The Indian Premier League:

Pay versus Performance

by

Krittivas Dalmia

An honors thesis submitted in partial fulfillment

of the requirements for the degree of

Bachelor of Science

Undergraduate College

Leonard N. Stern School of Business

Table of Contents

i. Television and Viewership ........................................................................10

i. Creating the Index ......................................................................................23

List of Tables and Figures:

Table 1 – Team Owners ...................................................................................................................9

Table 2 – Top 10 Paid Players of the Indian Premier League .......................................................14

Table 3 – Salaries for the Icon Players ..........................................................................................15

Table 10 – The Top Players According to the Combined Index ...................................................27

Table 11 – Teams Bang per Buck ..................................................................................................28

Table 12 – Bowling: Salary on Past Performance Metrics ............................................................30

Table 13 – Batting: Salary on Past Performance Metrics ..............................................................30

Table 14 – Batting: Salary on Past Performance (excluding A-List) ...........................................31

Table 15 – Bowling: Salary on Past Performance and Age ...........................................................31

Table 16 – Batting: Salary on Past Performance and Age .............................................................32

Table 17 – Batting: Salary on Past Performance and Age (excluding A-List) .............................32

Table 18 – Bowling: Salary on Past Performance, Age and Age

2

.................................................33

Figure 1 – Scatter Plot: Price vs. Batting Average ........................................................................37

Figure 2 – Scatter Plot: Price vs. Bowling Average .....................................................................37

Acknowledgements

First and foremost, I would like to thank my thesis advisor, Professor Raghu Sundaram, for

sharing the same enthusiasm and passion in my topic of research. In addition his undivided

attention, constant guidance, patience and immense knowledge were a source of motivation to

help me complete this thesis. I could not have hoped for a better thesis advisor and this would

not have been possible without his help.

I would also like to thank the faculty advisor for the Honors Program, Professor Marti

Subrahmanyam, for giving me an opportunity to be a part of this program and pairing me with

I thank my fellow thesis students, especially Veena Bontu, Misha Esipov, and Vinod Kasturi for

their constant criticism, motivation and enthusiasm towards my topic and the class, which always

pushed me to work on my thesis. Without their company in Friday morning class, in Stevenson

Lab or in the Leadership Suite till 4 A.M throughout the semester, this thesis would not have

been possible.

I would also like to thank Hannah Kim and Surekha Rao for their constant moral support.

Lastly, I would like to thank my parents and family for their love and constant support in all the

decisions I made in completing this thesis.

Abstract

In my thesis, I examine the initial auction process for the Indian Premier League, a recent and

immensely successful cricket league in India. The league is based on a new format of the game,

one that requires a very different set of skills compared to earlier and traditional forms. I look to

find the link between pay and performance for the top 89 players in the league. Since the league

is based on a new format of the game, adequate historical data on the performance of players in

performance. I use the index to compare performance rankings to pay. The second method looks

The results are mixed; they show that pay cannot be adequately explained by past performance

alone, nor are pay levels justified by future performance. The bids for players in the initial

auction appear to have been based on intangibles that are hard to quantify. This is not, perhaps,

entirely unexpected given the very new format of the game. However, the last three years have

revealed a substantial amount about the relevant skills for this format and the players who

possess them, and I expect that the pending renegotiations of the original contracts will be tied

closer to performance.

1. Introduction

Cricket is a bat and ball sport that originated in England and spread through the British Empire to

international cricket calendar. The ICC has three levels of members: full, associate and affiliate

members. The full members or the leading countries in the game are Australia, Bangladesh,

England, India, New Zealand, Pakistan, South Africa, Sri Lanka, West Indies, and Zimbabwe. In

addition there are 35 associate members and 59 affiliate members for a total of 104 members.

The format of international cricket has evolved considerably over time. In its early days,

international competitions took on the form exclusively of bilateral ―Test match‖ series between

countries. Test matches were multi-day affairs, sometimes even of indefinite length with the

match lasting until a result was reached. Post-World War II, Test matches were limited to five

days in length, a format that continues to hold even today. In the 1970s, a second international

format made its appearance: the one-day international or ODI. ODI matches were given a

In the 2000s, a third and even shorter format was introduced called Twenty20 cricket, and

abbreviated T20. Games in the T20 format last only 3 hours, and internationally this format saw

its first ―Word Cup‖ only in 2007. The huge success of the T20 World Cup, particularly in the

professional league, the Indian Premier League, the subject of this thesis.

IPL, were the players who were paid the most successful? In addition, since the format of the

game is relatively young and there is not enough historic performance data available, what were

the bids for the players based on? Each format of the game requires a very different set of skills.

These differences are so significant that nations have different teams for each format of the

game. It is also important to note that each format of the game has come to demand a specific

skill set from players and has allowed for the development of specialists. The auction took place

only five years after the first Twenty20 game was ever played. However, given this background

to acquire players and compare the top paid players of the league to the players in the National

Basketball Association (NBA) and the English Premier League (EPL). I will proceed to calculate

a performance index to measure the pay to performance and will publish the research and results

thereof.

2. Cricket and the IPL: A Background

The new format of the game, 20-20 cricket is responsible for the growth of club cricket. Before

this, club cricket was not as popular as it is today. Though the game is played primarily at the

international level, each country has intra country competitions in various forms of the game.

counties as in England or island nations as in the West Indies. Though this competition was in all

formats of the game, each format had some changes, for instance Test matches ranged from 3 – 5

days instead of the traditional 5 days. This internal competition is more of an effort to find

players for the national teams. There were no professional leagues and the concept of club

cricket was nothing close to what the Indian Premier League has made it out to be.

In the 1970’s, Kerry Packer, a rich and influential Australian tried to create a league for his

television network in order to secure exclusive broadcasting rights to Australian cricket. Even

though his bid was refused by the Australian Cricket Board, he setup individual agreements with

players from leading countries and started the league. However in light of court cases and new

developments, the league was short lived and ended after 2 years. Thus the first attempt to create

20 game was played. As the reigning ODI champions, Australia were the favorites to win the

World Cup. This is because ODIs are the closest form of cricket to Twenty20. However, India

ended up winning the World Cup setting up the backdrop for the Indian Premier League (IPL).

Moreover, India had played only one T20 game in its history and won the final against its arch

rival, Pakistan. The World Cup and the win popularized the format of the game in cricket’s

largest market; the Indian subcontinent. This sowed the seeds for the possibility of a successful

league and the Indian Premier League was born within a year.

(T20 Cricket). This form is the shortest version of the game and is completed within 3-4 hours

making it extremely interesting, exciting and enthralling to watch. Chaired by Mr. Lalit Modi in

collaboration with the Board of Cricket Control in India, the Indian Premier League was created

to emulate the professional sporting leagues in America specifically the National Football

League, the National Basketball Association and Major League Soccer. Still in the initial stages

of its life, the league comprises of 8 teams, each of them structured as a franchise and owned by

leading industrialists such as Mukesh Ambani and Vijay Mallya, corporate owners such as India

Delhi Daredevils, Kings XI Punjab, Kolkata Knight Riders, Mumbai Indians, Rajasthan Royals

and Royal Challengers Bangalore, each named after a big metropolitan city in India. For the 4

team. VVS Laxman decided to surrender his icon status to enable his team to have more money

to buy more players. However, the other 5 players retained their status. Thus, though these

players are some of the top paid players in the league, I have not included them in the above

In the 2009 and 2010 auction, players have been bought for higher prices than stated in the above

table. In the second auction, Kevin Peterson and Andrew Flintoff were bought for $1.55 million

into account the initial auction. This is because there is a learning curve with auctions, which

could have affected the prices, teams paid for these players.

Table 4 – Salaries of the Top 10 EPL Players

This table shows the top 10 paid players in the EPL according to the Portuguese agency, Futebol

Finance are: (all figures are converted to US Dollars using the exchange rate of 1.5 USD/

Pound.)

It also calculates per game salaries to be able to compare the salaries to those of the

Indian Premier League.

It is important to keep in mind that these are the top 10 players for the league not based on initial

contracts but on renegotiated deals and transfers. The salary per game has been calculated on 38

games, which is the number of games each team plays in a season.

The comparable salaries for the NBA for the top 10 paid players are in Table 5.

This table shows the top paid players of the NBA on an annual and per game basis.

The salary

per game has been calculated on the 82 games that each team plays in a season.

Thus, looking at Table 3, 4 and 5 and comparing just the salaries, we see that the Indian Premier

League players are paid huge amounts of money for a six week period. The amount is almost

second and third auction, the salary per game for the IPL is even higher with the highest per

week salary jumping to about $170,000. In addition, while looking at these salaries, it is

important to keep in mind that the highest paid cricket players in the world are Australians,

where the highest paid players get $1.5 million a year.

same amount to a player for a period of 6 weeks.

Indian Premier League

I will use this data to run regressions to try and find the correlation between pay and performance

amount of money that was paid to the players.

There are three main aspects to cricket; batting, bowling and fielding. Batting is how the team

scores runs in the game and there are a number of statistics pertaining to batting that speak to

how well a player is performing.

Runs – A run is a basic unit of batting. The basic objective of batting is to score as many runs as

possible.

Batting Average – It is the number of runs scored per innings played and is a first measure of the

Strike Rate – This is a measure of the number of runs scored per ball faced. It gives an idea as to

how fast the batsman is scoring his runs. Since each team plays only a limited number of balls,

scoring runs fast is important.

Like the other performance measures, the higher the number, the better the batsman.

to the objective of batting and a higher number indicates poor performance.

Bowling is the other major aspect of cricket. Bowling is how the team takes the wickets or gets

the other team out. If the bowling team takes 10 wickets in an innings, the other team’s innings is

over and the two teams switch roles.

Overs – An over is a set of six valid balls delivered by a single bowler. A valid ball is a ball

whose delivery meets certain specified requirements.

Maidens – A maiden is an over in which the bowler gave no runs. The objective of bowling is to

give the least runs possible and thus maidens are a good performance metric.

Wickets – A ―wicket‖ is getting the batsmen out and can be done in various ways. The objective

4W – This refers to ―Four Wickets‖ and is a record of how many times a bowler has taken 4 or

Average – The average in bowling refers to the number of runs given by the bowler per wicket

Economy Rate – The economy rate is the number of runs given per over. Since one of the

bowler.

Best – This refers to the best bowling performance by a bowler in his career. It is based on the

number of wickets and a higher number indicates a better performance.

they are not batting. Their objective is to prevent the batting team from scoring runs.

Catches – A catch is one way for a batsman to get out and is a good performance metric for the

fielders. It is a performance metric to judge the efficiency of the players in preventing batsmen

from scoring runs.

I begin my analysis with the first of the two questions: To what extent can the auction-based

statistics for the players. I divided the players into batsmen and bowlers and ran separate

included variables such as runs, average, strike rate, innings, catches, stumpings etc. For the

these regressions yielded insignificant results. This showed that the performance in the

subsequent seasons of the IPL was not matched to the amounts they were paid. It was a little

surprising to see that not one variable was significant considering I used most of the common

performance metrics.

This table summarizes the regression of all batting and fielding performance variables on price.

The R square for the regression was 22.4%.

Table 7 – Bowling – Regression Analysis: Salary on all Variables

This table summarizes the regression of all bowling performance variables on price.

The R square for the regression was 8.9%.

players. I also create a combined batting-and-bowling index by adding the two separate indices;

Where:

was limited to only 18 players, thus I did another regression with only the Test and One Day

International match statistics.

Table 14 – Batting: Salary on Past Performance (excluding A-List)

This table summarizes the regression of salary on past performance metrics in batting.

stated before, since this format of the game is extremely fast and dynamic, maybe the

performance was offset by the age of the players, which accounted for the variance of pay versus

performance.

Table 15 – Bowling: Salary on Past Performance and Age

This table summarizes the regression of salary on past performance metrics and age in bowling.

The R square for this regression was 48.8%. As can be seen in Table 15, this regression yielded a

new significant variable, A-List Wickets and strengthened the significance of the A-List

Average. The R square remained the same though.

Table 16 – Batting: Salary on Past Performance and Age

This table summarizes the regression of salary on past performance metrics and age in batting.

The R square for this regression was 47%. As can be seen in Table 16, adding age to the

The R square for this regression was 6.8%. As can be seen in Table 17, the R square for this

regression remained the same and there were no significant variables.

Since these regressions failed to show age as a significant explanatory variable, I changed age

from a linear to a quadratic variable. I also did this based on the assumption that teams in this

format of the game would value athleticism and experience more thus would pay more for the

younger and older players as compared to the middle aged players. Thus, I ran the same

regressions with age and age

Table 18 – Bowling: Salary on Past Performance, Age and Age

The R square for this regression is 52.1%. As can be seen in Table 18, adding age as a quadratic

variable increased the R square of the regression yet yielded only one significant variable, the A-

List Average.

This table summarizes the regression of salary on past performance metrics in batting and age as

a quadratic variable.

The R square for this regression is 47%. As can be seen in Table 19, though this increased the R

square for the regression greatly, it did not yield any significant variables.

Table 20 – Batting: Salary on Past Performance, Age and Age

(excluding A-List)

This table summarizes the regression of salary on past performance metrics in batting and age as

a quadratic variable.

Since the regressions continue to show that age was a fairly insignificant variable, I tested for

one last variable, captaincy. Some teams picked players for their team as leaders or as the captain

of the team. Thus I added a new variable captaincy in addition to age and past performance

measures to determine if pay is determined better. To do this, I assigned past captains a ―1‖ and

gave past players who had no captaincy experience a ―0‖. In addition, since making age a

quadratic variable did not help, I kept it as a linear variable in the remaining regressions.

This table summarizes the regression of salary on past performance metrics in bowling, age and

The R square for the regression is 48.8%. As can be seen in Table 21, this regression continued

This table summarizes the regression of salary on past performance metrics in batting, age and

captaincy.

yield no significant variables.

5. Conclusion

This research has investigated possible justifications (in terms of past performance) for the

amounts bid in the player-auctions in the Indian Premier League, and whether the amounts paid

were justified by subsequent performance in the league games themselves. What made the

auction process unusual was that the format of the league games (the so-called Twenty20 or T20

format) is a relatively new one in cricket and very little data on performance metrics in this form

of the game were available at the time of the auction.

The results are generally negative. I find that the highest-paid players were not necessarily the

the league. Secondly, player performances in other formats of the game do not fully explain the

of performance; however they show little relationship to pay. Regression analysis confirms this

result: The results of the regressions have little statistical significance with the price paid for

players. Generally, one would expect the better paid players to perform better and the past

performers to be paid better however, putting in specific factors, this was not the case. For

example, Ricky Ponting, an Australian cricketer and a legend in the other forms of the game was

paid a mere $400,000. This becomes more shocking when you take into account that he is

highest paid contracted cricketer in Australia (and the world since the Australians are the highest

paid cricketers in the world). On the whole however, higher paid batsmen performed better and

The figures are scatter plots of price vs. averages.

As far as past performance in concerned, the regressions for past bowling metrics still yielded A-

List Average as a significant variable in all the regressions suggesting that there was a

correlation between this number and the price paid for players. For the pre-auction batting

metrics however, the data failed to suggest significant variables. This is not too shocking because

it affirms the importance of different skill sets in the different formats of the game. On the

whole, the bowlers were better explained through the performance metrics in the other formats of

the game.

As suspected, this proves that the pay was not based on any significant statistical data but on

more intangible factors that are hard to quantify. While some players have clearly been paid less

than the amount they should have, others have failed to justify the amount paid for them. There

Barrett, Brad. "50 Top Paid Players in World Soccer for the 2009-2010 Season." Soccer

Cricinfo.com | Cricket News, Live Scores and Statistics. Web. 10 Nov. 2009.

Doley, Amrit. "The World's Top 10 Most Popular Team Sports - Sportingo." Sports News and

"Each IPL Squad Must Have 16 Cricketers." News. 19 Feb. 2008. Web. 20 Oct. 2009.

"The NBA's Top Ten Highest Player Salaries for 2009-2010 | Sportige." Sports News to Enrich

"Nita Ambani Wins the Tie-breaker, Bags Pollard." Reliance News. Web. 5 Apr. 2010.

"World Gazetteer: India - Largest Cities (per Geographical Entity)." World Gazetteer: World