The Myth of Bowling Depth


Previously, in this article, I looked at the effect of batting depth in the T20 Blast, and the research that I performed illustrated that having capable batsmen at 7/8, and a wicket keeper who could bat - particularly high up the order - was a huge asset, and a boost to a teams expectation of winning.  

Furthermore, in the T20 Blast Finals Day 2016 preview, I looked at the comparison between Jonny Bairstow and Andrew Hodd at Yorkshire, and examined the effect of having to play 
Hodd over Bairstow.  Yorkshire supporters may respond and say 'when Bairstow is away with England, we had no choice but to play Hodd', and that would be true, with the caveat that the club gave a batsman so ill-suited for white ball cricket a contract in the first place.

The Finals Day also demonstrated the embarrassing inability of a player - Scott Borthwick - who is obviously inept at scoring at a high enough scoring rate for the format.  My finals day preview had his adjusted strike rate expectation at 89.82 and I think anyone who watched his 10 (12) innings, when his team badly needed impetus, would agree that it was painful.  If he'd have been able to improve this, to say 10 (7), as suggested mandatory for bowlers below, Durham would have had five further deliveries to score 6-7 more runs, which would have given Northants 9-10 runs to get from the final over.  It could have been a different ending with just this marginal gain.

In short - batting depth is invaluable, but it seems that many teams do not either appreciate its value, or their squads are ill-equipped to have such batting depth.  This is quite amazing given the obvious value of teams even having bowlers who can play a 10 (7) style innings, and considering the huge rewards for a successful T20 player currently, bowlers would be well advised to put themselves in the best position they can by improving their batting to achieve this.  As money talks in every aspect of life, these rewards will only grow in the future as T20 inevitably takes over the world.  Purists will scoff but the cricket landscape will be dramatically different in possibly as little as five years time.

Anyway, enough about batting depth - what about bowling depth?  Is having more bowling options than necessary an asset or a hindrance?  Does the variety help a team prise out a well-set batsman, or bowl an economical over here and there, or are weaker bowlers considered weaker for a reason - they aren't very good?  

As is always the case, it's best to find out with data, so that's exactly what I did...

The table below illustrates the data for the 6th bowler of an innings in the five major domestic T20 leagues around the world in 2016, and also T20 internationals which involved test playing nations:-

6th

Blast

Big Bash

IPL

CPL

PSL

T20i

Overall

Overs

456

151

243

117

85

205

1257

Runs

3758

1187

2085

1037

587

1585

10239

Wickets

147

51

45

29

21

62

355

RPO

8.24

7.86

8.58

8.86

6.91

7.73

8.15

RPW

25.56

23.27

46.33

35.76

27.95

25.56

28.84


We can see that there was a big discrepancy between three leagues which had very poor figures- T20 Blast, the IPL and the CPL - and the Big Bash, PSL and T20 internationals, which had around or better than average figures.  In particular, the IPL and CPL batsmen punished 6th bowlers in the extreme, which would lend weight to the theory that big-hitting batsmen can punish poor bowlers significantly more than expectation.

For reference, the average runs per wicket in T20 is 25.10 with there being a mean of 8.08 runs per over scored, so we can also see that overall, the 6th bowler figures were worse than average for both metrics.  Runs per wicket was 1.15x higher than the average for the 6th bowler overall, and runs per over 1.01x the mean economy figures.

On this basis, we can start to build a picture that using a sixth bowler isn't particularly an asset, and the good leagues for 6th bowlers do have logical explanations - Australia have plenty of bowling depth to use in the Big Bash, and the Pakistan Super League was played in very bowler friendly conditions, whilst T20 international players are usually internationals for a reason - they are good cricketers.  It is logical that T20 international teams have natural depth.

However, this theory would be rubber-stamped if data on seventh and eight choice bowlers (yes, there were a few teams using eight bowlers, and captain Virat Kohli used nine in one IPL match this year) was also poor.  

Virat Kohli used nine bowlers in an easy win in the IPL this year...

The table below shows the data for 7th/8th bowlers combined, for the same sample:-

7th/8th

Blast

Big Bash

IPL

CPL

PSL

T20i

Overall

Overs

103

20

42

22

8

35

230

Runs

900

162

351

200

72

282

1967

Wickets

33

7

10

6

2

10

68

RPO

8.74

8.10

8.36

9.09

9.00

8.06

8.55

RPW

27.27

23.14

35.10

33.33

36.00

28.20

28.93


The overall data was even more damning - with a very poor 8.55 runs per over mean economy rate and a bowling average of 28.93 runs per wicket.  These figures were 1.06x and 1.15x the overall T20 bowler mean respectively.  

Again, the Big Bash and T20 internationals came up well for bowling options, but the other four leagues sampled recorded very poor data for 7th and 8th bowlers indeed. 

Data like this makes me wonder what actually possesses the minds of captains to bowl so many different bowlers, when the numbers are so poor for doing so.  Perhaps they are completely unaware of these numbers, or just think they know what they are doing, when it is very obvious they do not.  

During the course of this research, many of these 7th/8th bowlers were batsmen who the commentators phrase 'occasional bowler' would be very accurate.  I'd prefer to use the phrase 'sacrificial lamb', because these players are basically being thrown to the lions by their captains.

A number of bowlers seemed to be used as 6th/7th/8th bowlers quite often, and strangely many of them seemed to be West Indian - Kieron Pollard, Sulieman Benn, Darren Sammy and Darren Bravo featured regularly in the samples.  Chris Gayle also came up a few times and didn't have the worst numbers from a small sample, and Bravo's figures weren't horrific either.  I have no idea whether this finding was particularly significant, but it was interesting nonetheless.

As a sport/format of a sport, T20 is definitely one of the most solvable from a Game Theory style basis, and deep research shows it is clear that there are examples on almost a game by game basis that captains/management are running their teams on a far from optimal basis.  

The auction format for various franchises around the world - a format that will only grow - also is a very solvable problem, and my next article will look at the 2016 IPL auction and illustrate the best and worst signings of that auction.  

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