April 26, 2014

Chalk and cheese in the same ODI innings

Instances of huge contrasts between batting strike rates or bowling economy rates within the same ODI innings

Wavell Hinds and Runako Morton were both guilty of extreme slow-scoring in ODIs in which their team-mates scored far more quickly © Getty Images

At the end of the Indian innings in the 2014 World Twenty20 final, I did my usual strike-rate comparison between the best and worst batsmen in the team. I can do this calculation in my mind and I worked out that Virat Kohli was striking at around 2.5 times that of Yuvraj Singh. I remembered that Dwayne Bravo was also at a similar level against Marlon Samuels in their semi-final match against Sri Lanka. I was certain that Kohli and Bravo, having scored more runs at 2.5 times the strike rate of the other batsman, would be seething within, despite contrary public utterances, because the matches were lost. On a whim, I looked up the 2012 World Twenty20 final. Samuels' strike rate clocked in at 7.4 times that of Gayle. But all sins are forgotten in a Gangnam dance if the team wins. Then I suddenly realised that I had the material for a nice and easy-to-understand article. I also remembered Mohan's comment on my last article that he loved the simpler ones which he could understand clearly. I reminded myself that I should do these simple anecdotal and low-key analytical articles more often.

I took a major decision. I decided that I would go with the ODI analysis first. I wanted to wet my feet initially on the longer limited-overs format since the T20 format is a sharp and unforgiving format: both on the playing and analytical arenas. The time is so short that one performs or perishes. So my cut-offs as well as metrics in T20s have to be radically different from those of the ODI game. So I have covered the ODI game here. I promise you that the T20 analysis, which will be a far sharper one, will follow, in due course.

The idea is simple. Set reasonable cut-offs for the (batting) innings and (bowling) innspells. Find the best and worst and ratio between both as the HL_Index. Order these on the basis of the Index values. The great thing about the analysis is that anyone can determine the Index value in two minutes flat by perusing any scorecard. The USP of this article is its essential simplicity.

The pleasing aspect of this analysis is that it is a true peer comparison. I agree that there could be early-morning support for the pace bowlers, the drying of pitch and consequent help to spinners, the dew factor in the evening games, the post-rain skid factor and so on. But everything is capsuled into the 200 minutes of an ODI innings. Conditions might change by about 10% but not more. If the opening batsmen need some time to settle down, they also have attacking fields to help in scoring runs. If they take 20 balls to break their duck, they are expected to convert this to a 70-ball-50, not a 30-ball-1. And let us not forget that all comparisons are within a single innings.


But cut-offs have to be set. I cannot let Sunil Narine's (yes, you got the name right) cameo of 6-ball-24 or James Franklin's 8-ball-31 or Brendon McCullum's 9-ball-32 et al disturb the balance of the idea. These are freakish cameos. So I decided on a cut-off of 20 balls. This is essential since I wanted to cross the 20-ball mark to give the batsman some more time to get his innings back on track. We are not looking for low-scoring in 20 balls or so but getting out soon after. On the faster side, there is only one innings below 20 that merits inclusion: Afridi's 18-ball-55. I have checked this innings and made sure that the index for this team innings is below the selection mark.

Twenty balls represents more than 50% of the average innings size of 31 balls. It allows the batsman to settle down and score reasonably quickly to finish with a 100+ strike rate. On the slower side, 20 balls represents sufficiently long batting stint to settle down and score at least at around 33.3%.

They were batting on two pitches a continent apart: Index 20 and above
HL-Index Year ODI Team Res Batsman in zone BPos Score High S/R Batsman struggling BPos Score Low S/R
36.6920021898WinCL Hooper 738 (29)131.0WW HindsOB1 (28) 3.6
33.3320052276ZimCK Coventry 735 (21)166.7BRM TaylorOB1 (20) 5.0
32.0020062422WinDR Smith 830 (30)100.0RS Morton 30 (31) 3.1
30.4419991524IndWRobin Singh 745 (34)132.4SR TendulkarOB1 (23) 4.3
28.2920133401SafQ de KockOB27 (21)128.6F Behardien 61 (22) 4.5
27.161990 610PakWasim Akram 786 (76)113.2Imran Khan 51 (24) 4.2
25.2819961053UaeJA Samarasekera 829 (39) 74.4V Mehra 41 (34) 2.9
25.001984 274SlkWPA de Silva 750 (54) 92.6S WettimunyOB1 (27) 3.7
24.1519981335BngHasibul Hossain1021 (20)105.0Athar Ali KhanOB0 (22) 4.3
23.921983 182NzlBL Cairns 852 (25)208.0JV Coney 52 (23) 8.7
23.9019981335IndWSR TendulkarOB33 (29)113.8R Dravid 31 (21) 4.8
23.5020031973WinRR Sarwan 447 (44)106.8BC Lara 31 (22) 4.5
22.4420021807PakWAbdul Razzaq 746 (41)112.2Inzamam-ul-Haq 41 (20) 5.0
22.421983 171EngWAJ Lamb 4108 (106)101.9G FowlerOB0 (21) 4.5
22.3720001657ZimA Flower 451 (57) 89.5SV Carlisle 31 (25) 4.0
21.981991 685PakWSaleem Malik 487 (95) 91.6Aamer SohailOB1 (24) 4.2
21.5520113168SlkWDPMD JayawardeneOB79 (77)102.6AD Mathews 61 (21) 4.8
20.8020001591SafWL Klusener 852 (50)104.0G KirstenOB1 (20) 5.0
20.211994 938WinBC Lara 332 (38) 84.2PV Simmons 50 (23) 4.2
20.1220031951IndHarbhajan Singh 928 (32) 87.5R Dravid 41 (23) 4.3

The table represents the top entries while the potted scores below are for ten matches involving the major teams. While it is of interest to know that Samarasekera of UAE scored 25 times faster than Mehra, it is unlikely to be of further interest to anyone, including the UAE cricket followers.

At the turn of the century West Indies tried many opening batsmen. No one was really good enough to hold their place for a long time. This is reflected in a couple of entries at the top. Wavell Hinds was one and Runako Morton was another. Hinds laboured to 1 in 28 and Morton to 0 in 31 (taken as 1 in 32, for calculation purposes). Their scoring rates were abysmally low 3 or so. Carl Hooper scored at 131 and outscored Hinds by over 36 times. Smith scored at 100 and outscored by a mere 31 times. There is a personal take on the Hinds-Hooper match, chronicled later.

Robin Singh outscored his more illustrious team-mate Sachin Tendulkar by 30 times. Imagine Wasim Akram outscoring the normally quick-scoring Imran Khan by 27 times. And in another match, Tendulkar, despite opening the batting, outscored his second-wicket partner, Rahul Dravid, by 24 times. But it can be seen that many of the slow scoring batsmen have opened. Allan Lamb's performance is praiseworthy since he has scored a hundred at better than run-a-ball and outscored Graeme Fowler, who did not open his account in 21 balls, by a margin of 21 times.

It can be seen that more matches are lost than won in these matches. It is clear that the quicker scoring batsman has not always been able to undo damage done by the slow-scoring batsmen. The top teams seem to find ways of undoing the damage more than the teams in the lower rung.

Potted Scores

ODI # 1898. India vs West Indies. India won by 3 wickets.
Played on 21 November 2002 at Jodhpur. Mom: Agarkar A.B.
West Indies: 201 all out in 46.3 overs
   WW Hinds             1  in  28 (  3.6)
   CL Hooper           38  in  29 (131.0)
India: 202 for 7 wkt(s) in 46.2 overs

The Jodhpur match, the top one in the table, was one of the matches in which I did television work with Doordarshan. Just out of interest I looked back at my notes, still preserved carefully, for this match. I have written there "Hooper 36 times faster than Hinds. Discuss with Greenidge". I remember talking to Greenidge about it and Greenidge's discomfiture at a West Indian opening batsman dawdling. He also mentioned that Javagal Srinath was good but Ajit Agarkar and Sanjay Bangar were only medium pacers. His grouse was at Hinds getting out at 1. Little did I realise that one day this match would be the leading entry in a list.

What followed was even more painful to watch. Samuels scored 3 in 28 balls. So the 2 and 3 batsmen accumulated an incredible tally of 4 in 56 balls: less than half a run per over. West Indies lost the match and it is certain that Hinds and Samuels have to be considered responsible. Not to forget Ramnaresh Sarwan's contribution: 14 in 38 balls, making a total of 18 in 84 balls.

ODI # 2422. Australia vs West Indies. Australia won by 127 runs. 
Played on 24 September 2006 at Kuala Lumpur. Mom: Clarke.
Australia: 240 for 6 wkt(s) in 50.0 overs
West Indies: 113 all out in 34.2 overs
   RS Morton            0  in  31 (  3.1)
   DR Smith            30  in  30 (100.0)

What does one say of Morton? 31 balls to score 0 and then getting out. Granted, he walked in at 0 for 1 with Gayle getting out first ball. Granted, the bowling was top class: Glenn McGrath, Brett Lee, Nathan Bracken and Shane Watson. But surely a forgettable performance, resulting in a big loss. It is possible that even if Morton had scored a few more runs, West Indies would have lost.

ODI # 1524. India vs New Zealand.  India won by 14 runs.
Played on 11 November 1999 at Gwalior. Mom: Ganguly S.C.
India: 261 for 5 wkt(s) in 50.0 overs
   SR Tendulkar         1  in  23 (  4.3)
   Robin Singh         45* in  34 (132.4)
New Zealand: 247 for 8 wkt(s) in 50.0 overs

Tendulkar, opening, scored 1 in 23 and Robin Singh, coming in at 7, scored at 132. A factor of 30. However Sourav Ganguly scored a majestic run-a-ball 150-plus and India won well.

ODI # 3401. Sri Lanka vs South Africa. Sri Lanka won by 128 runs. 
Played on 31 July 2013 at Colombo. Mom: TM Dilshan.
Sri Lanka: 307 for 4 wkt(s) in 50.0 overs
South Africa: 179 all out in 43.5 overs
   Q de Kock           27  in  21 (128.6)
   F Behardien          1  in  22 (  4.5)

This was a recent match. Quinton de Kock, free and flowing in the opening position, scored at 128. Farhaan Behardien, coming in at 6, scored at below 5. Surely South Africa could do better than Behardien. The match was lost anyhow.

ODI # 610. Australia vs Pakistan. Australia won by 7 wickets. 
Played on 23 February 1990 at M.C.G. Mom: Not awarded.
Pakistan: 162 all out in 47.5 overs
   Imran Khan           1  in  24 (  4.2)
   Wasim Akram         86  in  76 (113.2)
Australia: 163 for 3 wkt(s) in 45.5 overs

The mercurial Imran Khan scored at just over 4 and Akram at 113. Pakistan lost comfortably. This was a peculiar match. Ijaz Ahmed scored a 34-ball-7 and Javed Miandad a 26-ball-2. So no one was comfortable, barring Wasim.

ODI # 274. Sri Lanka vs New Zealand. Sri Lanka won by 4 wickets. 
Played on 3 November 1984 at Colombo. Mom: de Silva P.A.
New Zealand: 171 for 6 wkt(s) in 45.0 overs
Sri Lanka: 174 for 6 wkt(s) in 39.4 overs
   S Wettimuny          1  in  27 (  3.7)
   PA de Silva         50* in  54 ( 92.6)

Sidath Wettimuny's dawdle at the beginning was offset by Aravinda de Silva's brilliant innings resulting in a win. Maybe the low target prompted Wettimuny to see through the fast bowlers. So we cannot really blame him.

ODI # 182. Australia vs New Zealand.  Australia won by 149 runs.
Played on 13 February 1983 at M.C.G. Mom: Hughes K.J.
Australia: 302 for 8 wkt(s) in 50.0 overs
New Zealand: 153 all out in 39.5 overs
   JV Coney             2  in  23 (  8.7)
   BL Cairns           52  in  25 (208.0)

This is chalk with a vengeance and cheese likewise. One batsman scores at 8 and the other at 200+. At 44 for 6, Lance Cairns threw his bat around, and connected. It did not matter. Neither innings could save New Zealand. Jeff Crowe scored a 48-ball-27 and Chatfield, at the end, a 36-ball-10. The bowling was awesome: Dennis Lillee, Rodney Hogg and Geoff Lawson.

ODI # 1335. India vs Bangladesh.
Played on 25 May 1998 at Mumbai. India won by 5 wickets. Mom: Kumble A.
Bangladesh: 115 all out in 36.3 overs
India: 116 for 5 wkt(s) in 29.2 overs
   SR Tendulkar        33  in  29 (113.8)
   R Dravid             1  in  21 (  4.8)

Now Tendulkar is on the other side. He blazes away and Dravid goes into his shell. The scoring rates speak for themselves. But let us not forget that the target was only 116. And Ajay Jadeja's 47-ball-17 led to a laboured win.

ODI # 1973. Sri Lanka vs West Indies. Sri Lanka won by 6 runs. 
Played on 28 February 2003 at Cape Town. Mom: Vaas WPUJC.
Sri Lanka: 228 for 6 wkt(s) in 50.0 overs
West Indies: 222 for 9 wkt(s) in 50.0 overs
   BC Lara              1  in  22 (  4.5)
   RR Sarwan           47* in  44 (106.8)

No one seems to be exempt from this malaise. One can never accuse Brian Lara of scoring slowly. But in this match against Sri Lanka, he dawdled and might even have caused West Indies the match. He scored a single in 22 balls as against Sarwan's fluent 44-ball-47. That too in the World Cup with possibly a place in the next round at stake.

ODI # 1807. Pakistan vs West Indies. Pakistan won by 4 wickets.
Played on 14 February 2002 at Sharjah. Mom: Abdul Razzaq.
West Indies: 190 all out in 48.3 overs
Pakistan: 193 for 6 wkt(s) in 46.1 overs
   Inzamam-ul-Haq       1  in  20 (  5.0)
   Abdul Razzaq        46* in  41 (112.2)

And the malaise runs wide. Inzamam-ul-Haq outscored 22 times by Abdul Razzaq. But the target was a low one and Inzamam's dawdle did not cost the match.


For the bowlers I decided that I would consider only the bowling accuracy for comparison. The strike rates do not mean much and bowlers who come at the end and capture couple of wickets in an over will cause havoc. The bowling accuracy, on the other hand, is a stable measure. The cut-offs here are dicey. After a lot of deliberation I have decided to go with two cut-offs. 5 overs for the accurate bowlers and 3 overs for the expensive bowlers. The reason is simple. I want the accurate bowlers to maintain their accuracy for an extra over or two. 5-0-10-0 is tougher to achieve than 4-0-8-0 or 3-0-5-0. On the other hand the expensive bowlers can easily concede tons of runs in 3 overs: (e-g) Ravi Rampaul 64 in 3, Sreesanth 48 in 3, Rangana Herath 43 in 3 and so on. Now on to the tables.

They were bowling on two pitches a continent apart: Index 7 and above
HL-Index Year ODI Team Res Bowler in zone Analysis Low RpO Bowler struggling Analysis High RpO
14.2920031990KenAY Karim8.2-6-7-3 0.84MA Suji3.0-0-36-012.00
13.0019991512WinWCEL Ambrose10.0-5-5-1 0.50SL Campbell8.0-0-52-2 6.50
11.731986 406PakWWasim Akram7.2-4-4-2 0.55Manzoor Elahi5.0-0-32-1 6.40
10.831981 104AusWDK Lillee5.0-2-3-1 0.60LS Pascoe4.0-0-26-1 6.50
10.371992 782WinWPV Simmons10.0-8-3-4 0.30KCG Benjamin9.0-1-28-2 3.11
9.751989 558IndN Kapil Dev7.0-4-4-0 0.57K Srikkanth4.4-1-26-3 5.57
9.5320011781ZimTR Gripper7.0-4-6-0 0.86HK Olonga6.0-0-49-0 8.17
9.0720092860KenPJ Ongondo7.0-3-9-0 1.29NN Odhiambo3.0-0-35-011.67
8.691984 268AusCG Rackemann8.0-4-7-3 0.88JN Maguire5.0-0-38-0 7.60
8.2120133389WinJO Holder10.0-4-13-4 1.30SP Narine3.0-0-32-010.67
7.9620021826SlkM Muralitharan10.0-3-9-5 0.90WPUJC Vaas6.0-0-43-0 7.17
7.8820113222PakWSaeed Ajmal7.0-4-6-2 0.86Shahid Afridi4.0-0-27-0 6.75
7.861992 726EngDA Reeve5.0-3-2-1 0.40PAJ DeFreitas7.0-1-22-2 3.14
7.801975 24IndWBS Bedi12.0-8-6-1 0.50M Amarnath10.0-0-39-2 3.90
7.7320072539IreWAC Botha8.0-4-5-2 0.62KJ O'Brien6.0-0-29-1 4.83
7.6720072640KenWTM Odoyo7.0-3-7-3 1.00NN Odhiambo3.0-0-23-0 7.67
7.5820021830SlkWWPUJC Vaas7.0-2-8-2 1.14ST Jayasuriya3.0-0-26-0 8.67
7.5020092876IreWAR Cusack5.0-2-3-0 0.60AR White4.0-0-18-1 4.50
7.2420001558SafWSM Pollock8.0-4-7-3 0.88N Boje3.0-0-19-0 6.33
7.2220052264SlkWMF Maharoof10.0-5-9-3 0.90UDU Chandana10.0-0-65-1 6.50
7.201984 237WinWMD Marshall6.0-4-5-1 0.83WW Daniel10.0-2-60-2 6.00
7.181973 9WinWLR Gibbs11.0-4-12-1 1.09KD Boyce6.0-0-47-0 7.83
7.1420082766ZimWRW Price10.0-5-7-0 0.70C Zhuwawo3.0-0-15-0 5.00
7.1420042165IndHarbhajan Singh10.0-2-14-2 1.40AB Agarkar6.0-0-60-110.00
Karim-Suji pair resulted in an index value of just above 14. The match was lost, as happened with many of the lesser teams.

Now look at what Curtly Ambrose did. He completed his innings spell with an RpO value of 0.5. The part-time bowling of Sherwin Campbell had an RpO value of 6.50. It could as well have been Nixon McLean who conceded 5 runs per over. We now come to Wasim Akram's way-out innspell of 7.2-4-4-2 compared with Manzoor Elahi's 5-0-32-1. Then comes Lillee's five-over burst conceding 3 runs against Len Pascoe's 26 runs in 4 overs. The two bowling efforts of Lillee and Pascoe disguise the match-winning effort by Greg Chappell who had figures of 9.5-5-15-5.

The next entry is, arguably, the most incredible one. During the 1992 World Cup, Phil Simmons bowled, what I consider one of the most devastating spells ever, against Pakistan. He finished with the unbelievable figures of 10-8-3-4. All four were top-order dismissals. This is the only instance of eight maiden overs in a ten-over spell. Note Bishan Bedi's bowling figures, albeit against East Africa. He shares the record of eight maidens with Simmons, although in 12 overs. Kenny Benjamin's 9 over spell at 3.11 was over 10 times more expensive than Simmons'. Pakistan had no answer to the extraordinary bowling effort of Simmons and lost by a big margin. Incidentally Jimmy Adams had figures of 4-2-2-1.

Kapil Dev's RpO was nearly 10 times better that of Kris Srikkanth. But that was certainly in vain as West Indies won quite comfortably. This is the first match, involving major teams, which has been featured here which was lost.

There is a missing entry which deserves special mention. It did not qualify because Courtney Walsh bowled 4.3 overs. This was the 1986 match in Sharjah between West Indies and Sri Lanka. How can anyone ever understand Walsh's analysis: 4.3-3-1-5? The nearest to this piece of magic was Herath's spell in the World T20 against New Zealand. With the type of bowling accuracy of Walsh, a figure of 0.22, almost any other spell would form the other half of the HL-Index pair. It fell to Malcolm Marshall's fairly accurate spell of 5-1-16-1 to complete the pair. The index works out to 14.4.

It can be seen that most of the matches have been won by the bowling teams. Hence I will not be presenting any potted scores.

The bowling side of this analysis leaves very little room for further nuanced looks. First the index values have a much narrower range than the batting index values. Look at the highest index values: 36+ and 14+. Possibly because the bowling performances are limited to 5/6 bowlers in an innings. Also most of the matches involving the A-teams have resulted in wins for the considered bowling combinations.

The reason could very well be that the bowling stint of the top bowler normally represents a fifth of the team effort and if that is outstanding, say like Simmons' 10-8-3-4, that effort straightaway puts the team on the ascendancy. And this spell is also indicative of a track which is somewhat helpful to the bowlers. The combination of these factors normally leads to wins. Of course, in a combination like Harbhajan Singh's 10-2-14-2 and Agarkar's 6-0-60-1, Agarkar's nightmare effort effectively cancelled Harbhajan's spell and the team lost. But the instances are far and few in between.

Incidentally for the 6874 innings, the average Batting HT-Index value is 2.59. The average Bowling HT-Index value is 2.14. These are the average spreads between the best and worst performances. There seems to be a more balanced distribution amongst the Bowling index values. There are 2638 Bowling index values above the average while there are only 2127 Batting index values above the average, even though the high Batting index values are much higher.

To download/view the part list of the HT_Index tables, 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