ODIs lost from the brink of victory September 28, 2013

Tripping up at the home stretch

A look at ODIs lost from the brink of victory
22

Charl Langeveldt arguably bowled the best last over in ODI history © Getty Images

This is a follow-up article to the previous one in which I had introduced a new concept of measuring the status of ODI matches using the resources available for the team. In the first article, I had looked at the wins by second batting teams, from perilous situations. In this article I will look at losses by teams batting second, from seemingly impregnable positions.

First, let me present the revised table of wicket resources. As suggested by Anshu Jain, I split the innings into first and second and determined the wicket resources. Since there seems to be a significant variation between the resources utilised at the fall of most wickets, exceptions being 3rd and 4th wickets, I will use the appropriate innings related wicket resource values in all further work.

Resource percentages
WicketAllInns******FirstInns******SecondInns******
ResUtilizedResAvlblResUtilizedResAvlblResUtilizedResAvlbl
1 16.72%83.28% 14.67%85.33% 18.81%81.19%
2 30.86%69.14% 29.57%70.43% 32.26%67.74%
3 45.34%54.66% 44.82%55.18% 45.95%54.05%
4 57.96%42.04% 58.69%41.31% 57.04%42.96%
5 68.83%31.17% 70.42%29.58% 66.66%33.34%
6 77.39%22.61% 79.25%20.75% 74.78%25.22%
7 84.37%15.63% 85.89%14.11% 82.28%17.72%
8 90.15% 9.85% 91.18% 8.82% 88.88%11.12%
9 95.33% 4.67% 95.72% 4.28% 94.92% 5.08%
10100.00% 0.00%100.00% 0.00%100.00% 0.00%

This part of the second-innings analysis is quite tricky and far more complex than the more straight-forward chasing wins analysis. Earlier I was looking for the worst situations. As such it was easy to look at the fall of the wicket and determine the TRF-W (Target Resources Factor-Wickets). Compare this with TRF-B (Target Resources Factor-Balls) and select the higher one which would indicate the more critical of the two resource situations. When the team wins from such difficult situations it was easy to gauge the quality of win correctly.

This is a different ball game. Wickets are fine and the wicket resources are determined similarly. Except for a minor but very important tweak. Let us say that the first wicket fell at 120, chasing 250. The wicket resources will be calculated at 120 for 0, not 120 for 1. This means we have to consider the situation before the ball was bowled, not after. The reason is obvious. 120 for 0 is better situation than 120 for 1 and since the team has plummeted into a loss, we have to take as the base the better situation, unlike in the last analysis wherein we looked at the worse situation. This is relatively easy to understand.

Here comes the tougher part. There is no way for me to ignore the balls resource available. If I tell you that a team is at 150 for 1, chasing 250, the sentence does not convey anything. One cannot venture a comment on the team situation without knowing the balls available information. While chasing 250, 150 for 1 in 25 overs is a 95% winning situation, 150 for 1 in 35 overs is a 65% winning situation, 150 for 1 in 40 overs is a 35% winning situation and 150 for 1 in 45 overs is a 5% winning situation. So the balls available data is a must.

So much so, I am forced to exclude the 650 or so chasing wins for which balls-at-wicket-fall data is not available. I tried doing this based on assumptions on scoring rates but the results were skewed because of the dynamics between the two types of resources. When I used the scoring rate at the fall of wicket, many matches were included and I am not sure which match situation is good or worse. I do not want to say that the team lost a certain win from 150 for 1 for someone to say that I am a fool since they had only 7 overs to score 100 runs. So let us leave it at that.

Now comes the other tricky part. The scoring rate at the wicket-fall stage cannot be really taken as the scoring rate for the later part of the innings. If a team is 150 for 1 in 35 over, chasing 250, I cannot really take the actual scoring rate of 4.2 and conclude that the required rate of 6.3 is 50% harder. With nine wickets in hand, this is really a cakewalk. On the other hand, if the score at the same stage had been 150 for 4, the win is not that straight-forward. So the number of wickets in hand plays a part in getting a handle on this scoring rate which can be achieved. I have adopted the following rates. Not necessarily arbitrary since some logic has gone into it. I did not want to over-complicate at this stage of analysis to do an actual scoring rate achieved. Over thousands of matches the scoring rates for the rest of the innings, at the fall of early wickets, is likely to be in a narrow band of 4.5 to 5.0 and that does not help. I have left that for a later improvement, if feasible. I have ranged this between 7.0, which I feel is the highest achievable scoring rate over a number of overs, and 5.0. In addition, these expected rates are incremented by 10% if fewer than ten overs remain. Changing 7.0 to 7.7 will not cause much of a change. Anything above is infeasible.

Wkts-in-hand   Exp-S/R
10 7.0 9 6.7 8 6.4 7 6.15 6 5.9 5 5.7 4 5.5 3 5.3 2 5.15 1 5.0

Let me now summarise the calculations.

1. Determine the wickets-resource left by looking at the number of wickets lost, just before the ball is bowled. Divide the target factor (Runs left/Target runs) by this value to arrive at the TRF-W.
2. Determine the Expected SR by looking at the above table, again just before the ball is bowled.
3. Determine the Actual SR required by dividing the number of runs by number of balls left.
4. Determine the TRF-B as the ratio between the above two scoring rates (Actual/Expected). This will have a low value (easy task) when fewer wickets have fallen and the scoring rate required is not high. It will have a high value when more wickets have fallen and the scoring rate has climbed up.
5. Determine the TRF-S (Target Resources Factor-Situation) as equivalent to 0.66667*TRF-W + 0.33333*TRF-B. Wickets are far more important to retain and this fact is recognized in this equation. These are not golden numbers. These are just weights and are driven by common-sense. I tried 0.5 and 0.5 and was not happy with quite a few of the situations. A choice of 0.75 and 0.25 lowered the impact of the Balls resource too much.
6. Select situations in which the TRF-S value is lower than 0.50. This is a very happy situation to be in. And let us not forget that the team proceeded to lose from this invincible position.

I will explain this with a few clear examples.

Let us say that a team is chasing 249. The target is 250. The score before the fall of the second wicket is 150 for 1. The Wkt-resource in front is 81.19%. The target in front is 40.0% (100/250). So the TRF-W value is 0.493 (40.00/81.19).

Now let me take two scenarios. The score of 150 for 1 has been reached in 25 overs. The Expected -RpO is 6.7 (from the table). The required RpO is 4.0 (100/25). The TRF-B value is 0.597 (4.0/6.7). The final TRF-S value is 0.66667*0.493 + 0.33333 *0.597. This works to 0.527. An excellent situation to be in. This match would just miss selection for this analysis, if the team lost.

Now let me say that the score of 150 for 1 was reached in 35 overs. The Expected-RpO remains at 6.7. The required RpO is 6.6667 (100/15). The TRF-B value is 0.995 (6.6667/6.7). The final TRF-S value is 0.66667*0.493 + 0.33333*0.995. This works to 0.660. Still a good situation to be in.

Finally let us say that the openers have dawdled and the score of 150 for 1 was reached in 41 overs. The situation changes drastically. The Exp-RpO changes to 7.37 (6.7 * 1.1). The required RpO is 11.111 (100/9). The TRF-B value is 1.507 (11.111/6.37). The final TRF-S value is 0.66667*0.493 + 0.33333*1.507. This works to 0.830. A reasonable situation only because of the 9 wickets in hand.

Change the situation to 150 for 4 in 40 overs. The TRF-W is 0.931(40.00/42.96). The TRF-B comes to 1.695(10.00/5.9). The weighted TRF-S is 1.186. The balance has clearly shifted to the fielding side. Note how delicately the whole situation is balanced and the interplay of the various factors.

Let me take a breath now.

Out of the 2100 or so matches considered, and the 1056 defending wins/ties therein, 21 matches qualify under these conditions. This time I have not gone on the TRF-S value to feature matches. I have featured the matches which had more than one such situation in the match. In other words, the losing team let go more than one opportunity. Makes these losses special, so to speak. I could get seven such matches. I have selected three more matches from the list of 21; matches that caught my eye. It can be seen that most of the matches have been lost by small margins. It seems logical. A team at 107 for no loss, chasing 209, is unlikely to lose by 50 runs.

Let me now present the table of 21 selected matches. The table is self-explanatory. The downloadable table is complete and presents all the multiple situations in the matches.

ODI matches lost from certain win situations
Match IdScoreWkt-Res %T1-ScoreT2-ScoreTarget-%TRF-WEquationReq-RpOExp-RpOTRF-BTRF-S
Featured
3080188/8 11.1%190/10189/10 1.6%0.141 3 in 460.395.660.0690.117
2600192/3 54.0%210/ 8209/ 6 9.0%0.167 19 in 284.076.770.6020.312
3120164/8 11.1%171/10165/10 4.7%0.418 8 in 172.825.660.4980.445
1450249/8 11.1%252/ 9249/10 1.6%0.142 4 in 73.435.660.6050.297
2642199/4 43.0%233/ 9219/1015.0%0.348 35 in 573.686.490.5680.421
2269196/9 5.1%198/10196/10 1.5%0.297 3 in 111.645.500.2980.297
2734271/4 43.0%282/ 8281/ 6 4.2%0.099 12 in 164.506.490.6930.297
2243281/6 25.2%284/ 6283/10 1.4%0.056 4 in 64.006.050.6610.257
1344272/4 43.0%307/ 6301/1011.7%0.272 36 in 395.546.490.8530.466
1514196/8 11.1%196/10196/10 0.5%0.046 1 in 61.005.660.1770.089
Included
1283221/5 33.3%241/ 9235/10 8.7%0.260 21 in 255.046.270.8040.441
1294125/0100.0%228/ 7227/ 945.4%0.454104 in 1703.677.000.5240.478
1405198/4 43.0%232/ 8222/1015.0%0.350 35 in 464.576.490.7030.468
1722196/3 54.0%242/ 8240/1019.3%0.358 47 in 594.786.770.7070.474
1941216/6 25.2%229/ 7224/10 6.1%0.241 14 in 174.946.050.8170.433
2520235/5 33.3%257/ 8252/ 9 8.9%0.267 23 in 265.316.270.8470.460
2535203/5 33.3%221/ 9221/10 8.6%0.257 19 in 383.006.270.4780.331
2682301/3 54.0%340/ 6340/ 711.7%0.217 40 in 386.326.770.9340.456
2826212/3 54.0%270/ 7244/ 714.2%0.262 35 in 346.186.770.9130.479
3135222/6 25.2%243/10225/10 9.0%0.358 22 in 522.546.050.4200.378
3215155/3 54.0%200/10174/1022.9%0.423 46 in 733.786.150.6150.487

Let us now see the featured matches now. The sequence is rather arbitrary. In general I have shown the matches where the teams lost despite many chances first.

1. ODI # 3080. South Africa vs. India.
Played on 15 January 2011 at New Wanderers Stadium, Johannesburg.
India won by 1 run. Mom: Munaf Patel
India: 190 all out in 47.2 overs
   Yuvraj Singh        53 ( 68)
South Africa: 189 all out in 43.0 overs
   GC Smith            77 ( 98)
   MM Patel            8.0  0  29  4
This is arguably amongst the most astonishing matches ever played. The more I see what happened in the match, the more I feel like I have been caught in the eye of a typhoon, going round and round. India, batting first, at the Wanderers, reached a very ordinary total of 190. South African, despite losing the first wicket early, were coasting. Starting at 152/4 (the fifth wicket falling at 25.2), South Africa were looking almost certain winners at 160/5, 163/6, 177/7, 188/8 and 189/9. On each of these situations their TRF-S was well below 0.50. Surprisingly the best situations were at 188/8 (0.117) and 189/9 (0.155). But they lost the match by 1 run. How? It is a question they might still be trying to find an answer for. A combination of panic, good bowling by Munaf Patel and good fielding contributed to the disaster.
2. ODI # 2600. Ireland vs. Netherlands.
Played on 11 July 2007 at Civil Service Cricket Club, Stormont, Belfast.
Ireland won by 1 run. Mom: Kevin O'Brien 
Ireland: 210 for 8 wkt(s) in 50.0 overs
   EJG Morgan          51 (112)
   D Langford-Smith    31*( 13)
Netherlands: 209 for 6 wkt(s) in 50.0 overs
   Mudassar Bukhari    71 (114)
This was an equally amazing match played between two talented Associate teams. Ireland scored a moderate 210 for 8. Netherlands looked certain to win when they were 138/1, 159/2, 192/3 (0.312) and 192/4. They lost a wicket at each of these situations and were finally at 198 for 5 (0.419), needing 13 to win in 14 balls. They managed to fall couple of runs short. The scores of the later four batsmen, 15 in 24, 10 in 15, 2 in 5 and 5 in 7, tell the complete story.
3. ODI # 3120. England vs. South Africa.
Played on 6 March 2011 at MA Chidambaram Stadium, Chepauk, Chennai.
England won by 6 runs. Mom: Bopara R.S.
England: 171 all out in 45.4 overs
   IJL Trott           52 ( 94)
   RS Bopara           60 ( 98)
South Africa: 165 all out in 47.4 overs
   HM Amla             42 ( 51)
   SCJ Broad           6.4  0  15  4
This match is fresh in everyone's memory. World Cup 2011, and the unfancied England played the eternal bridesmaids, South Africa. This time the South Africans had it all figured out. England could not make head or tail of Imran Tahir and Robin Peterson and were dismissed for 171. At 124/3, 160/7 and 164/8(0.445), South Africa looked very clear favourites to win. Although the numbers may not reveal this, they needed 48 in 108 balls with seven wickets at the first-mentioned situation. That was their best chance. But they let go of all these chances and were finally dismissed for 165, six runs short. Stuart Broad was unplayable at the end. But the real culprit was Peterson who scored 3 in 16 balls.
4. ODI # 1450. India vs. Zimbabwe.
Played on 19 May 1999 at Grace Road, Leicester.
Zimbabwe won by 3 runs. Mom: Grant Flower
Zimbabwe: 252 for 9 wkt(s) in 50.0 overs
   A Flower            68*( 85)
India: 249 all out in 45.0 overs
   S Ramesh            55 ( 77)
   HH Streak           9.0  0  36  3
   HK Olonga           4.0  0  22  3
This match was during the 1999 World Cup. Zimbabwe posted a very competitive total of 252. This was not like the earlier referred matches in which the losing team had established their ascendancy early in the innings. India struggled at the start and was behind the game at 103 for 5. Then they recovered and reached recovered and were very comfortably placed at 246 for 7, requiring only 7 runs in 10 balls with three wickets left. Then Henry Olonga struck. Even then, at 249 for 8 (0.297) and 249 for 9, they were well placed to win. But then fell 3 runs short. This was an amazing result considering the difference in strength between the two teams.
5. ODI # 2642. Pakistan vs. South Africa.
Played on 29 October 2007 at Gaddafi Stadium, Lahore.
South Africa won by 14 runs. Mom: Makhya Ntini
South Africa: 233 for 9 wkt(s) in 50.0 overs
   HH Gibbs            54 ( 61)
   JH Kallis           86 (130)
Pakistan: 219 all out in 46.3 overs
   Younis Khan         58 ( 65)
   Mohammad Yousuf     53 ( 88)
   M Ntini             9.0  0  61  4
   JA Morkel           8.3  0  44  4
South Africa set Pakistan a fair target of 234. They were well ahead of the game at 199 for 4 (0.421), requiring 35 in 57. Then the score became 209 for 5 and then 219 for 7. Even then 15 in 24 looked easy. At this point Albie Morkel captured three wickets in four balls and sent Pakistan crashing to a 14-run defeat. That too, at home.
6. ODI # 2269. Africa XI vs. Asia XI.
Played on 17 August 2005 at SuperSport Park, Centurion.
Africa XI won by 2 runs. Mom: Ashwell Prince
Africa XI: 198 all out in 44.3 overs
   AG Prince           78*(113)
Asia XI: 196 all out in 48.1 overs
   Abdul Razzaq        38 ( 77)
   SM Pollock         10.0  1  32  3
   JH Kallis          10.0  2  42  3
This time it is an Africa XI. Another low score of 198 against the motley collection of players known as Asia XI. Chasing 199, Asia XI lost wickets regularly and were down in the dumps at 59 for 4 and 96 for 7. Then they recovered and were at 193 for 8, requiring 6 in 16. Soon they lost the ninth wicket and were 196 for 9 (0.297), needing only 3 runs. But fell 2 runs short.
7. ODI # 2734. West Indies vs. Australia.
Played on 4 July 2008 at Warner Park, Basseterre, St Kitts.
Australia won by 1 run. Mom: Andrew Symonds
Australia: 282 for 8 wkt(s) in 50.0 overs
   A Symonds           87 ( 78)
   DJ Hussey           50 ( 51)
West Indies: 281 for 6 wkt(s) in 50.0 overs
   CH Gayle            92 ( 92)
   RR Sarwan           63 ( 79)
   S Chanderpaul       53 ( 71)
   B Lee              10.0  0  64  3
For a change this was a close result in a big-scoring match. Australia scored a very competitive total of 282 for 8. At 138 for 1 and 188 for 2 West Indies were quite comfortably ahead. Even when the score was 247 for 3, the situation was very good. However the best situation was at 271 for 4, West Indies requiring only 12 in 16 balls. They were very well placed at a TRF-S value of 0.297. But West Indies floundered inexplicably and scored only 10 runs in the next 12 balls. They ended just a run short.
8. ODI # 2243. West Indies vs. South Africa.
Played on 11 May 2005 at Kensington Oval, Bridgetown, Barbados.
South Africa won by 1 run. Mom: Charl Langeveldt 
South Africa: 284 for 6 wkt(s) in 50.0 overs
   HH Dippenaar       123 (129)
   JH Kallis           87 (109)
West Indies: 283 all out in 49.5 overs
   CH Gayle           132 (152)
   A Nel              10.0  0  42  3
   CK Langeveldt       9.5  0  62  5
This was a most extraordinary match, won by a single bowler, like Albie Morkel did against Pakistan. How often have South Africa figured in these matches on the other side of the fence? South Africa put up a most impressive total of 283. Aided by a top-class century from Chris Gayle, but still losing wickets steadily, West Indies were 281 for 6 (0.257) requiring just 4 runs for a win. Then they were 283 for 6, requiring only two runs. Charl Langeveldt produced, arguably, the best last over in ODI history and claimed a hat-trick, letting the South Africans win by 1 run. A hat-trick was badly needed and he produced it.
9. ODI # 1344. Sri Lanka vs. India.
Played on 7 July 1998 at R. Premadasa Stadium, Colombo.
India won by 6 runs. Mom: Sachin Tendulkar
India: 307 for 6 wkt(s) in 50.0 overs
   SC Ganguly         109 (136)
   SR Tendulkar       128 (131)
   A Jadeja            25 ( 15)
Sri Lanka: 301 all out in 49.3 overs
   PA de Silva        105 ( 94)
   AB Agarkar         10.0  0  53  4
This is the only featured match that produced two scores of over 300 runs. Helped by two centuries at the top, India scored 307 and looked comfortable winners. But Sri Lanka never gave up. Their best position was 272 for 4 (0.466). They needed only 36 in 39 balls. Reasonably comfortable position. However they started losing wickets regularly and fell 6 runs short. The last 5 batsmen scored 12 runs in 25 balls, a woeful effort indeed.

This match is very similar to ODI 2932, in which the mammoth Indian total of 414 was almost chased down by Sri Lanka, who fell 3 runs short. The only reason it does not get into this list is because at 401 for 6, chasing 414, the balls situation, 14 in 10, was not that favourable to Sri Lanka. And it proved difficult. They could not make those runs.

10. ODI # 1514. Pakistan vs. Sri Lanka.
Played on 15 October 1999 at Sharjah C.A. Stadium.
Match tied. Mom: Abdul Razzaq.
Pakistan: 196 all out in 49.4 overs
   Mohammad Yousuf     48 ( 90)
Sri Lanka: 196 all out in 49.1 overs
   RS Kaluwitharana    75 (108)
   RP Arnold           61 ( 93)
   Wasim Akram        10.0  1  38  3
   Abdul Razzaq        9.1  2  31  5
This is the only tied match featured here. Pakistan were dismissed for 196. Chasing the relatively modest target of 197, Sri Lanka looked certain winners at various positions such as 157/1, 173/3, 174/3, 177/4, 186/5, 186/6, 194/7 and finally 196/8. The TRF-R values were way under 0.500 in all these situations. Interestingly the best situation was at 196 for 8 (0.089) since the scores were level and 1 run was needed. However, they lost both wickets and the match was tied. The batsmen at positions 4-11 scored 25 runs in 71 balls. These included Aravinda de Silva, Sanath Jayasuriya, Mahela Jayawardene and Chamara Silva, all recognised batsmen.

I have created a document file with details of all matches in which the TRF-S values were below 0.50. This includes multiple occurrences within the same match. To download/view the document, please CLICK HERE.

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

Comments have now been closed for this article

  • AnanthNarayanan on September 30, 2013, 17:37 GMT

    [[
    I have created another document file with details of all matches in which the TRF-S values below 0.90. This includes multiple occurrences within the same match. To download/view the document, please CLICK HERE.
    Ananth
    : ]]

  • ThinkingCricket on September 30, 2013, 14:48 GMT

    Nice article, but I think some of your assumptions are strongly weighted towards situations that are not such huge-turnarounds. In particular you are overweighting turnarounds that happen with low sample sizes of 1 or 2 overs. Two or three tail-enders falling with 12 runs required isn't really an epic collapse, because resource percentages often miss the point that one bad over isn't a huge turnaround. Common sense can tell us that 177/7 chasing 190 isn't a particularly comfortable situation and I wouldn't expect that to result in a win more than 80 or 90% of the time, so for that to be one of the biggest turnarounds in history seems like an anomalous result. Intuitively I think anyone would agree that 157/1 chasing 196 (bottom match) is a far more spectacular collapse than 152/4 chasing 190, or 177/7 chasing 190 or 189/9 chasing 190 (the last of which is literally saying that No.11 being bowled with two runs required is a bigger surprise than a collapse from 157/1 to 196 all out).
    [[
    I have no problems with any of your statements. This is the first attempt and the next try will be based on additional insights such as yours. If you see my comments on the respective weights, it is quite possible that 75/25 is probably the more correct one. That would have over-valued the wickets-resources in hand. Having said that, 196 for 8, with 1 run needed is still in many ways an excellent position. A no ball, wide, bye, leg-bye, hit-and-run, to-the-keeper-and-run, lucky hit through slips, misfield: I could go on. 157 for 1 still required that the good quality effort had to be sustained for another 8-10 overs.
    But looking at it from a subjective angle, 157 for 1 would be the position most captains would have liked to be in. Let us see how to recognize these situations. It could be even that a position of 157 for 1, 30 needed in 87 balls, should completely discount the balls-available factor. The TRF-W works out to 0.250. This is lower than the current TRF-S of 0.304.
    But very good food for thought. Many thanks.
    Ananth
    : ]]

  • on September 30, 2013, 9:55 GMT

    Hi Ananth, Great work!! I was wondering if something like D/L method to estimate the requirement at a given stage not be a more standard way? If you do not like D/L method, take the other variant. But basically, you can say that given the situation (while chasing, e.g., 120/1 in 20 overs), you can also compute the par score and see what's the difference between the current score and par score to decide how much ahead/behind the chasing team is. That way, you will not have to make choices and defend them. In fact, where I was think of this was the analysis of how valuable the runs scored by a batsman are (in an ODI situation)? If you leave your team better off than when you arrive, your runs are valuable, else they count for much. What do you think of this idea?
    [[
    NM, there is one difference between D/L and my working. There is nothing in my methods which are like Rocket Science. It would take anyone 5 minutes to understand the method and apply it in real life.
    D/L is far more complicated and the purpose is different. They determine par scores at any time. I determine the winning chances. If the TRF-S is 0.3, what I really infer is that the chances of winning are 3.3 times that of the other team. If the TRF-R was 2.50, the chances of winning are 40% of those of the other side.
    As far as defending numbers are concerned, when have the learned duo defended their numbers. Those are taken for granted. I have a basis for every number. I may also very well do some work to determine the expected RpO also.
    Ananth
    : ]]

  • Anand_S on September 29, 2013, 14:28 GMT

    Ananth I was refering to match #2357. I made an error, I read 117-3 as 117-1. I take my question back now. Also my apologies for not noting your mention about the match in which Malinga and Langeveldt did magic for their teams. The other match India vs NZL I asked about was match # 1933. Chasing 200 INdia was 182-3 in 42 overs needing 18 to win with 7 wkts and 8 overs. They ended up winning by 1 wicket and one ball to spare. Had they lost where would this be in your analysis?
    [[
    This was some match. I think the best situation was at 141 for 1, needing 59 in 105 balls. The TRF-S was 0.397 (0.348 & 0.494), way below the cut-off point. Because of the number of such situations I am certain it would have been featured. Come to think of it, what was the situation at 197 for 7. The TRF-S was 0.1668 (0.095 & 0.3087).
    Ananth
    : ]]

  • on September 29, 2013, 10:57 GMT

    I would like to mention the following ODIs: 1. No 1202 WI v India, 1997, though ball by ball data is not available but Almanack says' India were seemingly heading towards another victory while Ganguly and Dravid were comfortably adding 130 in 29 overs for the second wicket, but then collapsed to defeat: their last eight fell for 46 in nine overs'
    [[
    Yes, the inference is that at 201 for 3, they were very comfortably placed. However there is no clear information on balls available. Let us not forget that even run-a-ball was tough those days.
    Ananth
    : ]]

    2. 1996 WC Semi final, No 1082 WI vs Aus:Alamanack reports'West Indies pulled off an extraordinary defeat, losing eight wickets in the final 50 minutes. After 41 overs, they were 165 for two, needing 43 from the last nine'
    [[
    Based on the numbers you have given and verified by me, the TRF-S was a very low 0.4263 (0.299 & 0.679). There is no doubt that West Indies threw away an easy win. Or should we say the Warne magic at work. Who know what would have happened in the Final.
    An interesting sideview is that this is one of three WC Semi-final heists by Australia. And they won two of these WCs. 1987 Sf when they won by 18 runs against Pakistan, the referred 1997 Sf and the 1999 Sf against South Africa.
    Ananth
    : ]]

  • 07sanjeewakaru on September 28, 2013, 20:53 GMT

    How about , Singer Triangular Series - 2nd match in 2000 SL vs SA where SA were 0/150 chasing SL's 249 and lost by 37(ODI 1604)
    [[
    Couple of nice matches, Sanjeewa.
    In 1604, the best situation was 150 for 0 (100 in 111 balls). Remarkably this situation is almost the same as the mythical situations I have used to work out the examples. The TRF-S was 0.524 (0.400 & 0.772). I would go to the extent of saying that this match should find a way into the featured matches. If only I were able to access my article for editing. Chandana was the unlikely hero.
    Ananth
    : ]]

    SL vs AUS 2nd ODI at Dambulla,2004 where Aus were 1/148 ,2/170 and 3/190 chasing SL's 244 and Vass bowled a wonderful last over against Symonds and Bevan (ODI 2103). Thanks million!.
    [[
    Remarkably similar match. The best sitution was at 190 for 3 (56 in 58 balls). The TRF-S was 0.566 (0.421 & 0.856). This time Sri Lnakans contained the Australians.
    Ananth
    : ]]

  • on September 28, 2013, 14:01 GMT

    Hey Ananth, Nice article. What about this match. http://www.espncricinfo.com/ci/engine/current/match/66192.html Chasing 262 SA were 192/1 in 37 overs when Gibbs retired. He was not out and was replaced by Rhodes. Soon they were 194/2 and 194/3 at the end of 39 overs. 211/3 at the end of 43 and they lost by 10 runs.
    [[
    ODi match1886. An excellent choice, Navin.
    At 194/1, the equation was 69 in 70 balls. The final values are 0.320(TRF-W), 0.870(TRF-B) and 0.503(TRF-S). So this match just missed the cut-off. Again the near run-a-ball requirement is the factor which derailed the chase.
    Ananth
    : ]]

  • kaushik_243 on September 28, 2013, 9:51 GMT

    what about india.vs australia hyderabad 2009 match ,where sachin scored 175 .we needed 50 at run a ball with 6 wickets in hand before losing it by 2 runs.where does that rank in the analysis .but great work overall .keep it up sir :)
    [[
    Match #2923.
    The best situation was at 299 for 4, needing 51 in 45 balls with 6 wickets in hand. The TRF-W at that time was 0.345. The TRF-B was 1.068. So the TRF-S is 0.586, just outside the cut-off. But worth looking at this match because of the ease of task ahead, especially as the wickets were concerned. But the killer was, as usual, the pressure of scoring at over run-a-minute.
    Just worked out that the situation was even better at 332 for 6. Only 18 in 18 balls and into the last 10 overs. The TRF values are 0.215, 1.108 and 0.513. So a still lower value.
    Ananth
    : ]]

  • KKTripathi on October 12, 2013, 9:07 GMT

    The analysis should be split by pre-t20 and post t-20 era. the weights should vary.

  • on October 7, 2013, 12:14 GMT

    Sir, can you please present your analysis for India vs Pakistan, played on 13th March 2004 at Karachi. During that time, this match was a record for the highest team totals. Ind scored 349. Pak reached 344 while chasing.

  • AnanthNarayanan on September 30, 2013, 17:37 GMT

    [[
    I have created another document file with details of all matches in which the TRF-S values below 0.90. This includes multiple occurrences within the same match. To download/view the document, please CLICK HERE.
    Ananth
    : ]]

  • ThinkingCricket on September 30, 2013, 14:48 GMT

    Nice article, but I think some of your assumptions are strongly weighted towards situations that are not such huge-turnarounds. In particular you are overweighting turnarounds that happen with low sample sizes of 1 or 2 overs. Two or three tail-enders falling with 12 runs required isn't really an epic collapse, because resource percentages often miss the point that one bad over isn't a huge turnaround. Common sense can tell us that 177/7 chasing 190 isn't a particularly comfortable situation and I wouldn't expect that to result in a win more than 80 or 90% of the time, so for that to be one of the biggest turnarounds in history seems like an anomalous result. Intuitively I think anyone would agree that 157/1 chasing 196 (bottom match) is a far more spectacular collapse than 152/4 chasing 190, or 177/7 chasing 190 or 189/9 chasing 190 (the last of which is literally saying that No.11 being bowled with two runs required is a bigger surprise than a collapse from 157/1 to 196 all out).
    [[
    I have no problems with any of your statements. This is the first attempt and the next try will be based on additional insights such as yours. If you see my comments on the respective weights, it is quite possible that 75/25 is probably the more correct one. That would have over-valued the wickets-resources in hand. Having said that, 196 for 8, with 1 run needed is still in many ways an excellent position. A no ball, wide, bye, leg-bye, hit-and-run, to-the-keeper-and-run, lucky hit through slips, misfield: I could go on. 157 for 1 still required that the good quality effort had to be sustained for another 8-10 overs.
    But looking at it from a subjective angle, 157 for 1 would be the position most captains would have liked to be in. Let us see how to recognize these situations. It could be even that a position of 157 for 1, 30 needed in 87 balls, should completely discount the balls-available factor. The TRF-W works out to 0.250. This is lower than the current TRF-S of 0.304.
    But very good food for thought. Many thanks.
    Ananth
    : ]]

  • on September 30, 2013, 9:55 GMT

    Hi Ananth, Great work!! I was wondering if something like D/L method to estimate the requirement at a given stage not be a more standard way? If you do not like D/L method, take the other variant. But basically, you can say that given the situation (while chasing, e.g., 120/1 in 20 overs), you can also compute the par score and see what's the difference between the current score and par score to decide how much ahead/behind the chasing team is. That way, you will not have to make choices and defend them. In fact, where I was think of this was the analysis of how valuable the runs scored by a batsman are (in an ODI situation)? If you leave your team better off than when you arrive, your runs are valuable, else they count for much. What do you think of this idea?
    [[
    NM, there is one difference between D/L and my working. There is nothing in my methods which are like Rocket Science. It would take anyone 5 minutes to understand the method and apply it in real life.
    D/L is far more complicated and the purpose is different. They determine par scores at any time. I determine the winning chances. If the TRF-S is 0.3, what I really infer is that the chances of winning are 3.3 times that of the other team. If the TRF-R was 2.50, the chances of winning are 40% of those of the other side.
    As far as defending numbers are concerned, when have the learned duo defended their numbers. Those are taken for granted. I have a basis for every number. I may also very well do some work to determine the expected RpO also.
    Ananth
    : ]]

  • Anand_S on September 29, 2013, 14:28 GMT

    Ananth I was refering to match #2357. I made an error, I read 117-3 as 117-1. I take my question back now. Also my apologies for not noting your mention about the match in which Malinga and Langeveldt did magic for their teams. The other match India vs NZL I asked about was match # 1933. Chasing 200 INdia was 182-3 in 42 overs needing 18 to win with 7 wkts and 8 overs. They ended up winning by 1 wicket and one ball to spare. Had they lost where would this be in your analysis?
    [[
    This was some match. I think the best situation was at 141 for 1, needing 59 in 105 balls. The TRF-S was 0.397 (0.348 & 0.494), way below the cut-off point. Because of the number of such situations I am certain it would have been featured. Come to think of it, what was the situation at 197 for 7. The TRF-S was 0.1668 (0.095 & 0.3087).
    Ananth
    : ]]

  • on September 29, 2013, 10:57 GMT

    I would like to mention the following ODIs: 1. No 1202 WI v India, 1997, though ball by ball data is not available but Almanack says' India were seemingly heading towards another victory while Ganguly and Dravid were comfortably adding 130 in 29 overs for the second wicket, but then collapsed to defeat: their last eight fell for 46 in nine overs'
    [[
    Yes, the inference is that at 201 for 3, they were very comfortably placed. However there is no clear information on balls available. Let us not forget that even run-a-ball was tough those days.
    Ananth
    : ]]

    2. 1996 WC Semi final, No 1082 WI vs Aus:Alamanack reports'West Indies pulled off an extraordinary defeat, losing eight wickets in the final 50 minutes. After 41 overs, they were 165 for two, needing 43 from the last nine'
    [[
    Based on the numbers you have given and verified by me, the TRF-S was a very low 0.4263 (0.299 & 0.679). There is no doubt that West Indies threw away an easy win. Or should we say the Warne magic at work. Who know what would have happened in the Final.
    An interesting sideview is that this is one of three WC Semi-final heists by Australia. And they won two of these WCs. 1987 Sf when they won by 18 runs against Pakistan, the referred 1997 Sf and the 1999 Sf against South Africa.
    Ananth
    : ]]

  • 07sanjeewakaru on September 28, 2013, 20:53 GMT

    How about , Singer Triangular Series - 2nd match in 2000 SL vs SA where SA were 0/150 chasing SL's 249 and lost by 37(ODI 1604)
    [[
    Couple of nice matches, Sanjeewa.
    In 1604, the best situation was 150 for 0 (100 in 111 balls). Remarkably this situation is almost the same as the mythical situations I have used to work out the examples. The TRF-S was 0.524 (0.400 & 0.772). I would go to the extent of saying that this match should find a way into the featured matches. If only I were able to access my article for editing. Chandana was the unlikely hero.
    Ananth
    : ]]

    SL vs AUS 2nd ODI at Dambulla,2004 where Aus were 1/148 ,2/170 and 3/190 chasing SL's 244 and Vass bowled a wonderful last over against Symonds and Bevan (ODI 2103). Thanks million!.
    [[
    Remarkably similar match. The best sitution was at 190 for 3 (56 in 58 balls). The TRF-S was 0.566 (0.421 & 0.856). This time Sri Lnakans contained the Australians.
    Ananth
    : ]]

  • on September 28, 2013, 14:01 GMT

    Hey Ananth, Nice article. What about this match. http://www.espncricinfo.com/ci/engine/current/match/66192.html Chasing 262 SA were 192/1 in 37 overs when Gibbs retired. He was not out and was replaced by Rhodes. Soon they were 194/2 and 194/3 at the end of 39 overs. 211/3 at the end of 43 and they lost by 10 runs.
    [[
    ODi match1886. An excellent choice, Navin.
    At 194/1, the equation was 69 in 70 balls. The final values are 0.320(TRF-W), 0.870(TRF-B) and 0.503(TRF-S). So this match just missed the cut-off. Again the near run-a-ball requirement is the factor which derailed the chase.
    Ananth
    : ]]

  • kaushik_243 on September 28, 2013, 9:51 GMT

    what about india.vs australia hyderabad 2009 match ,where sachin scored 175 .we needed 50 at run a ball with 6 wickets in hand before losing it by 2 runs.where does that rank in the analysis .but great work overall .keep it up sir :)
    [[
    Match #2923.
    The best situation was at 299 for 4, needing 51 in 45 balls with 6 wickets in hand. The TRF-W at that time was 0.345. The TRF-B was 1.068. So the TRF-S is 0.586, just outside the cut-off. But worth looking at this match because of the ease of task ahead, especially as the wickets were concerned. But the killer was, as usual, the pressure of scoring at over run-a-minute.
    Just worked out that the situation was even better at 332 for 6. Only 18 in 18 balls and into the last 10 overs. The TRF values are 0.215, 1.108 and 0.513. So a still lower value.
    Ananth
    : ]]

  • KKTripathi on October 12, 2013, 9:07 GMT

    The analysis should be split by pre-t20 and post t-20 era. the weights should vary.

  • on October 7, 2013, 12:14 GMT

    Sir, can you please present your analysis for India vs Pakistan, played on 13th March 2004 at Karachi. During that time, this match was a record for the highest team totals. Ind scored 349. Pak reached 344 while chasing.

  • vinjoy on October 4, 2013, 3:55 GMT

    Ananth, how about the India-SL 2nd ODI on 12 August 1993 (See http://www.espncricinfo.com/ci/engine/match/64434.html). Chasing 205, India were cruising smoothly at 152 when we lost Azhar and team folded by losing 8 wickets within 44 runs.
    [[
    The query on this match (no 834) has already been anaswered in my reply to Sanjeewa. You have to click on the Comments: 20 at the top to see all the comments.
    Unfortunately no fall-of-wkt-balldata is available. In 833, when Sri Lanka were 161 for 2, we do not know how many balls were available. Similarly in 834, when India were 152 for 2, chasing 205.
    Ananth
    : ]]

  • harshthakor on September 30, 2013, 17:05 GMT

    My most notable were

    1.India v Australia at Madras in the 1987 Reliance Cup when India appeared cruising home before losing by one run.

    2.Australia v.South Africa 1st Carlton and United final game in 1997-98 when Alan Donald snatched a win for the Proteas from the jaws of defeat. by just 6 runs.

    3.the 1996 semi-final at Chandigarh where West Indies lost by 6 runs after needing just 38 runs of 8 overs with 7 wickets in hand.

    4.India v.West Indies in the 1989 Nehru cup where India collapsed after looking to cruise home.

    5.South Africa v .Pakistan in the 2000 league game at Sharjah where the Proteas middle order collapsed to give Pak a sensational win.

    6.India v Pakistan in the 1985 Rothmans cup where India defended a target of a mere 125 runs.

    7.Sri Lanka v.India in 1998 final of tornament in Sri Lanka where chasing a target of above 300 the Lankans seemed to be cruising home

    8.1996 singer cup final in Singapore between Sri Lanka and Pakistan
    [[
    Harsh, some of these have already been covered in response to earlier comments. Pl also see my featured comment providing the link for a more extensive file.
    Ananth
    : ]]

  • jokerbala on September 30, 2013, 16:19 GMT

    One Glaring omission I can see is the India Vs South Africa match of the champions league 2002.(http://www.espncricinfo.com/ci/engine/current/match/66192.html)l India made some 260 and Saffers were 190 for 1 before choking at the home stretch. Or perhaps did not meet your criteria
    [[
    Match no 1886.
    At various points during the chase, the TRF-S was between 0.5 and 1.0. The best situation was at 194 for 1, when the value was o,503. Hence it just missed the cut. It is quite possible that when I make the tweak for the low-wicket-fall situations (see featured comment of ThinkingCricket) this would get in.
    Ananth
    : ]]

  • Ukri82 on September 30, 2013, 15:10 GMT

    I see that you spent lot of efforts to explain the methodology. Thanks for that. I struggled a bit to grasp the third point, but finally I managed to get it, I think. To test myself, I would ask another question. "The only reason it does not get into this list is because at 401 for 6, chasing 414, the balls situation, 14 in 10, was not that favourable to Sri Lanka. And it proved difficult. They could not make those runs. " Well in this situation, Sri Lanka managed to make 401 runs in 48.2 overs, so the remaining 14 in 10 shouldn't be difficult, right? In other words, shouldn't you consider the current run-rate to define the difficulty of the required run-rate?
    [[
    Unni, my initial working was based on the current scoring rate. However that did not work well in low-scoring matches. Maybe a better idea is to take the Expected RpO or Actual RpO whichever is higher.
    Ananth
    : ]]

  • Ram2212 on September 30, 2013, 5:57 GMT

    Hi Ananth, As usual, amazing work. Thanks for making us see from a different perspective. I am eager to know how a particular match ranks as per your model - Semi Finals between WI and Australia in 1996 World Cup, when WI who were cruising so well, suddenly were on the losing side. Thanks.
    [[
    Ram, this was match no 1082.
    This has been covered in my response to Pawan's comment, which is a featured one.
    Ananth
    : ]]

  • indianpunter on September 29, 2013, 23:43 GMT

    Ananth, this is incredible stuff. it is simply amazing that u think of such situations. would u be able to do a "best batsmen in run chases" list? I suspect that it will show up Rahul dravid as a poor finisher as i cant even remember 1 of his 12 ODI hundreds in a chase India won. Just a thought. Keep the magical stuff coming.
    [[
    Yes, now I think it would be possible to relate such matches to individual performances rather than team efforts. Let us not put down Dravid. He was not the designated finisher at any time. He was the anchor. Finishers have different tasks to fulfill. Becan, Hussey and Dhoni have never been anchors in international matches.
    Ananth
    : ]]

  • 07sanjeewakaru on September 29, 2013, 8:45 GMT

    @ by Anand_S- How close would Malinga's 4 in 4 been had SL won? I think he mentions that 4 wicket in a trot by Malinga @ 2007 wordcup super8 match in Guyana vs SA. If Malinga or Vass had taken that last wicket it should've been the first on this list by a long way.(ODI 2556) Thanks millions for constant feedback and humbleness! Longlive Sir!
    [[
    Yes, I think so. I have already covered the match.
    Ananth
    : ]]

  • calcu on September 29, 2013, 8:37 GMT

    What about the tied match between England and India in the 2011WC? I remember that England were in a very good position of winning when Strauss was batting before collapsing for a tie.
    [[
    Match no 3110. An excellent choice
    At various times in the chase, The TRF-S was between 0.5 and 1.0. The best situation was at 281 for 2 when the TRF-S value was 0.543 (0.253 & 1.123). The problem was 58 in 44.
    Ananth
    : ]]

  • on September 29, 2013, 7:07 GMT

    Hi Ananth Thanks once again for an amazing article. One small question, following up on Anand-S, where would this match have been if Sri Lanka could have got one more wicket? Lasith Malinga took 4 wickets in 4 balls in this match- http://www.espncricinfo.com/wc2007/engine/match/247482.html
    [[
    Yes, I understand Malinga's spell now. My apologies to Anand. I thought 4 for 4 was the final analysis.
    The TRF-S was 0.3966 (0.365 & 0.460) at 160 for 2 (53 needed in 108 balls). If Sri Lanka had taken one more wicket this would have been featured without any doubt.
    Just realized that Langeveldt captured 3 wickets in 5 balls in the Sri Lankan first innings.
    Ananth
    : ]]

  • Anand_S on September 29, 2013, 5:09 GMT

    Ananth:

    Amazing analyses as one can always expect from you. I tried computing some factors for these matches, but looked like they will fall short of your cut off, but still like to know your thoughts..

    (i) India chasing 271 against Aus in 1987 Reliance cup match #3. Needed 15 to win off last 4 with 4 wickets in hand and lost by one run. At one stage were cruitsing at 205-2 or something like that with an asking rate of 5 or there abouts. I dont have the score card to back up this stuff but a youtube video is available to confirm this.
    [[
    Match no 453. No Fall-of-wkt-balldata available. From 207 for 2 India collapsed to 269 all out. Probably a good candidate.
    Ananth
    : ]]

    (ii) How close would Malinga's 4 in 4 been had SL won?
    [[
    Malinga does not have a 4 for 4 spell. Only Dilshan had one. And that was a 139 run win.
    Ananth
    : ]]

    (iii) SL vs Pak in 1996 at Singapore. Jayasurya smashed a 17 ball 50 (my guess is still the fastest 50 in ODIs). SL were 100-1 in 10 overs chasing 212 and bowled out for 172 in 33.5 overs
    [[
    Match no 1093. Again no data is available. Problem is, that despite Jayasuriya's 28-ball-76, Sri Lanka kept on losing wickets at 70, 96, 100, 106 and so on. So they were probably around 0.5 at one of these points.
    Ananth
    : ]]

    (iv) Another game India vs NZL in 2003. NZL scored 199-9. India collapsed from 182-3 to 199-9. How would this match qualify in case of a tie or loss for India? (v) How about Eng vs India 1st ODI 2006?
    [[
    Match no 2357 is nowehere near the qualification criteria. Probably you are reeferring to some other match.
    Ananth
    : ]]

    Anand

  • 07sanjeewakaru on September 28, 2013, 20:21 GMT

    Hei Mr.Ananth,Very interesting analysis.You are right.That last over is the best in the ODI history. I remember 2 ODIs payed in Premadasa SL vs Ind in 1993..very much of this kind;.I think both of them are fit in to this top 10.1st ODI SL found themselves on the loosing end and in the next India were there. Check,ODI 833,and 834....
    [[
    Unfortunately no fall-of-wkt-balldata is available. In 833, when Sri Lanka were 161 for 2, we do not know how many balls were available. Similarly in 834, when India were 152 for 2, chasing 205.
    Ananth
    : ]]

    And how close the 3rd natwest ODI in 2010 to the top 10 if it Australia had won.Eng were 3/185 chasing 212 and collapse to 9/203.
    [[
    Match no 3003 does not even qualify because England won. If they had lost their 10th wicket at 205 or so, this would have comfortably qualified.
    Ananth
    : ]]

    Thanks million,You are brilliant!Love your perseverance and hardworking.It's very inspirational.

  • on September 28, 2013, 16:25 GMT

    Awesome Analysis. Brilliant explanation and an eye opener as far as watching matches. Will be be looking at all matches keeping this aspect of matches in back of mind.

  • on September 28, 2013, 16:25 GMT

    Awesome Analysis. Brilliant explanation and an eye opener as far as watching matches. Will be be looking at all matches keeping this aspect of matches in back of mind.

  • 07sanjeewakaru on September 28, 2013, 20:21 GMT

    Hei Mr.Ananth,Very interesting analysis.You are right.That last over is the best in the ODI history. I remember 2 ODIs payed in Premadasa SL vs Ind in 1993..very much of this kind;.I think both of them are fit in to this top 10.1st ODI SL found themselves on the loosing end and in the next India were there. Check,ODI 833,and 834....
    [[
    Unfortunately no fall-of-wkt-balldata is available. In 833, when Sri Lanka were 161 for 2, we do not know how many balls were available. Similarly in 834, when India were 152 for 2, chasing 205.
    Ananth
    : ]]

    And how close the 3rd natwest ODI in 2010 to the top 10 if it Australia had won.Eng were 3/185 chasing 212 and collapse to 9/203.
    [[
    Match no 3003 does not even qualify because England won. If they had lost their 10th wicket at 205 or so, this would have comfortably qualified.
    Ananth
    : ]]

    Thanks million,You are brilliant!Love your perseverance and hardworking.It's very inspirational.

  • Anand_S on September 29, 2013, 5:09 GMT

    Ananth:

    Amazing analyses as one can always expect from you. I tried computing some factors for these matches, but looked like they will fall short of your cut off, but still like to know your thoughts..

    (i) India chasing 271 against Aus in 1987 Reliance cup match #3. Needed 15 to win off last 4 with 4 wickets in hand and lost by one run. At one stage were cruitsing at 205-2 or something like that with an asking rate of 5 or there abouts. I dont have the score card to back up this stuff but a youtube video is available to confirm this.
    [[
    Match no 453. No Fall-of-wkt-balldata available. From 207 for 2 India collapsed to 269 all out. Probably a good candidate.
    Ananth
    : ]]

    (ii) How close would Malinga's 4 in 4 been had SL won?
    [[
    Malinga does not have a 4 for 4 spell. Only Dilshan had one. And that was a 139 run win.
    Ananth
    : ]]

    (iii) SL vs Pak in 1996 at Singapore. Jayasurya smashed a 17 ball 50 (my guess is still the fastest 50 in ODIs). SL were 100-1 in 10 overs chasing 212 and bowled out for 172 in 33.5 overs
    [[
    Match no 1093. Again no data is available. Problem is, that despite Jayasuriya's 28-ball-76, Sri Lanka kept on losing wickets at 70, 96, 100, 106 and so on. So they were probably around 0.5 at one of these points.
    Ananth
    : ]]

    (iv) Another game India vs NZL in 2003. NZL scored 199-9. India collapsed from 182-3 to 199-9. How would this match qualify in case of a tie or loss for India? (v) How about Eng vs India 1st ODI 2006?
    [[
    Match no 2357 is nowehere near the qualification criteria. Probably you are reeferring to some other match.
    Ananth
    : ]]

    Anand

  • on September 29, 2013, 7:07 GMT

    Hi Ananth Thanks once again for an amazing article. One small question, following up on Anand-S, where would this match have been if Sri Lanka could have got one more wicket? Lasith Malinga took 4 wickets in 4 balls in this match- http://www.espncricinfo.com/wc2007/engine/match/247482.html
    [[
    Yes, I understand Malinga's spell now. My apologies to Anand. I thought 4 for 4 was the final analysis.
    The TRF-S was 0.3966 (0.365 & 0.460) at 160 for 2 (53 needed in 108 balls). If Sri Lanka had taken one more wicket this would have been featured without any doubt.
    Just realized that Langeveldt captured 3 wickets in 5 balls in the Sri Lankan first innings.
    Ananth
    : ]]

  • calcu on September 29, 2013, 8:37 GMT

    What about the tied match between England and India in the 2011WC? I remember that England were in a very good position of winning when Strauss was batting before collapsing for a tie.
    [[
    Match no 3110. An excellent choice
    At various times in the chase, The TRF-S was between 0.5 and 1.0. The best situation was at 281 for 2 when the TRF-S value was 0.543 (0.253 & 1.123). The problem was 58 in 44.
    Ananth
    : ]]

  • 07sanjeewakaru on September 29, 2013, 8:45 GMT

    @ by Anand_S- How close would Malinga's 4 in 4 been had SL won? I think he mentions that 4 wicket in a trot by Malinga @ 2007 wordcup super8 match in Guyana vs SA. If Malinga or Vass had taken that last wicket it should've been the first on this list by a long way.(ODI 2556) Thanks millions for constant feedback and humbleness! Longlive Sir!
    [[
    Yes, I think so. I have already covered the match.
    Ananth
    : ]]

  • indianpunter on September 29, 2013, 23:43 GMT

    Ananth, this is incredible stuff. it is simply amazing that u think of such situations. would u be able to do a "best batsmen in run chases" list? I suspect that it will show up Rahul dravid as a poor finisher as i cant even remember 1 of his 12 ODI hundreds in a chase India won. Just a thought. Keep the magical stuff coming.
    [[
    Yes, now I think it would be possible to relate such matches to individual performances rather than team efforts. Let us not put down Dravid. He was not the designated finisher at any time. He was the anchor. Finishers have different tasks to fulfill. Becan, Hussey and Dhoni have never been anchors in international matches.
    Ananth
    : ]]

  • Ram2212 on September 30, 2013, 5:57 GMT

    Hi Ananth, As usual, amazing work. Thanks for making us see from a different perspective. I am eager to know how a particular match ranks as per your model - Semi Finals between WI and Australia in 1996 World Cup, when WI who were cruising so well, suddenly were on the losing side. Thanks.
    [[
    Ram, this was match no 1082.
    This has been covered in my response to Pawan's comment, which is a featured one.
    Ananth
    : ]]

  • Ukri82 on September 30, 2013, 15:10 GMT

    I see that you spent lot of efforts to explain the methodology. Thanks for that. I struggled a bit to grasp the third point, but finally I managed to get it, I think. To test myself, I would ask another question. "The only reason it does not get into this list is because at 401 for 6, chasing 414, the balls situation, 14 in 10, was not that favourable to Sri Lanka. And it proved difficult. They could not make those runs. " Well in this situation, Sri Lanka managed to make 401 runs in 48.2 overs, so the remaining 14 in 10 shouldn't be difficult, right? In other words, shouldn't you consider the current run-rate to define the difficulty of the required run-rate?
    [[
    Unni, my initial working was based on the current scoring rate. However that did not work well in low-scoring matches. Maybe a better idea is to take the Expected RpO or Actual RpO whichever is higher.
    Ananth
    : ]]

  • jokerbala on September 30, 2013, 16:19 GMT

    One Glaring omission I can see is the India Vs South Africa match of the champions league 2002.(http://www.espncricinfo.com/ci/engine/current/match/66192.html)l India made some 260 and Saffers were 190 for 1 before choking at the home stretch. Or perhaps did not meet your criteria
    [[
    Match no 1886.
    At various points during the chase, the TRF-S was between 0.5 and 1.0. The best situation was at 194 for 1, when the value was o,503. Hence it just missed the cut. It is quite possible that when I make the tweak for the low-wicket-fall situations (see featured comment of ThinkingCricket) this would get in.
    Ananth
    : ]]