The purpose of this exploratory data analysis is to identify potential investments in one of the top-performing English Premier League clubs. Several established clubs are already owned by competitors, so the management of the firm is seeking to identify clubs with great potential for future success.
The dataset comprises crucial information about all the clubs that have participated in premier league tournaments. Here is a data dictionary that outlines the information we have about the clubs and their performance.
⚽ Club: Name of the football club
⚽ Matches: Number of matches the club has played in the Premier League
⚽ Wins: Number of matches won by the club in the Premier League
⚽ Loss: Number of matches lost by the club in the Premier League
⚽ Clean Sheets: Number of matches in which the club has prevented the opposing side from scoring
⚽ Draws: Number of matches drawn by the club in the Premier League
⚽ Team Launch: Year in which the club was founded
⚽ Winners: Number of times the club has won the Premier League
⚽ Runners-up: Number of times the club has finished as runners-up in the Premier League
⚽ lastplayed_pl: Year in which the team last played in the Premier League
<aside> 📌 Dataset has been downloaded from the internet using multiple sources. All the credit for the dataset goes to the original creator of the data
</aside>
<aside> ➡️ Data cleaning is the process of identifying and correcting or removing errors, inconsistencies, and inaccuracies in a dataset. Some variables may need data type conversion, converting the existing data to a particular format, and not all null values need to be replaced with mean/median/mode. We need to treat the null values based on domain knowledge and formulate a default value.
</aside>
Reference from the project: Winners and Runner column (null values are treated based on domain knowledge i.e. default value as zero), last_played (formatted to ‘YYYY’).
<aside> ➡️ Feature engineering refers to the process of using domain knowledge to select and transform the raw data into relevant information that can be used in the analysis. Some variables might be not relevant which gets transformed.
</aside>
Reference from the project: Win Rate, Loss Rate, Drawn Rate, Clean Sheet Rate, Average No. of Goals from Wins, Loss, Draws, Clean Sheets, Goals and Matches Played.
<aside> ➡️ Outliers are unusual values in your dataset which can distort statistical analyses. We identify and remove them during EDA. But, that’s not always the case. Removing the outliers might not provide us with the relevant insight we are looking for.
</aside>
Reference from the project: The outliers are the ones that showed the clubs that are significant from others, with the high potential to win.
<aside> ➡️ We built a framework by assigning scores to each club from the inference obtained during EDA, which led us to a conclusion. To avoid being deceived by the weightage result, a thorough examination of the whole dataset should be done by checking if there might be any columns or correlations to re-consider in the dataset.
</aside>
Reference from the project: We recommended the second highest team as the first team was inconsistent and not currently active which was discovered after closely examining the last_played column (contains the information when it played last).