Decision Review System (DRS) data analysis - IPL 2024

Background

The Decision Review System (DRS) is a technology-based process to assist on-field match officials with their decision-making. Team captains may request that the third umpire consider a decision of the on-field umpires also known as Team review. Each team is limited to 2 unsuccessful reviews per innings.

Data Collection

The data has been collected from ESPNCricinfo - Match Flow section. Here is an example of Match 31: Kolkata Knight Riders vs Rajasthan Royals

Data Science Process

1. Ask an interesting question. (i) What is the goal? (ii) What would you do if you had all the data? (iii) What do you want to predict or estimate?
2. Get the data. (i) How were the data sampled? (ii) Which data are relevant? (iii) Are there any privacy issues?
3. Explore the data. (i) Plot the data (ii) Are there anomalies (iii) Are there patterns?
4. Model the data. (i) Build a model. (ii) Fit the model. (iii) Validate the model.
5. Communicate and visualize the results. (i) What did we learn? (ii) Do the results make sense? (iii) Can we tell a story?

Required Libraries:

1. Python
2. Pandas
3. Seaborn
4. Matplotlib
5. BeautifulSoup4 (for data extraction)

Questions answered:

1. Day 1: Total number of reviews taken in each innings.
2. Day 2: Total number of reviews taken in each over.
3. Day 3: Total number of reviews taken per type.
4. Day 4: Total number of reviews taken by batting or fielding sides.
5. Day 5: Total number of reviews taken per umpire.
6. Day 6: Total number of reviews taken in each match.
7. Day 7: Total number of reviews taken per team and Number of successful/unsuccessful reviews by each team.

Future work:

1. How has the trend of review taking changed over the course of the season?
2. Running average number of reviews over the course of the season?