here we are.

This is mostly an objective exercise. I am not going to select the most famous wickets - Warne dismissing Gatting or Holding's brutal over to Boycott. Those much-hyped wickets have historical significance but they did not significantly alter the course of a match or series. England lost both those Tests, and both series, comfortably. The pronouncements that Warne turned the ball anywhere between 12 to 30 inches or that Holding toyed with Boycott before dismissing him make for great drama and fine reading but are totally subjective. For that matter why could one not say that Warne's dismissal of Andrew Strauss at Edgbaston in 2005 was a greater delivery? Can one say that Muttiah Muralitharan has not claimed such a scalp or that Curtly Ambrose or Glenn McGrath or Waqar Younis have not bowled equally great overs?

As far as I am concerned, a wicket selected should change the course of the Test (and series) dramatically. It should change the winning percentage of the losing team from say 10% to 20% in favour of a win at that point in time or thereafter. There will be no subjectivity other than my judgement to select from among the shortlisted wickets.

To do this I use a concept developed jointly with Milind Pandit, my collaborator in many key areas of discovery. We analysed innings in which all ten wickets were captured and deduced the following about the resources the bowling side had available at the fall of each wicket.

1. 11.74%
2. 24.30%
3. 37.32%
4. 50.26%
5. 62.19%
6. 73.16%
7. 82.08%
8. 89.52%
9. 95.36%
10. 100%

These numbers are clear in explaining the value of the top-order wickets, especially the second, third and fourth. It is thus possible for us to make projections about expected innings-end scores. For instance, a team at 140 for 3 will be expected to score 375 (140/0.3732). A team at 200 for 5 will be expected to score 321 (200/0.6219). Yes, I know they could be dismissed for 250 or go on to score 450. But these are projections based on completely dependable historical data.

These values are computed from 5657 completed innings (until the end of Test No. 2220 in Centurion in late August). There was an option to do innings-specific projections but the real problem was the fourth innings. There are only 527 completed fourth innings and all but two have ended in losses. It was clear that making projections using data from lost matches would be error-prone. Hence, I decided to average across four innings.

The expected runs are derived using these values and adjusted if a significant partnership has been in progress. Using the target in front and the expected runs, I determine, with a fair degree of accuracy, the winning percentage of the batting team. Everything flows from this.

One subtle point. I look at the match status just before the fall of each wicket for the first to sixth wickets and just after the fall of the wicket for the latter wickets. The reasons are obvious: the quality of batsmen at the crease.

I have given below the example of this working for the Alan Davidson wicket in the tied Test in 1960.

Score 226 for 6, chasing 233. Partnership of 134 in progress
Projection: (226/0.8208) - 226 = 49 (later wicket, so status after fall of wicket)
Runs needed: 7
Partnership: 134
Partnership factor = 134/7 = 19.1 (limited to 5 and mapped to 1.5)* Expected runs: 49x1.50 = 73
Win chances: 73/(73+7) = 91.3%

*Partnership factor has a scale of 1-5. If >5, partnership factor is taken as 5. It is then mapped on a scale of 1 to 1.5 to determine expected runs. Partnership factor is an indicator of the momentum of the match and also of the tenure of the batsmen at the crease.

Holding v Boycott: great over, fine dismissal, but how much did it matter in the context of the match?

Holding v Boycott: great over, fine dismissal, but how much did it matter in the context of the match? © Patrick Eagar

The top three wickets selected themselves. Then I shortlisted 15 wickets and made my final selection of ten out of these, and have presented these as equals in reverse chronological order. I completed my selections from this table of wickets after carefully studying the scorecards and series situations and determining how easy or difficult run-making was.

The wickets need not be off great deliveries. It is immaterial whether they came off long hops, through awful strokes or even umpiring mistakes. I am not looking for aesthetics here but for decisive wickets that had great impact on the match and series.

Now the selection criteria, by innings.

Fourth innings
For the fourth innings, I only consider situations in which the capture of the wicket transforms the situation dramatically from an overwhelming win for the batting team to a narrow win for the bowling team or a tie. I do not go blindly for the last wicket that falls, even though that might be the decisive one. I look for the key wicket, earlier in the innings, that started a slide. Usually this wicket will be that of a much better batsman. However, there are situations in which the last wicket is the most important one. A key feature of this situation is the presence of a long tenth-wicket partnership. In fact, the most valuable wicket in Test history belongs to this genre.

The fourth innings is the easiest of all since the target is crystal clear. The objective, however, might vary: to go for the target or aim for a draw? Therefore it is quite easy to get a clear handle on the situation. I work out the winning percentage of the batting team. I have a high cut-off for this value and if the fall of this wicket is directly responsible for the bowling team's win, the match goes into the pot.

Third innings
The third innings is a totally different matter. Here I look for wickets that caused a significant collapse, leading to a third innings that finished way below the total that should have been realised.

First, an example. Let us give absolute parity to the two teams in the first innings. From 200 for 2, the team batting third collapses to 249 all out. Now if the team batting last scores 250 for 2, we don't have any wicket candidates from this match. But if the team batting last scores 250 for 8 to win, then we are in business. Maybe another 30 runs would have done the trick.

That means I look for collapses in third innings and narrow wins (by one or two wickets) in the fourth innings. That also means the decisive wicket has to be a top-order wicket. Fortunately there are only seven such Tests and I have selected two wickets from them for inclusion.

Second innings
The second innings is a beast, more a mind-bender than an analytical exercise. First, the team batting second should have lost the match. But they should have lost after being in a brutally commanding position in their first batting effort. How do I define this? After a lot of trials, I found that no team had gone past the other team's first-innings score without losing a wicket and still lost the match. And there were only two instances of a team crossing the opening team's score for the loss of one wicket and still losing the match. There I had my answer. I selected one of these matches.

That was close: West Indies celebrate after the Adelaide win in 1993

That was close: West Indies celebrate after the Adelaide win in 1993 © Getty Images

First innings
I studied the first innings very carefully. There was no way I could select a first-innings wicket that caused a collapse and conclude that it was a golden wicket. The simple reason is that there are three more innings yet to be played, unlike in the second innings, where a match situation of "X: 120, Y: 130 for 1" is decisive. Take Virender Sehwag's dismissal for 195 at the MCG in 2003. There is no doubt that the wicket sparked India's collapse from 311 for 3 to 366 all out. But then Australia took a first-innings lead of nearly 200 runs and won on the last day with nine wickets and 70 overs to spare. So how can anyone conclude that Sehwag's dismissal was almost solely responsible for India's loss? What if he had scored another 50 to 100 runs? Australia might still have won.

The bottom line is that there are no qualifying first-innings wickets.

Top three wickets in Test history

1. McDermott c Murray b Walsh 18
Test No. 1210 (1993)
Australia v West Indies, Adelaide
West Indies 252. Australia 213. West Indies 146. Australia 184
West Indies won by one run

West Indies saved the first Test of the series by the skin of their teeth. Australia won the next Test comfortably. The Sydney Test was a high-scoring draw. So they moved to Adelaide with the series in the balance.

West Indies finished with a first-innings score of 252. Curtly Ambrose's 6 for 74 gave West Indies a lead of 39. From 124 for 5, West Indies were spun out for 146, leaving Australia a target of 186 on the fourth day. Australia started poorly, losing wickets too often and were struggling at 102 for 8. Justin Langer and Tim May added 42 for the ninth wicket. When Langer was dismissed, Australia still needed 42 and the fat lady had started to exercise her vocal chords. But May, on the attack, and Craig McDermott, defending, had other ideas. They batted for an hour and a half and added 40, leaving the match tantalisingly close at 184 for 9. At this point, two runs were needed and the expected runs, based on the resources and ongoing partnership was 13. This led to a high winning percentage of 86.7 for Australia.

Then a rising ball from Walsh brushed McDermott's glove to give West Indies the closest victory ever, by one run. Australia's high winning chance, the ongoing partnership and the importance this wicket had for the match and series, gave this the honour of the most valuable wicket in Test history. A drop by Junior Murray could have led to a tie or a win.

West Indies went on to win the last Test and take the series 2-1.

2. Davidson run out 80
Test No. 498 (1960)
Australia v West Indies, Brisbane
West Indies 453. Australia 505. West Indies 284. Australia 232
Match tied

The first Test of one of the greatest series of all time was played in Brisbane. Two huge first innings ended with a 52-run lead for Australia. Davidson was instrumental in dismissing West Indies for a middling total, 284. Australia had to score 233, with most of the last day available.

Wes Hall was devastating, as Australia lost five wickets for 57 and then six for 92. A huge win for West Indies was on the cards. Richie Benaud joined Davidson and they had a brilliant partnership for well over three hours, with Davidson playing the aggressor. They added 134 and took the score to 226 for 6. Seven runs were needed with four wickets in hand. At this point, the expected runs were 73. The chances of an Australian win were 91.2%.

West Indies prayed for miracles, and it came in the form of brilliant fielding. Davidson was run out when Benaud called for a sharp single, but Joe Solomon's direct hit was decisive. Incidentally, Australia were in great shape even after the loss of Davidson's wicket. At 228 for 7, with Benaud at the crease, the win was on (the winning percentage was 90.9%). Then Hall dismissed Benaud and two more wickets fell for the addition of only four runs. Benaud's wicket could have been selected as well.

The series ended tantalisingly close: a 2-1 win for Australia.

Edgbaston 2005:

Edgbaston 2005: "Jones!... Bowden!..." © Getty Images

3. Kasprowicz c Jones b Harmison 20
Test No. 1758 (2005)
England v Australia, Edgbaston
England 407. Australia 308. England 182. Australia 279
England won by two runs

The Ashes in 2005 was one of the all-time great series. A huge Australian win at Lord's indicated a rout in the offing. The caravan moved across the M1 to Edgbaston. Two middling first innings meant that England led by 99. Shane Warne spun his magic to dismiss England for 182, leaving a target of 282.

Australia started poorly and were down in the dumps at 137 for 7. Michael Clarke was dismissed off the last ball of the third day's play and Australia were tottering on the brink at 175 for 8, with over 100 runs still to get. The next day, Warne and Brett Lee added 45 before Warne was dismissed with Australia on 220. It seemed like curtains with one wicket available and over 50 to get.

Michael Kasprowicz joined Lee and they attacked the bowling, adding 59 in 12 overs. In between, Kasprowicz was dropped by Simon Jones. At 279 for 9, with three runs needed, the expected runs were 21. The chances of an Australian win were 87.5%. Then a rising Harmison delivery took Kasprowicz's glove, and the wicketkeeper, Geraint Jones, completed the catch. The final wicket of this Test was very similar to that of the Adelaide classic of 1993.

England went on to win the Ashes 2-1. The series became one for the gods when Kasprowicz's wicket was taken. If Australia had gone 2-0 ahead, the series could have ended as an easy win for them.

Next ten most important wickets in Test history
These are presented in reverse chronological order. There is no attempt to rank them.

4a. Ponting c Southee b Bracewell 16
Test No. 2021 (2011)
Australia v New Zealand, Hobart
New Zealand 150. Australia 136. New Zealand 226. Australia 233
New Zealand won by seven runs

Australia needed 241 for a tough win. David Warner was magnificent, and useful contributions from Phil Hughes, Usman Khawaja and Ricky Ponting took the team to 159 for 2. The target was 82 runs away and the projection was 269. Australia were sitting pretty at 76.5%. Strains of "Waltzing Matilda" were heard in the distance.

Hobart 2011: Ponting c Southee b Bracewell

Hobart 2011: Ponting c Southee b Bracewell William West / © AFP

Then, in his farewell innings in Hobart, Ponting was caught at extra cover off Doug Bracewell. No one thought it was the beginning of the end, but it was. Bracewell took five more wickets to dismiss Australia for 233: a narrow loss by seven runs. An all-time great innings by Warner was thwarted by an all-time great bowling effort by Bracewell.

4b. Mohammad Yousuf c Haddin b Johnson 46
Test No. 1945 (2010)
Australia v Pakistan, Sydney
Australia 127. Pakistan 333. Australia 381. Pakistan 139
Australia won by 36 runs

This is the only selection from the second innings. In reply to Australia's 127, Pakistan were 144 for 1 and 205 for 2. I have selected Mohammad Yousuf's dismissal at 237 for 3 as the golden wicket since he was one of the world's best batsmen at the time. He was caught behind off Mitchell Johnson, though this did not spell disaster immediately. Pakistan still managed a lead of over 200. But that was wiped out by a Mike Hussey classic before Nathan Hauritz and Johnson ran through Pakistan to win by 36 runs. Pakistan's winning chances were very high before the fall of Yousuf's wicket.

4c. Tendulkar c Wasim Akram b Saqlain Mushtaq 136
Test No. 1442 (1999)
India v Pakistan, Chennai
Pakistan 238. India 254. Pakistan 286. India 258
Pakistan won by 12 runs

To describe this innings is to gild the lily. Suffice to say that at 254 for 6, it looked like curtains for Pakistan (79.5% for India). Tendulkar was dismissed for 136 and Pakistan secured a miracle win.

4d. Mark Waugh b Waqar Younis 61
Test No. 1268 (1994)
Pakistan v Australia, Karachi
Australia 337. Pakistan 256. Australia 232. Pakistan 315 for 9
Pakistan won by one wicket

This is the first of two third-innings wickets. Playing in tough conditions in Karachi, Australia were 171 for 2 after securing a first-innings lead of 81. An insurmountable 400-run target seemed likely. Australia's winning chance was 75.7% (a target of 92 versus projection of 287). The notional target was 263. When Mark Waugh was bowled by a Waqar Younis special, very few could have anticipated that there would be a horrendous collapse with no one else reaching double figures. Australia lost eight wickets for 61.

Mohammad Yousuf's dismissal in the Sydney Test of 2010 did not wreck Pakistan's first innings, but the loss of his wicket meant his side's chances of a win plummeted

Mohammad Yousuf's dismissal in the Sydney Test of 2010 did not wreck Pakistan's first innings, but the loss of his wicket meant his side's chances of a win plummeted Mark Nolan / © Getty Images

Although Australia set Pakistan a decent target, Inzamam-ul-Haq's magnificent innings, with help from the lower order, took Pakistan to a one-wicket win. Twenty-five more runs would have been enough for Australia. Let me reiterate that this match would not have qualified if Pakistan had scored 315 for 4 to win the game. Readers can identify the difference in dynamics between the third and fourth innings.

4e. PA de Silva c Border b McDermott 37
Test No. 1194 (1992)
Sri Lanka v Australia, Colombo. Australia 256. Sri Lanka 547 for 8d. Australia 471. Sri Lanka 164
Australia won by 16 runs

Sri Lanka dominated the first two innings, and were set a target of 181. They were coasting at 127 for 2 (winning chances 79.8%) when disaster struck. De Silva, who was batting fluently, went for an almighty slog and the skier was brilliantly caught by Border at mid-on. What happened next was similar to the collapse in the Karachi Test in 1994: seven single-digit dismissals and the loss of eight wickets for 37 runs. Australia walked on water through the combined efforts of Greg Matthews and Warne.

4f. Thomson c Miller b Botham 21
Test No. 943 (1982)
Australia v England, Melbourne
England 284. Australia 287. England 294. Australia 288
England won by three runs

Three days, three innings. The fourth day started with Australia needing 292 for a win. They were 218 for 9 and then came the stand between Border and Jeff Thomson. The two took the score to 288 for 9. Now Australia, with winning chances of 84% (the target down to four and a projection of 21), were on top. Thomson edged an away delivery and Geoff Miller took the catch after it popped out of Chris Tavaré's grasp. One of the greatest Ashes Tests ended with less than a boundary separating the teams.

Melbourne 1982: Miller catches Thomson to end a cliffhanger

Melbourne 1982: Miller catches Thomson to end a cliffhanger © Getty Images

4g. Kapil Dev c Gooch b Willey 0
Test No. 854 (1979)
England v India, The Oval
England 305. India 202. England 334 for 8d. India 429 for 8
Match drawn

England set India 438 for a win. On the last day, a run a minute was needed. The top three took India to 366 for 2, at which point an upset seemed probable. Sunil Gavaskar was batting beautifully and the chance of a win was 90%, although time was becoming a crucial factor. Kapil was sent up: a questionable decision. When he got out, the win had just about slipped away. Gavaskar's wicket at 389 for 3 could also have been considered. I selected Kapil's wicket since he scored 0. Fifteen from the last over proved too much and India downed the shutters. This is the only drawn Test in this selection.

4h. Border b Sarfraz Nawaz 105
Test No. 849 (1979)
Australia v Pakistan, Melbourne
Pakistan 196. Australia 168. Pakistan 353 for 9d. Australia 310
Pakistan won by 71 runs

Australia were set 382. Border, ably supported by Andrew Hilditch and Kim Hughes, took the score to 305 for 3 and a terrific win was in the offing. The win percentage for Australia was 79.7 (a target of 77 versus projection of 302). Border was bowled for 105 by Sarfraz, who then launched into one of the greatest spells of fast bowling: seven wickets for one run in 33 balls. Australia were dismissed for the addition of just five runs. This was arguably Pakistan's greatest away win, matching their famous victory in Port-of-Spain in 1958.

4i. Dexter c Grout b Benaud 76
Test No. 510 (1961)
England v Australia, Manchester
Australia 190. England 367. Australia 432. England 201
Australia won by 54 runs

Bolstered by a last-wicket stand of 98, Australia set England 256. England's chase began with a good opening stand, followed by Ted Dexter, imperious as ever, attacking his way to 76. England's chances, at 150 for 1, were 81.5%. Then Benaud, using the rough created by Fred Trueman's and Jack Flavell's footmarks, dismissed Dexter, Peter May, Brian Close and Raman Subba Row to leave England at 163 for 5. Australia won comfortably but it was Dexter's wicket that started England's downward spiral.

England were well on course at Old Trafford in 1961 before this wicket: Dexter c Grout b Benaud

England were well on course at Old Trafford in 1961 before this wicket: Dexter c Grout b Benaud © Getty Images

4j. Cowper c Indrajitsinhji b Nadkarni 81
Test #567 (1964)
India v Australia, Bombay
Australia 320. India 341. Australia 274. India 256 for 8
India won by two wickets

This is the second match where the wicket selected was in the third innings. Two middling first innings left Australia with a deficit of 21. In the third innings, Australia were well placed at 246 for 3 and the notional target was 300. The win chance for Australia was 82.1% (a target of 53 and projection of 243). Then Bob Cowper was caught behind off the persistent Bapu Nadkarni and the cat was set among the pigeons. Australia lost their next five wickets for 28. India reached the target of 254 for the loss of eight wickets. If Dilip Sardesai had scored a hundred and India had scored 256 for 4, this Test would not have been considered.

None of the 13 wickets selected is from before the first World War. There were three or four such matches in the first shortlist. However, in most of these Tests the targets were low and with the prevailing low-scoring trends, batting was extremely difficult. In Test No. 9, the scores were 63, 101 and 122. Despite the low target of 85, runs were hard to come by and England managed only 77. The match that came closest was Test No. 19, where the final innings was almost identical to the one in the first tied Test.

Surprisingly there was nothing of note during the inter-war period, possibly because of the strong batting sides that dominated play in this era.

The cornerstone of this analysis is that the wickets picked needed to effect a complete transformation in the match. The preliminary selections are objective and the final selections are made after a perusal of related factors. Readers will have their own selections of golden wickets and I am interested in hearing of their selections.

Below is a list of the wickets that missed out on the final list.

Test No. 19. Read b Spofforth 56
Test No. 149. Catterall b Macaulay 76
Test No. 611. Simpson c Goddard b Pollock 65
Test No. 1052. Chetan Sharma c McDermott b Bright 23
Test No. 1787. Gayle c Fleming b Astle 82
Test No. 2148. Vijay lbw Lyon 99

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