PremiershipPlayerAnalysis

Using Python to scrape some basic player information from www.premierleague.com and then use Pandas to analyse said data. Note : My understanding is the squad data on this site can change at any time so your results might be different

Improvement : Calculate age to finer degree than just years

The was developed in Jupyter Notebook and this walkthrough willl assume you are doing the same

Once you have ran the scraping

original = pd.DataFrame(playersList) # Convert the data scraped into a Pandas DataFrame 

original.to_csv('premiershipplayers.csv') # Keep a back up of the data to save time later if required 

df2 = original.copy() # Working copy of the DataFrame (just in case) 


df2.info()


   
    
RangeIndex: 578 entries, 0 to 577
Data columns (total 11 columns):
 #   Column       Non-Null Count  Dtype 
---  ------       --------------  ----- 
 0   club         578 non-null    object
 1   name         578 non-null    object
 2   shirtNo      572 non-null    object
 3   nationality  562 non-null    object
 4   dob          562 non-null    object
 5   height       500 non-null    object
 6   weight       474 non-null    object
 7   appearances  578 non-null    object
 8   goals        578 non-null    object
 9   wins         578 non-null    object
 10  losses       578 non-null    object
dtypes: object(11)
memory usage: 49.8+ KB

   

*** A total of 578 player. ***

6 without shirt number

16 without nationality listed

16 without dob listed

78 without height listed

104 without weight listed

Cleanup Data

1. Remove spaces and newline from dob, appearances, goals, wins and losses columns

2. Change type of dob to date

3. change type of appearances, goals, wins, losses to int

    df2['dob'] = df2['dob'].str.replace('\n','').str.strip(' ')
    df2['appearances'] = df2['appearances'].str.replace('\n','').str.strip(' ')
    df2['goals'] = df2['goals'].str.replace('\n','').str.strip(' ')
    df2['wins'] = df2['wins'].str.replace('\n','').str.strip(' ')
    df2['losses'] = df2['losses'].str.replace('\n','').str.strip(' ')
   
    # change type of dob, appearances, goals, wins, losses
    from datetime import  date
   
    df2['dob'] = pd.to_datetime(df2['dob'],format='%d/%m/%Y').dt.date
    df2["appearances"] = pd.to_numeric(df2["appearances"])
    df2["goals"] = pd.to_numeric(df2["goals"])
    df2["wins"] = pd.to_numeric(df2["wins"])
    df2["losses"] = pd.to_numeric(df2["losses"])
    df2['height'] = df2['height'].str[:-2]
    df2["height"] = pd.to_numeric(df2["height"])
    
    
    # Create age column
   
    today = date.today()
   
    def age(born):
        if born:
            return today.year - born.year - ((today.month, 
                                          today.day) < (born.month, 
                                                        born.day))
        else:
            return np.nan
   
    df2['age'] = df2['dob'].apply(age)
   

10 Oldest Players

    df2.sort_values('age',ascending=False).head(10)

10 Youngest Players

    df2.sort_values('age',ascending=True).head(10)

Squad Sizes

    df2.groupby(['club'])['club'].count().sort_values(ascending=False)

Team's Average Player Age

    plt.ylim([20, 30])
    df2.groupby(['club'])['age'].mean().sort_values(ascending=False).plot.bar()

Burnley appear to not only have one of the highest average player ages but also the owest number of registered players

Top 10 Premiership Appearances

    df2.sort_values('appearances',ascending=False).head(10)

Collective Premiership Appearances per Club

    df2.groupby(['club'])['appearances'].sum().sort_values(ascending=False)

    df2.groupby(['club'])['appearances'].sum().sort_values(ascending=False).plot.bar()

10 Tallest Playes

    df2.sort_values('height',ascending=False).head(10)

10 Shortest Playes

    df2.sort_values('height',ascending=True).head(10)

Nationality totals of Players

    pd.set_option('display.max_rows', 100)
    df.groupby(['nationality'])['club'].count().sort_values(ascending=False)

Nationality totals per club

    pd.set_option('display.max_rows', 500)
    df.groupby(['club','nationality'])['nationality'].count()