The Effect of Batting Depth in the T20 Blast


As one of the easiest domestic T20 leagues to bat in (discussed here) it is clear that the T20 Blast is one of the weaker standard T20 domestic leagues throughout the world.

This is down to a number of factors with a lack of foreign superstars and more severe restrictions on overseas players than overseas domestic T20 tournaments, and financial issues for the English county teams at the forefront.  A very poor level of television coverage is also a significant negative.  

On the subject of financial issues, it seems that the UK cricket system needs to work out whether counties are to aim to be a successful teams whilst also being solid businesses, or merely feeders for the England team.  I feel it's impossible to be both and this split loyalty also reminds me of the horse racing industry which is propped up by betting.  

Sport teams cannot be sustained on crowds of one man and their dog, which four day county matches mostly have, yet it would appear that authorities are keen to have as many four day matches as possible to help the England team.  It can't work both ways.

Given this lack of finances, it's not difficult to envisage that many teams will suffer from a lack of strength in depth in their squads, and a good measure of this is how good their 7th and 8th batsmen perform with the bat, as well as their wicketkeepers.

Focusing on wicketkeepers first, it is generally accepted that in the modern T20 format, wicketkeepers are more associated with being 'batsmen who keep wicket', yet despite this, some clubs do not have the necessary playing staff to adhere to this rule, as the table below of some selected T20 clubs illustrates.  Data is from this year's T20 Blast group stages:-

Team

WK






C/Inns

Runs

BF

Ave

SR







Kent

11

164

142

14.91

115.49

Hampshire

11

143

125

13.00

114.40

Surrey

7

90

89

12.86

101.12

Essex

6

50

44

8.33

113.64

Yorkshire

4

25

29

6.25

86.21

Somerset

6

31

36

5.17

86.11

Overall

128

2903

2197

22.68

132.13


We can see that this table that the batting average for wicketkeepers in this year's T20 Blast group stages is 22.68, with a strike rate of 132.13.

However, these six teams had wicketkeeping batting performances significantly worse than average, with Yorkshire and Somerset having incredibly poor stats (mainly Andy Hodd and Ryan Davies, respectively).  Essex (James Foster) were little better, whilst it is a surprise that one of the teams with a bigger budget, Surrey, cannot find a decent wicketkeeper-batsman.  It's difficult to make a case for these players being good enough to play T20 cricket.

With just 16 completed innings for wicketkeepers amongst these three teams, it's clear that these three teams have wicketkeepers who are hidden down the batting order, which has a major issue in that it places more pressure on teams to find more batsmen who can play further up the order, as opposed to a Jos Buttler or Riki Wessels, or even Ben Cox, Phil Mustard or Adam Rossington, for example, who can compete as normal batsmen, to some extent.  It also means that bowlers need to be able to bat at a higher position in the batting order, which they may be incapable of doing well.

Interestingly, Kent, despite having one of the highest rated young wicketkeeper batsmen in the game, Sam Billings, also struggled, and this data backs up my assertions in this article, where I make the point that data indicates Billings is severely over-rated.

The records for the group stages of the T20 Blast are below:-

Team





Won

Lost

NRR





Kent

6

8

-0.643

Hampshire

4

8

-0.691

Surrey

7

7

0.153

Essex

7

6

0.174

Yorkshire

7

5

0.223

Somerset

3

10

-0.660


In total, these clubs won 34 matches and lost 44 (43.59% win rate), and had a mean net run rate of -0.241, so it looks obvious that having a wicketkeeper with a lack of batting ability has a definite effect on a teams chances of winning a match.

This is information that we can use in the betting/trading markets to our advantage, not to mention the obvious benefit that teams would have if they were aware of this data.

A further measure of strength in depth is the quality of batsmen in the 7th/8th positions in the batting order.  Ideally, these players should have the ability to score runs at a fast strike rate given that they are likely to be batting at a time of the innings where the need to do this is at its greatest.

The table below indicates the averages of all teams in the T20 Blast group stages for their 7th/8th batsmen and wicket-keepers combined, a real test of strength in depth (sorted by combined mean differential, explained below):-

Team




Overall 7/8/WK





Overall 7/8/WK




Won

Lost

NRR

C/Inns

Runs

BF

Ave

SR

Ave Mean Diff

SR Mean Diff

Combined Mean Diff













Nottinghamshire

8

2

0.741

9

356

240

39.56

148.33

2.08

1.13

1.21

Worcestershire

5

7

-0.862

15

492

345

32.80

142.61

1.72

1.09

0.81

Durham

6

6

-0.050

20

569

426

28.45

133.57

1.50

1.02

0.52

Lancashire

6

7

0.200

17

358

218

21.06

164.22

1.11

1.25

0.36

Leicestershire

4

8

-0.180

13

338

260

26.00

130.00

1.37

0.99

0.36

Middlesex

7

6

0.395

21

474

341

22.57

139.00

1.19

1.06

0.25

Gloucestershire

10

3

0.518

8

164

128

20.50

128.13

1.08

0.98

0.06

Glamorgan

8

3

1.005

15

319

298

21.27

107.05

1.12

0.82

-0.06

Sussex

5

6

-0.053

15

267

204

17.80

130.88

0.94

1.00

-0.06

Derbyshire

5

7

0.021

14

258

205

18.43

125.85

0.97

0.96

-0.07

Yorkshire

7

5

0.223

13

208

152

16.00

136.84

0.84

1.04

-0.11

Northamptonshire

7

5

0.265

22

361

271

16.41

133.21

0.86

1.02

-0.12

Surrey

7

7

0.153

20

382

340

19.10

112.35

1.00

0.86

-0.14

Birmingham/Warwickshire

6

7

-0.215

21

298

216

14.19

137.96

0.75

1.05

-0.20

Somerset

3

10

-0.660

17

239

193

14.06

123.83

0.74

0.95

-0.32

Hampshire

4

8

-0.691

31

401

310

12.94

129.35

0.68

0.99

-0.33

Kent

6

8

-0.643

27

328

282

12.15

116.31

0.64

0.89

-0.47

Essex

7

6

0.174

17

178

143

10.47

124.48

0.55

0.95

-0.50

Overall 7/8/WK




315

5990

4572

19.02

131.01

1.00

1.00

0.00


To explain the mean differential, it is calculated for each of the average and strike rate using the following method:-

Team Average or Strike Rate divided by Overall Average or Strike Rate.

So, for example, to find the average differential for Nottinghamshire's average for batting positions 7, 8 and wicket-keeper, we divide 39.56 (Nottinghamshire's average) by 19.02 (league positional average), which generates the 2.08 as shown.

Then, we add the average and strike rate differential together and subtract 2 (as 1.00 is average for each metric) to get the overall mean differential, which the table is sorted by.

If we look at the results of the most successful nine teams using the mean differential for positions 7/8/WK (Nottinghamshire to Sussex), we can see that they won 59 and lost 48 matches (55% win rate), whilst the least successful nine teams won 52 and lost 63 (45%), so performance with the bat in these three positions had a clear impact on a team's success.

Furthermore, the top nine teams had an average net run rate of 0.190, whilst the bottom nine teams struggled at -0.153.

In addition, if we look at the five teams (Birmingham, Somerset, Hampshire, Kent and Essex) with a -0.20 differential or worse, they won 26 matches and lost 39 (40%) with a mean net run rate of -0.407, so those teams with poor batsmen in these roles struggled badly.   

For example, Kent's regular 7th/8th batsmen were Fabian Cowdrey, Matt Coles, James Tredwell or Kagiso Rabada - not exactly batsmen who would strike fear into the hearts of opposition bowlers.  The effect of this lack of depth is a player like Alex Blake, who would be well suited to this role, is pushed too high in the batting order due to a lack of other options.

This article shows the value to traders and cricket teams alike of having individual player data in many areas, and is just a small part of what will be possible with cricket data once I have finished building and testing my models, which should be complete in the very near future.

If you enjoyed reading this article, please feel free to make a donation towards the upkeep of the website.  
This will also help me to prioritise the time to write further articles and previews.  





Comments