Fantasy Sports Team Selection Requires Higher Skill Than Mutual Fund Management

 

Is It Luck or Skill?’ is a joint academic study by Prof. Vishal Misra & Prof. Devarat Shah from MIT and Columbia University

National, August 26, 2020:  Today, Researchers from MIT (Massachusetts Institute of Technology) and Columbia University published a joint academic study titled ‘Is It Luck or Skill?’. The study applies tests to data obtained from Fantasy Sports platforms for cricket and basketball (2013 to 2016) and various Mutual Funds (2005 to 2018). The study conclusively demonstrates that luck plays a bigger role in the success of a Mutual Fund Manager compared to that of a Fantasy Sports player. The study includes over 16 million data points collected over 4 years.

The rigorous data-driven study, undertaken by the researchers at MIT and Columbia University, uses a robust statistical framework for evaluating the hypothesis of luck and applies a novel test that addresses the question of whether fantasy sports is a game of skill or luck. The study was performed by Prof. Vishal Misra, Columbia University, Prof. Devavrat Shah, MIT and Dr. Sudarsan V. S. Ranganathan from MIT. 

Key Findings:

1. The research study quantifies the degree to which fantasy sports and mutual funds relatively exhibit a predominance of skill
2. Data as provided by Dream11 for this study strongly indicates a predominance of skill, in that luck may have more role to play in Mutual Fund Management than that in Team Selection in Fantasy Sports format as offered by Dream11 since the variance in skill is higher in Fantasy Sports than that in Mutual Fund Management. 
3. Dream11’s fantasy cricket data indicates a skill predominance by a factor of 10, while the mutual fund data indicates skill predominance by a factor of 3
4. The hypothesis of luck playing a role in the fantasy sports team selection is overwhelmingly rejected. In particular, the data present strong evidence that the outcomes of fantasy sports (cricket and basketball) administered by the Fantasy Sports operators, like Dream11, are not driven by pure luck and skill has a predominant role to play
5. The number of users in Dream11’s fantasy sport, with predominant skill, is at least 9x higher compared to that if skill had no role to play

Talking about the study, Vishal Misra, Professor in the Computer Science and Electrical Engineering Departments at Columbia University, said, “Online Fantasy Sports (OFS) is a rapidly growing industry, especially in India. Time and again the OFS industry has presented a challenge for the regulatory bodies across the globe: do they represent gambling or games of skill? Our mission was to address this question scientifically and devise a technique that can answer this question rigorously. Our analysis included data sets from Fantasy Sports as well as a vocation that requires significant analysis and research, Stock Market - Mutual Funds. Based on our comprehensive research with a data set containing millions of events, it is safe to say that picking a team for fantasy sports requires a significant amount of skill and the variations in skill levels of Fantasy Players exceeds that of the Mutual Fund Managers we analyzed.”

Talking about the methodology applied to the study, Devavrat Shah, Professor of Artificial Intelligence and Decision as well as Director of Statistics and Data Science at MIT, said, “The question of whether skill has a role to play in fantasy sports presented us with an opportunity to address age-old questions by developing novel mathematically grounded data-driven test: the failure of the test, as it happens to be the case data observed on Dream11 platform, undeniably establishes that their skill has significant enough role to play in the outcomes. Indeed, it makes sense to check whether mutual funds, which are traditionally considered to be driven by skill, do manage to fail such a test. But the intensity of failure in the context of Dream11, for example, is much more significant. That is, we are way more confident about the role of skill in the outcome of Dream11 compared to the performance of mutual funds!”  

A similar study was undertaken by Dr. U Dinesh Kumar, Professor, Decision Sciences, IIM Bangalore in 2019 which, through a detailed analysis of the data set with sufficient evidence, established that Dream11’s format of fantasy sports is skill dominant. The inferences indicated that player selection has a huge influence on achieving high scores, therefore establishing that performance on the Dream11 platform is skill dominant. The students at IIM Bangalore also go through this unique case-study as a part of their curriculum (https://www.iimb.ac.in/iimb-research/case-studies). This case has been published in the Harvard Business Review and is available in 75 countries across more than 410 universities. Link: https://store.hbr.org/product/fantasy-sports-a-game-of-skill-or-chance/IMB781

About Prof. Vishal Misra- Vishal Misra is a Professor in the Computer Science and Electrical Engineering Departments at Columbia University, where he has been on the faculty since 2001. He is an ACM and IEEE Fellow, and his research focus is on modelling, analysis, and design of network systems. His work is used by hundreds of millions of users daily, whether it is in the form of commercial products via the companies he started (ESPNCricinfo and Infinio), Internet standards (bufferbloat control mechanism in every cable modem as part of DOCSIS 3.1) or regulations (India has the strongest net neutrality regulations anywhere in the world, which follow his recommendations). He received his B Tech from IIT Bombay and MS and PhD degrees from UMass-Amherst. Honours include NSF CAREER Award, DoE CAREER Award, IBM and Google Faculty Awards, Distinguished Alumnus Award (IIT Bombay) and Distinguished Young Alumnus Award (UMass-Amherst).

About Prof.Devavrat Shah- Devavrat Shah, a faculty member of Electrical Engineering and Computer Science at MIT since 2005, is a member of the Laboratory for Information and Decision Sciences (LIDS) and the Institute for Data, Systems and Society (IDSS). He is the founding director of the Statistics and Data Science Center (SDSC) with the mission to set up a 21st-century academic unit in Statistics and Data Science at MIT.  His work has received multiple recognitions - Rising Star Award from the Association for Computing Machinery (ACM) Special Interest Group for the computer systems performance evaluation community (SIGMETRICS), the Erlang Prize from the Applied Probability Society of INFORMS in addition to paper prize awards including the Best Publication Award from the Applied Probability Society of INFORMS, Best Paper Award from Manufacturing and Service Operations Management Society of INFORMS, N(eur)IPS Best Paper Award and ACM SIGMETRICS Best Paper Award.  He received the 2019 ACM SIGMETRICS Test-of-Time Paper award. He received NSF CAREER Award and is a distinguished young alumni of his alma mater IIT Bombay. In 2013, he founded the machine learning start-up Celect, Inc. which helps retailers with optimizing inventory by accurate demand forecasting. In August 2019, Celect, Inc. was acquired by Nike, Inc.