Managing the Overhaul: Future ODI Squad Planning for South Africa & Pakistan

12th August, 2019.



Following the World Cup, a number of countries who failed to qualify from the group stage have parted company with their coaching staff.  Two of the most notable examples are Mickey Arthur (Pakistan) and Ottis Gibson (South Africa) with the latter's team in particular falling short of many people's expectations.  I say many people, because the performances of the teams didn't particularly fall short of my own expectations, which I've mentioned before and were driven on historical pre-requisites from successful World Cup teams in recent history.

I've written about these before, but just to recap, historical data - available, of course, prior to the World Cup - illustrates that every single one of the last five winning World Cup teams has an overall strike rate at least 15% greater than the tournament average strike rate (Mean deviation 15% or greater), as shown below:-

Year

Winner

Tournament Batting SR

Winning Team Batting SR

Winning Team SR Mean Deviation






2015

Australia

88.98

109.80

1.23

2011

India

78.39

91.19

1.16

2007

Australia

75.88

102.35

1.35

2003

Australia

72.64

83.55

1.15

1999

Australia

64.85

75.16

1.16


Across the 2019 World Cup, the overall tournament batting strike rate was 88.15 - marginally lower than the 2015 tournament - and looking at the numbers above, this was the first tournament of these last six to have a lower (albeit extremely marginally) overall strike rate than the previous edition of the World Cup.  

Bearing in mind the general trend of increasing strike rates in World Cups, and indeed, in cricket in general, it's reasonable to assume that the overall tournament strike rate would be at least around the high 80s, and therefore, continuing this trend would mean that the winning team batting strike rate would probably be around or in excess of 1.15x that figure.  Applying this multiple to the 2019 World Cup mean strike rate of 88.15 generates 101.37, so assuming the trend was going to hold, teams should have been aware that they'd need a World Cup batting strike rate of just over the 100 mark to have a decent chance of success.

As we can see from the table below, only one team managed a strike rate above 100 - tournament winners, England.

Team

Runs

Balls Faced

Strike Rate





England

3059

3045

100.46

Australia

2757

2899

95.10

Bangladesh

2145

2291

93.63

India

2432

2644

91.98

Pakistan

1934

2144

90.21

West Indies

1834

2053

89.33

South Africa

1838

2175

84.51

Sri Lanka

1503

1880

79.95

New Zealand

2023

2602

77.75

Afghanistan

1716

2363

72.62



In addition, England boasted by far the highest strike rate in ODI matches against the other World Cup teams from 1st January 2018, to 29th May 2019 (all pre-World Cup matches from the start of 2018 onwards):-

Team

Runs

Balls Faced

Strike Rate





England

8117

8076

100.51

New Zealand

5676

6263

90.63

Australia

6622

7389

89.62

India

7520

8410

89.42

South Africa

5013

5626

89.10

Sri Lanka

4476

5227

85.63

West Indies

4606

5402

85.26

Pakistan

6265

7348

85.26

Bangladesh

4471

5641

79.26

Afghanistan

1603

2034

78.81



Not only were England the only team in the World Cup to have a strike rate over the 100 mark, they were also the only team to manage this (and by some distance) in ODIs as the ten teams in the World Cup prepared for the tournament.  In short, they were the only team in the tournament to have demonstrated their ability to meet this strike rate trend either during the World Cup and also in advance of it.  

Whether England's strategy was pre-ordained via looking at historical data is something I obviously can't comment on, as I've not worked with the national team at all and therefore have no insight into their strategies.  Whether their World Cup success makes them the best ODI team in the world currently is also debatable - if you needed any further convincing of the variance-heavy nature of knockout events, then the last half an hour or so of the final against New Zealand was the perfect illustration - although it is pretty probable that they are.  What is less debatable is that they came into the World Cup best prepared to succeed, from a batting strike rate perspective at least.

In my opinion, the most incredible thing to take from England's World Cup triumph is no other team appeared willing to replicate their obviously successful playing style, as well as the dynamics from previous World Cups despite the fact that they went into the tournament as the number one ranked team in the world in the format, and demonstrated they were the most likely winners of the tournament based on their batting displays in the preceding 18 months or so.  The other teams, even including India - who have numerous prodigiously talented batsmen and an incredible depth of batting talent in their domestic game - deigned to take more of a safety-first approach during the World Cup, and as we can see from the data above, in ODIs in advance of the World Cup also.  It's difficult to know why other countries seem unwilling to follow the English batting model, with potential explanations possibly being a lack of talent or aggressive mindset among the playing pool, poor player selection, a fear of losing badly/unconventionally and therefore being criticised by the media and/or in social media, or perhaps even some coaches being oblivious to the advantages that such an approach could bring - it's often a struggle to act against 'conventional wisdom', which England did successfully.

Looking at the strike rate data for the World Cup countries in the table above, it's easy to see why my World Cup expectations for South Africa, and in particular, Pakistan, were low.  South Africa were ranked fifth in a congested pre-World Cup batting strike rate table - their issues in posting big totals were fairly predictable.  Pakistan had even greater strike rate issues prior to the World Cup, and there's actually a case to be made that they overperformed with the bat during the World Cup despite failing to qualify.  It would have been extremely unlikely for them - and would have required them to bowl absolutely magnificently on a consistent basis - to succeed in the World Cup with such batting strike rates.

As well as these batting strike rates, another obvious major issue facing South Africa was the age of their squad in advance of the World Cup.  I've done a lot of work on expected age curves for both young players and old, and despite it being the modus operandi of several teams around the world, what I can confidently state (as is the case with other sports as well) is that having an ageing team with numerous players past peak age and getting older at the same time is not generally a recipe for success.

The chart below illustrates ODI appearances v age for South African players from 1st January 2018 onwards:-
We can see here that the vast majority of the big names of South African cricket - players such as Imran Tahir, Hashim Amla, JP Duminy, Dale Steyn, AB De Villiers as well as captain Faf Du Plessis are the wrong side of 30 by some distance, and effectively, these players were allowed to get old at similar times.  Four of the ten players playing 20+ ODI matches for South Africa from 2018 onwards are aged 35 or greater.  Not only this, but 15 of the 30 players used by South Africa (50%) played 10 or fewer ODIs during this time period.  Compare that to England, who played more matches but used fewer players (25), and 11 of their players - who were all in the World Cup squad - played 25 ODIs or more during that same time period.  As well as the aforementioned aggressive batting mindset regardless of the risks, there was a another clear mantra from England - consistency of player selection - which South Africa arguably didn't take heed of.

Something South Africa would be well advised to look at in the immediate future is the low percentage of players used aged 28 or below - these were just nine players of the 30 used (Andile Phehlukwayo, Lungi Ngidi, Kagiso Rabada, Aiden Markram, Quinton de Kock, Heinrich Klaasen, Wiaan Mulder, Anrich Nortje and Duanne Olivier - now a Kolpak player) and this would indicate that they need a considerable squad overhaul by giving younger players greater opportunities.  The five players highlighted in red in chart above are around or below peak age as well as having been a consistent squad member, and it would make logical sense to use this quintet as core squad members in that overhaul.  

In sport, there often seems to be a general mindset of letting older players feature in a big event, almost in order to reward them for years of service.  However, as is also often the case in sport, emotionally-driven decision making tends to offer negative expectation, and allowing the older generation of South African players one last tournament in the 2019 World Cup also had the negative knock-on effect of denying younger players exposure and experience in big tournaments.  How much would Wiaan Mulder, for example, have benefited from playing big matches against high-quality opposition?  While he might not have been 100% ready for the tournament, he's a young player of extremely high potential and in any case, it's difficult to argue that his current version is much worse than a similar, older player in Dwaine Pretorius.  Now he has a long wait for major tournament experience.

However, it's not all negative for South Africa - they have plenty of talented, younger, players waiting for opportunities - they just need the selection mindset to give them consistent opportunity.  As I wrote previously here, we can look at profiling potential batsmen/all-rounders who could help South Africa both from an age perspective and also from a more aggressive batting mindset point of view.  I entered several filters into a database which contained the last two seasons of the Momentum One Day Cup.  For example, if we add age <=27 (higher potential improvement/upside), boundary % >10%, strike rate >90, minimum 200 balls faced, and having played both seasons, we obtain the following list of players who could be capable of making the step up to international level across the next few years:-

Existing regular players:-

Andile Phehlukwayo

Other players who fitted the filter:-

George Linde
Heinrich Klaasen
Kyle Verreyne
Patrick Kruger
Theunis de Bruyn

Close to filter:-

Gihahn Cloete
Janneman Malan
Wiaan Mulder
Reeza Hendricks
Sarel Erwee
Temba Bavuma


Moving onto Pakistan, some readers may be a little surprised as to why I had low pre-World Cup expectations for them, but the chart below accurately illustrates why (showing team batting strike rate and team bowling economy against all other World Cup teams in ODIs from 1/1/18 to 29/5/19):-

Here we can see that four teams in a cluster (Australia, South Africa, India & New Zealand) all had a better batting strike rate and better bowling economy in ODIs between the start of 2018 to the start of the 2019 World Cup, while England had worse bowling economy but a hugely better batting strike rate.  Pakistan were similar to both West Indies and Sri Lanka, and what the chart doesn't illustrate is that Pakistan also had an ODI bowling average in excess of 50 during this time period.

These numbers clearly showed World Cup success for Pakistan would be extremely difficult, despite both Champions Trophy success in 2017 as well as a reputation for being able to perform at a high level on their given day.  

In this time period, Pakistan played 33 ODI matches and in what looks a case of 'interesting' selection policy, used 30 players - the same as South Africa, with just eight players featuring in 20 or more of these matches.  One of these, Faheem Ashraf, did not feature in the World Cup.  14 players played eight ODIs of fewer, which perhaps indicates that the decision-makers didn't have a concrete idea as to their best team or squad.  This could also be evidenced by the fact that immediately prior to the World Cup, they jettisoned Ashraf, as well as Junaid Khan and Abid Ali from their initial World Cup squad, including 34 year old Wahab Riaz (who hadn't previously played an ODI since June 2017), Mohammad Amir and Asif Ali instead.

Compared to South Africa, Pakistan had less core players who were aged well in excess of peak age, but did have a lot of fringe players aged 30+.  However, they have six highlighted core players (20+ appearances from 2018 to start of World Cup) of around or below peak age, and these - as well as numerous talented bowlers at or below peak age - should be likely to form the core of their upcoming squads.

It is my view that Pakistan's upcoming ODI squads should contain options which have considerably more upside in terms of boundary-hitting.  The table below illustrates the boundary percentage and non-boundary strike rates of Pakistan batsmen who have faced 200+ balls in ODI matches from the start of 2018 onwards:-

Player

Boundary %

Non-Boundary Strike Rate




Imad Wasim

15.33

64.07

Asif Ali

14.64

64.85

Fakhar Zaman

12.52

47.95

Mohammad Hafeez

9.73

50.45

Babar Azam

9.05

54.16

Haris Sohail

9.04

51.97

Mohammad Rizwan

8.73

58.57

Imam Ul Haq

7.18

54.94

Sarfaraz Ahmed

6.87

64.36

Shoaib Malik

6.02

56.55

Shadab Khan

5.16

46.89


To provide context, England's numbers during the same time period look like this:-

Player

Boundary %

Non-Boundary Strike Rate




Jonny Bairstow

16.66

45.73

Jason Roy

15.07

52.71

Jos Buttler

14.77

64.57

Eoin Morgan

12.51

50.45

Alex Hales

12.38

44.82

Moeen Ali

10.31

50.37

Chris Woakes

10.25

56.17

Ben Stokes

9.21

51.16

Adil Rashid

9.00

54.69

Joe Root

7.45

62.05



While non-boundary strike rates are similar, and show that England's ability to rotate the strike aren't dissimilar (or is perhaps a touch worse) compared to Pakistan's, England's boundary hitting is markedly higher in general and show the sort of numbers required in order to obtain team strike rates of around 100 or greater.  This would also bring into question whether Pakistan can afford to play Imam Ul Haq, Haris Sohail, Sarfaraz Ahmed and Babar Azam all in the same team and captain Sarfaraz Ahmed - a high level rotator of strike in ODI cricket - looks much better suited to a middle-overs accumulator role as opposed to a lower-order hitter.

Given this, and the fact that of the World Cup top/middle order batsmen, only Fakhar Zaman, Babar Azam and Imam Ul Haq are below 30 years of age, a gradual overhaul of Pakistan's ODI batsmen looks logical.  Running the same filter as I did for South African batsmen, this time using Pakistan's domestic Quaid-E-Azam One Day Cup/Pakistan Cup/Departments One Day Cup/National One Day Cup data instead with a higher balls faced sample size (age <=27 (higher potential improvement/upside), boundary % >10%, strike rate >90, minimum 300 balls faced, and having played both seasons - 2017/18 and 2018/19) I was able to generate the following list of potential, mostly young and low-profile, Pakistan batsmen with future upside:-

Existing regular players:-

Asif Ali

Other players who fitted the filter:-

Ahmed Shehzad
Ahsan Ali
Ali Imran
Danish Aziz
Hasan Mohsin
Khushdil Shah
Mukhtar Ahmed
Saad Ali
Saud Shakeel

Close to filter:-

Iftikhar Ahmed
Saif Badar
Sami Aslam
Shan Masood
Umar Akmal

As we saw with South Africa, there are a number of young players with upside who could potentially improve the Pakistan setup in years to come, and it wouldn't be a particularly difficult process for me to create an algorithm to grade performances in each of the different domestic competitions in order to quantify for opposition quality, and thereby understanding which players have the greatest future, and more immediate upside.  I won't go into that detail here, but naturally, interested coaches or organisations can get in touch if interested via the contact details below.

While researching these players, Shan Masood cropped up as a particular player of interest.  He didn't quite fit the filter due to age (29) but both long-term and more recently, he's a run-scoring machine.  Despite this, Pakistan have only ever given him opportunities in five ODI matches and his profile is perhaps similar to Dimuth Karunaratne, who initially struggled in ODIs for Sri Lanka and was dropped from 2015 to 2019, despite scoring an incredible amount of runs in Sri Lanka's domestic 50 over cricket.  Naturally, judging a player from a small sample size of data is not recommended and there's certainly a reasonable argument to suggest Masood should have been given more opportunities in the past by the national team.

If this article has given you insight into the data that Sports Analytics Advantage can offer cricket franchises around the world in formulating team strategies, draft or auction plans, or any other work, please feel free to enquire at sportsanalyticsadvantage@gmail.com.

Comments