2018 Afghanistan Premier League T20 - Draft & Squad Analysis

3rd October, 2018.


If this article has given you insight into the data that Sports Analytics Advantage can offer cricket teams around the world in formulating draft or auction plans, selection strategies or tactics, please feel free to chat to us at sportsanalyticsadvantage@gmail.com.


The inaugural Afghanistan Premier League T20 commences on Friday, so with the first match two days away, I wanted to discuss the dynamics of the tournament, particularly focusing on the draft and the decision-making of franchises in advance of the start.

With the tournament being new on the cricket calendar, it was, without a doubt, critical for franchises to ascertain as much information as possible to make decisions of the highest expected value.  For this particular tournament, the main information which could have been obtained in advance focuses on conditions and player strength, both overseas and domestic.

Venue Data Analysis:

All the matches will be played at Sharjah Cricket Stadium, in the United Arab Emirates, a venue also used by the Pakistan Super League.  In 2018, there were 13 PSL matches played in Sharjah, and the following data from these matches was generated:-

Mean First Innings Score: 145.50
First Innings Runs Per Wicket: 20.54
First Innings Batting Strike Rate: 125.43
Second Innings Runs Per Wicket: 29.91
Second Innings Batting Strike Rate: 128.85
Overall Runs Per Wicket: 24.26
Overall Batting Strike Rate: 127.08
Spin Overs Bowled: 45.76%
Pace Overs Bowled: 54.24%
Spin Runs Per Wicket: 25.38
Spin Runs Per Over: 7.03
Pace Runs Per Wicket: 25.82
Pace Runs Per Over: 7.71

In addition, the mean venue score batting first across 2016 and 2017 combined in the PSL was 150.37, with a similar runs per wicket figure.  With this in mind, it would be reasonable to classify the venue as a fairly low scoring venue which, despite having a very high spin over bowled percentage, doesn't have a huge bias towards spin.  While the spin runs per over at Sharjah in PSL 2018 was considerably lower than that of pace, this is very standard throughout T20 events and many venues/tournaments actually have a bigger difference in bowling economy rates between spin and pace.

Across the matches in Sharjah from 2016-2018 in the PSL, teams scoring in excess of 160 when batting first won 9/12 times (75.0%).

Venue Actionable Advance Strategy:

In the fairly unlikely event that franchises had this information in advance of the draft, what player acquisition strategies could they work out?  

Firstly, teams need to recruit batsmen with the abilities to propel innings past this 160 mark (expected strike rate in excess of 135), and looking at the squad lists for the franchises, none really possess more than four batsmen with this expected strike rate figure who can be anticipated to play regularly.  Certainly, a number of overseas batsmen have been recruited who are unlikely to be able to achieve such levels, and I'd definitely question those signings (full team data at the bottom of this article).

Secondly, players with strong bowling economy (expected economy under 7.50) should be valued by franchises.  Nangahar look best equipped in this respect.

Player Draft Dynamics:

Moving on to player analysis, squad formation for the draft was relatively straightforward initially.  Each team was permitted one icon player, three gold players, two diamond players, six silver players, an associate player, and further emerging players.  Teams had to take the same number of domestic and overseas players in each bracket, so, for example, a team could not prioritise overseas players as icon/gold/diamond players, although there is little information available regarding any limit on playing overseas players.  The draft format, and scarcity of domestic quality, is worth discussing in a little more detail.

A number of months ago, I was approached by one of the General Managers of an Afghanistan Premier League franchise to prepare a report to pitch to a Hong Kong Blitz franchise owner.  I have very little idea what happened to it - the Hong Kong Blitz was eventually postponed - but looking at the General Manager in question's squad composition for the Afghanistan Premier League, I'm not entirely sure he paid a great deal of attention to it anyway.  Some relevant points from that report to discuss here are listed in italics below:-

With such a quality bias towards overseas players from both a batting and bowling perspective in Hong Kong, I recommended recruiting overseas all-rounders who would have expected batting and bowling data (based on my algorithm) which would be better than the average overseas player.  Doing this, a Hong Kong Blitz team would be able to have at least four batsmen capable of scoring quickly, while being able to bowl 16 of the 20 allotted overs, reducing the need to use lower-quality domestic bowlers as much.   Identifying an above-average associate player would also be key, and in effect, doing so would actually mean that domestic players would not necessarily need to bowl at all.

Domestic recruitment should focus on Hong Kong batsmen who can score quickly, particularly those capable of hitting boundaries at a rate of at least one per six balls on average. Recruiting domestic batsmen unable to score around the mean strike rate, or bowlers likely to concede higher than average should be absolutely avoided.

Such a strategy has some application for the Afghanistan Premier League, although considerably less than for Hong Kong, given the higher quality of domestic Afghanistan players.  The following domestic players (not emerging players - I'll discuss these in more detail later) were graded above-average, by my algorithm, based on the competition expected mean figures:-

Balkh: Mohammad Nabi, Gulbadin Naib.
Kabul: Rashid Khan, Hazratullah Zazai, Muslim Musa
Kandahar: Najibullah Zadran, Asghar Afghan, Hamza Hotak, Karim Sadiq, Sayer Shirzad
Nangahar: Shafiqullah Shafaq, Mujeeb Zadran, Rahmat Shah, Naveen Ul Haq, Zahir Khan
Paktia: Mohammad Shahzad

These 16 players look the best non-emerging domestic players, with the majority (but not all) being national team regulars.  With just 16 domestic players looking likely to add decent value to the average team, acquiring as many of these as possible is crucial, so as not to dilute a team's quality with perceived below-average domestic players.  Certainly, I anticipate the average overseas player to be able to add considerable value over the average domestic player - while this is the case for almost all T20 leagues, it does appear that leagues in the subcontinent are especially exposed to this dynamic.

Icon players were Chris Gayle, Brendon McCullum, Shahid Afridi, Shoaib Malik, Andre Russell and Rashid Khan.  In truth, it is difficult not to think that there are occasions where icon players in drafts have been categorised for marketing purposes, and it is quite commonplace for my data to consider some non-icon players to be better than icon players.  The latter two - Rashid Khan and Andre Russell - look the best choices, with it being extremely difficult to understand why any team with the first pick in this category would not want to choose the Afghanistan spin-bowling allrounder.

The other four players are aged closer to 40 than 30, and there are certainly arguments to suggest the first three, in particular, are not at previous peak levels.  

This year in T20s, the 39 year old Chris Gayle has struck at just 127 - a figure considerably lower than his career strike rate.  Shahid Afridi, now aged 38, didn't hugely contribute overall with the bat in the PSL, or in the Global T20 in Canada, this year, although he did have a very successful PSL campaign with the ball - much less so in Canada, however.    As I discussed here previously, McCullum's strike rate data is less impressive in the subcontinent than in other locations, and he looks to have a particular issue against spin bowling in the subcontinent.  

While it didn't quite happen here, a draft where an astute franchise will not want to choose ANY of the available icon/marquee players may not be that far away.  I'd much prefer a completely open draft system where players state their minimum required salary level as a reserve price, and then let market forces take over.  I feel that such a system would give an even bigger competitive advantage to franchises willing to go the extra mile with player recruitment.

Unsigned Players:

Despite my doubts over the abilities of a number of overseas signings, some quality players went unsigned in the draft.  I can understand why the likes of Joe Denly, Dawid Malan, James Vince and Liam Dawson were unsigned despite their obvious abilities in this format, as there will have been worries from franchises that they'd be lost to England duty.  If this concern was disregarded, they'd look a considerable upgrade on many overseas players.

In addition, Pakistan players look like they have not been given no objection certificates to participate.  Given that a number of Pakistan players were originally signed by franchises, it looks unlikely to have been a particular worry in advance of the draft, making the lack of interest in Junaid Khan and Imad Wasim a real head-scratcher.  Rumman Raees is another who went unsigned, but injury concerns possibly were a factor here.

Non-English overseas players who could have added quality, but who went unsigned initially, included Christiaan Jonker, Imran Tahir, Samuel Badree, Kesrick Williams and Anton Devcich, although Devcich has subsequently been picked up by Nangahar as a replacement.  Jonker and Tahir were likely to be required by South Africa at points (they are in the South Africa team for today's ODI against Zimbabwe).

English players (non national team) in particular seemed to be undervalued by franchises, which is strange for several reasons.  Firstly, some are VERY good players, who my algorithm makes better than some overseas players who were signed.  In addition, franchises often have a recency bias in draft/auction decision-making, by over-valuing recent performances because they are fresh in the memory.  Rather strangely, the effect here looks actually to be the opposite.

Certainly, the likes of Tom Kohler-Cadmore, Benny Howell, Jordan Clark, Ross Whiteley, Harry Gurney, Richard Gleeson, Josh Cobb and Pat Brown can feel pretty unlucky that they were not picked up, having entered the draft.

Associate Players:

Each team was required to sign one player from an associate nation, and Ryan Ten Doeschate, Ali Khan, Mohammad Naveed, Sandeep Lamicchane and Calum McLeod were selected.  Naveed, and in particular Lamicchane, look the best picks, although Ali Khan also has some potential.   The 38 year old Ten Doeschate's powers look on the wane, with mediocre expected batting data, and since a very successful 2016 T20 Blast campaign, hasn't bowled much.

I would have considered the likes of Rohan Mustafa, Craig Meshede, Roelof van der Merwe, or the Hong Kong duo of Nizakat Khan and Anshuman Rath.  All looked likely to be pretty capable in this particular event.

Emerging Players:

The requirement to pick emerging players complicates matters a little.  While this would be relatively easy in a country such as England, where there is a considerable amount of lower-level player data to interpret, it is tougher in this particular league, with the majority of the emerging players in the draft having little in the way of T20 data.  With this in mind, I listed all emerging players' data as N/A in the squad lists below.  However, I have listed below the emerging players that I anticipate as having the highest likelihood of success in the upcoming event (all players have relatively small data sample sizes):-

Nasir Totakhil - Left-Arm Spinner, Kabul. 
Waqar Salamkheil - Left-Arm Medium-Pace Bowler, Kandahar.
Waqarullah Ishaq - Right-Hand Batsman, Kandahar.
Imran Janat - Right-Hand Batsman, Nangahar.
Ibrahim Zadran - Right-Hand Batting All-Rounder, Nangahar.
Fazak Zazai - Right-Hand Batsman, Paktia.

Squads with Expected Batting/Bowling Data based on the Sports Analytics Advantage Algorithm:

As mentioned above, there are doubts over the participation of Pakistan domestic players in the event.  Where a replacement player is known, the squad has been updated from the original draft list:-

Balkh:

Balkh

Afghanistan T20 Expected Batting Average

Afghanistan T20 Expected Batting Strike Rate

Afghanistan T20 Expected Bowling Average

Afghanistan T20 Expected Bowling Economy






Aftab Alam

10.71

97.55

25.82

8.45

Asadullah Matani

N/A

N/A

N/A

N/A

Ben Laughlin

3.87

86.48

21.56

7.74

Chris Gayle

32.43

135.77

50.44

6.52

Colin Munro

34.82

152.04

32.66

9.30

Darwish Rasooli **

N/A

N/A

N/A

N/A

Farhan Zakhil

N/A

N/A

N/A

N/A

Gulbadin Naib

21.31

121.24

23.04

7.89

Ikram Ali Khil

13.21

107.25

N/A

N/A

Kamran Akmal

32.13

135.50

N/A

N/A

Mohammad Irfan

2.90

43.23

25.05

6.80

Mohammad Nabi

23.42

148.76

24.04

6.73

Mohammaz Nawaz

19.18

100.79

27.69

7.00

Qais Ahmad

10.22

103.95

30.04

8.19

Ravi Bopara

24.77

121.60

31.71

8.33

Ryan ten Doeschate

19.70

113.98

28.84

8.63

Samiullah Salarzai

N/A

N/A

N/A

N/A

Tariq Stanikzai

N/A

N/A

N/A

N/A

Usman Ghani

26.99

99.38

N/A

N/A

** Silver player, Darwish Rasooli, has little T20 experience, but from a fairly small sample size, some superb first-class and List A batting data.

Kabul:

Kabul

Afghanistan T20 Expected Batting Average

Afghanistan T20 Expected Batting Strike Rate

Afghanistan T20 Expected Bowling Average

Afghanistan T20 Expected Bowling Economy






Afsar Zazai

20.32

80.93

N/A

N/A

Ali Khan

N/A

N/A

25.12

8.17

Colin Ingram

33.25

137.04

31.45

7.77

Farid Ahmad (Malik)

N/A

N/A

24.23

8.66

Fitratullah Khawari

N/A

N/A

N/A

N/A

Hazratullah Zazai

38.80

122.79

N/A

N/A

Javed Ahmadi

29.63

105.44

N/A

N/A

Laurie Evans

31.21

113.79

N/A

N/A

Luke Ronchi

29.95

153.03

N/A

N/A

Muslim Musa

16.51

136.27

23.18

6.09

Nasir Totakhil

N/A

N/A

N/A

N/A

Rashid Khan

11.03

129.93

17.23

6.33

Shahid Mandozai

N/A

N/A

N/A

N/A

Shawkat Zaman

N/A

N/A

N/A

N/A

Sohail Tanvir

18.44

125.58

33.25

7.56

Usman Adil

N/A

N/A

N/A

N/A

Wayne Parnell

21.97

109.44

24.34

7.75

Zahir Shirzad

N/A

N/A

N/A

N/A

Zamir Khan

N/A

N/A

N/A

N/A


Kandahar:

Kandahar

Afghanistan T20 Expected Batting Average

Afghanistan T20 Expected Batting Strike Rate

Afghanistan T20 Expected Bowling Average

Afghanistan T20 Expected Bowling Economy






Abdul Baqi

N/A

N/A

N/A

N/A

Amir Hamza (Hamza Hotak)

10.01

95.86

26.95

6.92

Asghar Afghan

32.45

121.05

N/A

N/A

Brendon McCullum

26.59

130.56

N/A

N/A

Iftikhar Ahmed

23.84

97.39

53.31

10.31

Karim Janat

12.55

79.28

26.63

8.16

Karim Sadiq

28.10

120.52

27.70

6.84

Kevin O’Brien

15.11

111.80

27.02

8.91

Mohammad Naveed

15.62

117.94

22.59

7.23

Najibullah Zadran

32.81

152.18

N/A

N/A

Nasir Jamal

29.94

104.71

N/A

N/A

Paul Stirling

24.99

138.98

32.43

8.54

Sam Billings

22.55

122.75

N/A

N/A

Sayed Shirzad

N/A

N/A

19.63

7.31

Taskin Ahmed

5.81

57.95

35.39

9.56

Tymal Mills

2.35

68.52

27.45

7.87

Wahidullah Shafaq

N/A

N/A

N/A

N/A

Waqar Salamkheil 

N/A

N/A

N/A

N/A

Waqarulah Ishaq

N/A

N/A

N/A

N/A


Nangahar:

Nangarhar

Afghanistan T20 Expected Batting Average

Afghanistan T20 Expected Batting Strike Rate

Afghanistan T20 Expected Bowling Average

Afghanistan T20 Expected Bowling Economy






Andre Fletcher

28.81

119.71

N/A

N/A

Andre Russell

25.24

152.57

23.75

8.55

Anton Devcich

30.65

136.35

24.93

7.56

Ben Cutting

23.47

142.63

41.19

9.18

Fazal Haque

N/A

N/A

N/A

N/A

Fazal Niazai

N/A

N/A

N/A

N/A

Ibrahim Zadran 

N/A

N/A

N/A

N/A

Imran Janat

N/A

N/A

N/A

N/A

Johnson Charles

26.23

126.92

N/A

N/A

Khaibar Omar 

N/A

N/A

N/A

N/A

Mitchell McClenaghan

5.80

115.65

32.58

8.75

Mohammad Hafeez

22.35

109.39

22.33

6.65

Mujeeb Zadran

5.70

93.81

24.29

6.27

Mushfiqur Rahim

29.82

134.10

N/A

N/A

Najib Tarakai

24.60

119.58

N/A

N/A

Naveen Ul Haq

N/A

N/A

24.61

6.72

Nusrat Qureshi

N/A

N/A

N/A

N/A

Rahmat Shah

21.46

123.93

48.15

7.31

Sandeep Lamicchane

N/A

N/A

24.07

7.16

Shafiqullah Shafaq

28.26

167.75

N/A

N/A

Tamim Iqbal

35.74

125.54

N/A

N/A

Zahir Khan

N/A

N/A

16.13

7.05


Paktia:

Paktia

Afghanistan T20 Expected Batting Average

Afghanistan T20 Expected Batting Strike Rate

Afghanistan T20 Expected Bowling Average

Afghanistan T20 Expected Bowling Economy






Azmatullah Omarzai

N/A

N/A

N/A

N/A

Calum McLeod

23.56

110.68

N/A

N/A

Cameron Delport

23.16

126.48

27.44

7.95

Chris Jordan

14.35

101.99

24.84

8.05

Faheem Ashraf

19.83

127.14

21.22

7.96

Fazal Zazai

N/A

N/A

N/A

N/A

Ihsanullah Janat

38.28

128.15

N/A

N/A

Luke Wright

23.74

123.39

N/A

N/A

Mohammad Shahzad

41.72

141.20

N/A

N/A

Rahmanullah Gurbaz 

N/A

N/A

N/A

N/A

Samiullah Shinwari

24.86

107.66

38.50

6.98

Shahid Afridi

14.19

154.11

22.52

6.76

Sharafuddin Ashraf

13.84

136.62

33.98

6.58

Sikandar Raza

26.99

127.87

32.30

7.78

Tahir Adil

N/A

N/A

N/A

N/A

Thisara Perera

19.15

152.29

38.99

9.35

Yousuf Zazai

N/A

N/A

N/A

N/A

Zia Ur Rahman

N/A

N/A

N/A

N/A

Zia Ur Rahman Akbar

N/A

N/A

30.57

6.83




If this article has given you insight into the data that Sports Analytics Advantage can offer cricket teams around the world in formulating draft or auction plans, selection strategies or tactics, please feel free to chat to us at sportsanalyticsadvantage@gmail.com.
Comments