Cricket Analytics

Translate qualitative questions about player strategies, performances, umpire biases, team predictions and everything else cricket into rigorous quantitative insights and visualize the findings
Data science
Cricket

Some projects


Batter strategies and their impacts in the IPL

Spend time or hit out? How have different strategies affected teams’ winning chances?
Read more

How have different IPL players contributed to their team’s winning chances?

Dissecting player impacts on the match outcomes by using the win probability index published by ESPNCricinfo.
Read more

Predicting shot played from text commentary

Train machine learning models to predict shot played by batters from text commentary.
Read more

Quantifying context-aware extra runs scored by batters in the IPL

Develop a context-aware measure of extra runs scored by IPL batters and classify batters based on this metric.

Coming soon

What does “Umpire’s Call” tell us about umpire bias?

Spend time or hit out? How have different strategies affected teams’ winning chances?

Coming soon

Other ideas

  1. How crucial is toss in winning a match? Does this depend on particular aspects of the match like day vs night, venue etc.
  2. Are umpires biased as revealed by DRS data?
  3. What opener strategies exist and how do they impact match outcomes?
  4. What finisher strategies exist and how do they impact match outcomes?
  5. What is the impact-per-dollar of different players in the IPL and other leagues around the world? Who are the players that have had more impact-per-dollar than their peers in a similar price class?
  6. How good is ESPNCricinfo’s score prediction and win probability?

If you have more ideas that you want to see explored, or if you are interested in discussing/collaborating with me, please feel free to reach out.