Veterans in T20 - Cricket's Biggest Waste of Resources?

Email: Sportsanalyticsadvantage@gmail.com

15th February, 2018.

Over the next few weeks, I'm going to produce a series of articles focusing on age curve analysis in T20.  Essentially, it is my belief that understanding the relationship between age and ability is somewhat misunderstood by T20 teams and franchises, with young players likely to exhibit further improvement, and older players begin to decline as they edge closer to retirement.

Despite football still not necessarily embracing quality analytics as much as it could, it is a sport which is considerably ahead of cricket in this respect, and this is reflected in the transfer fees paid to players - something strongly dependent on age.  

Generally speaking, top-level football teams are unwilling to pay big transfer fees for players in their thirties, and typically also give players in this age bracket much shorter contracts than they would consider for younger players.  It's also not uncommon for teams to stockpile young players (for example, Chelsea), loan them to smaller clubs and then sell them for a profit despite them not breaking through to the first team of their parent club.  Despite there being speculation regarding the ethics of such a transfer strategy, it's important to note that players are not forced to sign for a particular club (unlike in T20 drafts or auctions) and also it is undoubtedly a solid business strategy.

While it is fair to suggest that future resale value is a big factor in football, and isn't such a consideration in cricket, there are still several critical elements which remain the same across the sports.  Examples of these include the previously mentioned improvement/decline depending on age, and also the fact that younger players are likely to be more loyal, given that the team that first selects them to play for the first team has essentially given them their chance on the big stage.

Given this, it should be reasonable to expect that T20 franchises should be cautious about older players, and give the correct value to young players of high potential.  However, as Freddie Wilde discussed in this article following the eventually ill-fated T20 Global League Draft in September 2017, the opposite is actually the case.

In his piece, Wilde established that the average salary rose 'steadily through five-year age brackets', and noted that 'This is a trend that is generally evident in drafts but it is unsurprisingly prevalent in a new league in which new sets of management are drafting relatively unknown players.'

Realistically, there is no excuse for management - whether for new or old franchises - not having the skill-sets to assess players correctly.  There are enough T20 tournaments around the world for virtually any player's expectations to be assessed for any competition around the world - for example, my algorithm is even capable of giving expected data for an English Second XI player in the extremely unlikely event of him playing in the IPL.  

Such bias towards older players was also evident in the recent IPL auction, several months in advance of the upcoming tournament:-

Age

Mean Salary (Lakh)

% of Players >= 1 Crore Price




<= 20

216.96

52.17

21-25

236.33

48.98

26-30

375.00

72.37

31-34

411.11

85.19

35+

308.57

85.71


Here we can see that older players (26+) are valued considerably higher than younger players, with the age bracket of 31-34 having the peak purchase price of 411.11 lakh (around $640,000).  

While the IPL didn't exhibit quite the dynamics of the Global T20 League, where purchase prices rose throughout the age brackets, examining the 35+ age bracket was interesting.  It actually had the highest percentage of players signed for a purchase price of greater than or equal to 1 Crore, and overall, 85.37% of players aged 31 or greater were purchased for at least 1 Crore.  While only one player over 35 years of age was signed for more than 4 Crore (MS Dhoni, for 15 Crore), 11 of the 14 players in this age bracket were signed for priced between 1 and 4 Crore.

With this in mind, it appears that, perhaps unsurprisingly, the IPL is understands the age-decline curve marginally more than the potential South African franchises did, but there is still much to learn in this respect.

As older players frequently attract high prices in T20 drafts and auctions, it would be reasonable for the average person to assume that they performed considerably better than their younger counterparts.  However, when assessing data, this wasn't the case at all.

I examined the data from the last editions of the Big Bash, T20 Blast, Bangladesh Premier League, Caribbean Premier League, the Indian Premier League and the Pakistan Super League, and found that actually, players aged 35 or greater performed at a level very similar indeed to the T20 mean across these events.

Overall, players aged 35 or greater generated the following statistics:-

Batting Average

25.60

Batting Strike Rate

130.34

Bowling Average

26.33

Bowling Economy Rate

7.75


With a bowling average marginally greater than their batting average, and conceding 129.17 runs per 100 balls when bowling, almost identical to their batting strike rate of 130.34, it is clear that the average 35+ year old player in these leagues was merely an average player - despite having a high chance of commanding a larger salary than average.  

Tournament

Batting Average

Batting Strike Rate

Bowling Average

Bowling Economy Rate






Big Bash

23.68

127.13

31.61

7.42

T20 Blast

25.71

137.11

25.76

7.92

BPL

30.03

129.68

23.07

7.49

CPL

32.54

132.88

25.13

7.58

IPL

24.34

124.05

27.68

7.90

PSL

20.70

124.08

25.02

7.88


The above table illustrates how over 35 year olds performed in the last edition each league, and demonstrates that despite over 35 year olds being extremely valued in the IPL, they performed considerably below the level of an average player, with a batting average of 24.34 and strike rate of 124.05 - very poor data in what is one of the most batting-orientated T20 leagues around the world.  With the ball, economy rates and bowling averages for over 35s in the IPL also generated the 2nd worst data across all the leagues.  

Players in this age bracket performed much better in the BPL and CPL, and also from a batting strike rate perspective in the batting-orientated T20 Blast in England.

It's also worth evaluating how some of the serial T20 franchise batsmen (there are fewer T20 bowlers playing multiple leagues in this age bracket) aged 35 or greater have performed in the above events and T20 internationals from 2017 onwards, compared to in 2015 and 2016. 

Player

Year

T20 Batting Average

T20 Batting Strike Rate





Brendon McCullum

2015

37.05

169.95

Brendon McCullum

2016

23.00

135.29

Brendon McCullum

2017+

29.82

144.59

Chris Gayle

2015

62.20

167.56

Chris Gayle

2016

30.74

155.38

Chris Gayle

2017+

37.59

140.90

Kevin Pietersen

2015

36.75

129.52

Kevin Pietersen

2016

33.94

142.42

Kevin Pietersen

2017+

32.96

138.77

Kumar Sangakkara

2015

40.06

130.96

Kumar Sangakkara

2016

25.07

131.32

Kumar Sangakkara

2017+

28.71

128.99

Marlon Samuels

2015

41.79

125.54

Marlon Samuels

2016

43.69

108.40

Marlon Samuels

2017+

29.10

116.57

Michael Klinger

2015

57.65

137.64

Michael Klinger

2016

46.28

130.16

Michael Klinger

2017+

29.11

120.74

Shane Watson

2015

39.44

135.50

Shane Watson

2016

27.11

137.76

Shane Watson

2017+

23.22

132.28


Of these players, not a single one managed their best season from 2017+, and in a number of cases, had considerably worse data as their careers have progressed.

Brendon McCullum: Appears to have tailed off considerably after 2015, when his data was absolutely elite level.  After a mediocre 2016, his 2017 data shows he's still an above-average T20 batsman, but is no longer world-class.

Chris Gayle: For the time period assessed, he has shown a declining strike rate year on year, and if it wasn't for several match-winning displays in the latter stages of the BPL, would have recorded a barely above-average strike rate since 2017.

Kevin Pietersen: Pietersen has stated that this is his last season playing professional cricket, and data shows he's declining a little, with a slightly lower average in recent years, although his strike rate has remained fairly consistent, although far from at a world-class level.

Kumar Sangakkara: Very stable strike rate around the worldwide T20 mean, but average from 2016 onwards is also around average.  Data would suggest that he's no longer much better than the average T20 batsman.

Marlon Samuels:  Impressive average data until 2017, but woeful strike rates from 2016 onwards.  He also has a very poor strike rate against spin, which contributes to him having the potential to stagnate a team's innings when batting.

Michael Klinger:  Never the best from a strike-rate perspective, Klinger's asset was usually giving a very strong foundation to the start of an innings when opening.  However, now aged 37, his data from 2017 onwards shows a big decline, with his average now barely above average, in conjunction with a below-average strike-rate.

Shane Watson: Consistent strike-rate above the 130 mark, but a dropping batting average year on year.  From 2017 onwards as well, Watson's bowling has also been worse than the average player, taking wickets at a cost of 31.92 runs per wicket, and going for 9.07 runs per over.  From 2015 onwards, Watson's economy rate in T20 is 8.55.

Evidently, big name players have a point in their careers where they hit their peak and then decline at variable rates subsequently.  Given this phenomenon, when purchasing a player in their thirties, a T20 team should establish that the player in question is still close to his peak levels.  However, with many of the players above still perceived to be a valuable commodity in T20 cricket, it is far from certain that T20 teams currently do so.  

In future articles, I'll be looking at profiling the age/ability curve in more detail, in an attempt to try and assess the likely age that players peak at, before declining.

If this article has given you insight into the data that Sports Analytics Advantage can offer cricket franchises around the world in formulating draft or auction plans, please feel free to enquire for bespoke draft and auction strategies via sportsanalyticsadvantage@gmail.com.
Comments