📜 Bangladesh Premier League Analysis

🎯 AIM

To analyze player performances in the Bangladesh Premier League by extracting insights from batsmen, bowlers, and match data—ranging from toss outcomes to overall match results.

📊 DATASET LINK

https://www.kaggle.com/abdunnoor11/bpl-data

📓 KAGGLE NOTEBOOK

https://www.kaggle.com/code/avdhesh15/bpl-analysis

Kaggle Notebook

⚙️ TECH STACK

Category	Technologies
Languages	Python
Libraries/Frameworks	Matplotlib, Pandas, Seaborn, Numpy
Tools	Git, Jupyter, VS Code

---

📝 DESCRIPTION

What is the requirement of the project?

• This project aims to analyze player performance data from the Bangladesh Premier League (BPL) to classify players into categories such as best, good, average, and poor based on their performance.
• The analysis provides valuable insights for players and coaches, highlighting who needs more training and who requires less, which can aid in strategic planning for future matches.

How is it beneficial and used?

• For Players: Provides feedback on their performance, helping them to improve specific aspects of their game.
• For Coaches: Helps in identifying areas where players need improvement, which can be focused on during training sessions.
• For Team Management: Assists in strategic decision-making regarding player selection and match planning.
• For Fans and Analysts: Offers insights into player performances and trends over the league, enhancing the understanding and enjoyment of the game.

How did you start approaching this project? (Initial thoughts and planning)

• Perform initial data exploration to understand the structure and contents of the dataset.
• To learn about the topic and searching the related content like what is league, About bangladesh league, their players and much more.
• Learn about the features in details by searching on the google or quora.

Mention any additional resources used (blogs, books, chapters, articles, research papers, etc.).

• Articles on cricket analytics from websites such as ESPNcricinfo and Cricbuzz.
• https://www.linkedin.com/pulse/premier-league-202223-data-analysis-part-i-ayomide-aremu-cole-iwn4e/
• https://analyisport.com/insights/how-is-data-used-in-the-premier-league/

---

🔍 PROJECT EXPLANATION

🧩 DATASET OVERVIEW & FEATURE DETAILS

📂 bpl.csv

• There are 19 features in BPL Dataset

Feature Name	Description	Datatype
id	All matches unique id	int64
season	Season of the match (20XX-XX)	object
match_no	Number of matches (For eg, Round 1st, 3rd, Final)	object
date	Date of the match	object
team_1	Name of the first team	object
team_1_score	Scoreboard of the team 1 (runs/wickets)	object
team_2	Second Team	object
team_2_score	Scoreboard of the team 2 (runs/wickets)	object
player_of_match	Player of the match	object
toss_winner	Which team won the toss?	object
toss_decision	Toss winner team decision took either 'field first' or 'bat first'	object
winner	Name of the team who won the match	object
venue	Venue of the match	object
city	City of the match	object
win_by_wickets	Team win by how many wickets	int64
win_by_runs	Team win by how many runs	int64
result	Conclusion of win_by_wickets & win_by_runs	object
umpire_1	Name of the first umpire	object
umpire_2	Name of the second umpire	object

🛠 Developed Features from bpl.csv

Feature Name	Description	Reason	Datatype
team_1_run	Run scored by the team 1	To covert team_1_score categorical feature into numerical feature	int64
team_1_wicket	Wickets losed by the team 1	To covert team_1_score categorical feature into numerical feature	int64
team_2_run	Run scored by the team 2	To covert team_2_score categorical feature into numerical feature	int64
team_2_wicket	Wickets losed by the team 2	To covert team_1_score categorical feature into numerical feature	int64

📂 batsman.csv

• There are 12 features in Batsman Dataset

Feature Name	Description	Datatype
id	All matches unique id	int64
season	Season of the match (20XX-XX)	object
match_no	Number of matches (For eg, Round 1st, 3rd, Final)	object
date	Date of the match	object
player_name	Player Name	object
comment	How did the batsman get out?	object
R	Batsman's run	int64
B	How many balls faced the batsman?	int64
M	How long their innings was in minutes?	int64
fours	No. of fours	int64
sixs	No. of sixes	int64
SR	Strike rate (R/B)*100	float64

📂 bowler.csv

• There are 12 features in Bowler Dataset

Feature Name	Description	Datatype
id	All matches unique id	int64
season	Season of the match (20XX-XX)	object
match_no	Number of matches (For eg, Round 1st, 3rd, Final)	object
date	Date of the match	object
player_name	Player Name	object
O	No. of overs bowled	float64
M	No. of middle overs bowled	int64
R	No. of runs losed	int64
W	No. of wickets secured	int64
ECON	The average number of runs they have conceded per over bowled	float64
WD	No. of wide balls	int64
NB	No. of No balls	int64

---

🛤 PROJECT WORKFLOW

Project workflow

  graph TD
    A[Start] --> B[Load Dataset]
    B -->|BPL Data| C[Preprocess BPL Data]
    B -->|Batsman Data| D[Preprocess Batsman Data]
    B -->|Bowler Data| E[Preprocess Bowler Data]

    C --> F{Generate Queries?}
    D --> F
    E --> F

    F -->|Yes| G[Graphical Visualizations]
    G --> H[Insights & Interpretation]
    H --> I[End]

    F -->|No| I[End]

Step 1Step 2Step 3Step 4Step 5

• Read and explore all datasets individually.
• Started with bpl.csv, analyzing its structure and features.
• Researched dataset attributes through Google and Bangladesh cricket series data.
• Reviewed relevant Kaggle notebooks to gain insights from existing work.

• Performed basic EDA to understand data distribution.
• Identified and handled missing values.
• Converted categorical features into numerical representations to enrich analysis.
• Examined dataset properties using info() and describe() functions.

• Developed a custom function plotValueCounts(df, col, size=(10,5)) to visualize feature distributions.
• This function, built using Seaborn and Matplotlib, plays a key role in extracting insights.
• Generates count plots with labeled bar values for better interpretability.

• Refined bpl.csv by merging teams with their respective divisions to prevent duplicate counting.
• Addressed inconsistencies:
  • Resolved two tie matches recorded in the dataset.
  • Extracted team runs and wickets from match scoreboards and transformed them into numerical features.

• Implemented automated querying using the plotValueCounts() function.
• Performed grouped analysis of winners by seasons.

❓ KEY QUERIES

• Key Queries on bpl.csv Dataset:
  • Won the toss, took the bat first and won the match.
  • Won the toss, took the bat first.
  • Winning rate by taking bat first vs field first.
• Key Queries on batsman.csv Dataset:
  • Batsman who scored more than 1 century or more in a match.
  • Batsman who faced 50 balls or more in a match.
  • Batsman hit more than 10 fours.
• Key Queries on bowler.csv Dataset:
  • Bowler conceded 50 runs or more
  • Bowler took more than 4 wickets
  • Bowler delivered 4 wide or more

---

🖥 CODE EXPLANATION

plotValueCounts() function

import seaborn as sns
import matplotlib.pyplot as plt

def plotValueCounts(df, col, size=(10, 5)):
    val_counts = df[col].value_counts()
    plt.figure(figsize=size)
    ax = sns.countplot(y=col, data=df, order=val_counts.index, palette="pastel")
    for container in ax.containers:
        ax.bar_label(container, fmt='%d', label_type='edge', padding=3, fontsize=10, color='black')
    plt.title(f"Value Counts of {col}")
    plt.show()

• It displays the visualization graph of value counts of any feature of the dataset.

---

⚖️ PROJECT TRADE-OFFS AND SOLUTIONS

Trade Off 1Trade Off 2Trade Off 3Trade Off 4

• Trade-off: The dataset includes multiple sources, but some features required external validation (e.g., team names, score formats).
• Solution: Standardized team names and resolved inconsistencies by merging duplicate team entries based on divisions.

• Trade-off: Performing detailed feature engineering (e.g., extracting runs, wickets, and match insights) can increase computational overhead.
• Solution: Optimized preprocessing by transforming scoreboard data into structured numerical features, reducing redundant calculations.

• Trade-off: Count plots with high-cardinality features could lead to cluttered or unreadable visualizations.
• Solution: Implemented plotValueCounts() with bar labels and sorting to enhance clarity while retaining key insights.

• Trade-off: Writing fixed queries could limit adaptability when new data is introduced.
• Solution: Developed a function-based querying approach (e.g., plotValueCounts(df, col)) to enable flexible, real-time insights.

---

🖼 SCREENSHOTS

Visualizations and EDA of different features

Toss WinnerWinner TeamsWin Match By WicketsWin Match By Runswinners By SeasonVenue of MatchTop 10 Player of MatchWinning Rate By Toss DecisionBatsman scored more than 1 centuryBowler took more than 4 wickets

---

✅ CONCLUSION

🔑 KEY LEARNINGS

Insights gained from the data

• Extracted meaningful player statistics by analyzing batting and bowling performances across multiple seasons.
• Identified trends in match outcomes based on factors like toss decisions, innings strategies, and scoring patterns.
• Enhanced data visualization techniques to present key insights effectively for better decision-making.

---

🌍 USE CASES

Application 1Application 2

Team Strategy & Selection - Helps coaches analyze player performance for better team selection and match strategies.

Player Performance Tracking - Assists in monitoring player trends to improve training and development.