NBA player analysis

1. get data

data=pd.read_csv("./data/nba_2017_nba_players_with_salary.csv")
data.head(10)

2. Data analysis

2.1 Data relevance

data_cor = data.loc[:, ['RPM', 'AGE', 'SALARY_MILLIONS', 'ORB', 'DRB', 'TRB',
                       'AST', 'STL', 'BLK', 'TOV', 'PF', 'POINTS', 'GP', 'MPG', 'ORPM', 'DRPM']]
data_cor.head()

#  Get the size comparison of the data correlation in the table 
corr=data_cor.corr()
#h Get the correlation between the two columns of data 
corr.head()

plt.figure(figsize=(20,8),dpi=100)
sns.heatmap(corr,square=True,linewidths=0.1,annot=True)  # According to relevance   Get the thermodynamic diagram of two columns of data  

2.2 Basic data ranking analysis

# Rank according to efficiency value 
data.loc[:,["PLAYER", "RPM", "AGE"]].sort_values(by="RPM",ascending=False)   #ascending null 

#  Rank by player salary 
data.loc[:,["PLAYER", "RPM", "SALARY_MILLIONS"]].sort_values(by="SALARY_MILLIONS",ascending=False)   #ascending null 

2.3 Seaborn Three commonly used data visualization methods

2.3.1 Univariate

#  utilize seaborn Medium distplot Draw a picture to see the salary of players 、 Efficiency value 、 The distribution of age 
sns.set_style("darkgrid")#  Set the style of the image 
plt.figure(figsize=(10,10))  # Set the size of the canvas 
plt.subplot(3,1,1)  # Usage method ：subplot（m,n,p） perhaps subplot（m n p）. subplot It's a tool for drawing multiple pictures on one plane . among ,m It's a graph arrangement m That's ok ,n The graphs are arranged in n Column , That is the whole figure There is n The two graphs are arranged in a row , altogether m That's ok , If m=2 That is to say 2 Line graph .p Show the location of the diagram ,p=1 Represents the first position from left to right, top to bottom .
sns.distplot(data["SALARY_MILLIONS"])
plt.ylabel("salary")

plt.subplot(3,1,2)  # Usage method ：subplot（m,n,p） perhaps subplot（m n p）. subplot It's a tool for drawing multiple pictures on one plane . among ,m It's a graph arrangement m That's ok ,n The graphs are arranged in n Column , That is the whole figure There is n The two graphs are arranged in a row , altogether m That's ok , If m=2 That is to say 2 Line graph .p Show the location of the diagram ,p=1 Represents the first position from left to right, top to bottom .
sns.distplot(data["RPM"])
plt.ylabel("RPM")

plt.subplot(3,1,3)  # Usage method ：subplot（m,n,p） perhaps subplot（m n p）. subplot It's a tool for drawing multiple pictures on one plane . among ,m It's a graph arrangement m That's ok ,n The graphs are arranged in n Column , That is the whole figure There is n The two graphs are arranged in a row , altogether m That's ok , If m=2 That is to say 2 Line graph .p Show the location of the diagram ,p=1 Represents the first position from left to right, top to bottom .
sns.distplot(data["AGE"])
plt.ylabel("AGE")

2.3.2 Bivariate

sns.jointplot(data.AGE,data.SALARY_MILLIONS,kind="hex")

2.3.3 Multivariable

multi_data = data.loc[:, ['RPM','SALARY_MILLIONS','AGE','POINTS']]
multi_data.head()

sns.pairplot(multi_data)

2.3.4 Some visualization practices of derived variables - Take age as an example

def age_cut(df):
    ''' Age division '''
    if df.AGE<=24:
        return "young"
    elif df.AGE>=30:
        return "old"
    else:
        return "best"

data["age_cut"]=data.apply(lambda x:age_cut(x),axis=1)   
#apply effect ： The first parameter is the pointer of the custom function , Pass the function in , then apply The function will data A line in is passed into the function as a parameter ,axis=1  Represents a line by line calculation  
data.head()       

#  Easy to count 
data["cut"]=1
data.loc[data.age_cut=="best"].SALARY_MILLIONS.head()

#  Analyze player salary and efficiency based on age 
sns.set_style("darkgrid")
plt.figure(figsize=(10,10),dpi=100)
plt.title("RPM and Salary")

x1=data.loc[data.age_cut=="old"].SALARY_MILLIONS
y1=data.loc[data.age_cut=="old"].RPM
plt.plot(x1,y1,"^")

x2=data.loc[data.age_cut=="best"].SALARY_MILLIONS
y2=data.loc[data.age_cut=="best"].RPM
plt.plot(x2,y2,"^")

x3=data.loc[data.age_cut=="young"].SALARY_MILLIONS
y3=data.loc[data.age_cut=="young"].RPM
plt.plot(x3,y3,".")

multi_data2 = data.loc[:, ['RPM','POINTS','TRB','AST','STL','BLK','age_cut']]
sns.pairplot(multi_data2,hue="age_cut")  # hue Continue to display in color according to this column 

3. Team data analysis

3.1 Team salary ranking

data.groupby(by="age_cut").agg({
    "SALARY_MILLIONS":np.mean})  # Group by age , Then aggregate according to the average 

data_team=data.groupby(by="TEAM").agg({
    "SALARY_MILLIONS":np.mean})  # Group according to the team , Then aggregate according to the average 
data_team.sort_values(by="SALARY_MILLIONS",ascending=False).head()#  Sort by salary in descending order  ascending=False For the descending order 

3.2 According to the team By age group , The number of players on the list is the same , Then arrange them in descending order according to the efficiency value

data_RMP=data.groupby(by=["TEAM","age_cut"]).agg({
    "SALARY_MILLIONS":np.mean,"RPM":np.mean,"PLAYER":np.size})
data_RMP.head()

data_RMP.sort_values(by=["PLAYER","RPM"],ascending=False).head(10)

3.3 Rank according to the comprehensive strength of the team

data_rpm2=data.groupby(by=["TEAM"],as_index=False).agg({
    "SALARY_MILLIONS":np.mean,
                                                        "RPM":np.mean,
                                                        "PLAYER":np.size,
                                                        "POINTS":np.mean,
                                                        "eFG%":np.mean,
                                                        "MPG":np.mean,
                                                        "AGE":np.mean})
data_rpm2.head()

data_rpm2.sort_values(by="RPM",ascending=False)

3.4 Use box diagram and violin diagram for data analysis

sns.set_style("whitegrid")#  Set the style of the drawing board 
# Get the corresponding data 
data_team2=data[data.TEAM.isin(['GS','CLE','SA','LAC','OKC','UTAH',"CHA",'TOR','NO','BOS'])]
data_team2.head()

# Make corresponding drawings 
plt.figure(figsize=(20,10))# Set the size of the sketchpad 
plt.subplot(3,1,1)
sns.boxplot(x="TEAM",y="SALARY_MILLIONS",data=data_team2)

plt.subplot(3,1,2)
sns.boxplot(x="TEAM",y="AGE",data=data_team2)

plt.subplot(3,1,3)
sns.boxplot(x="TEAM",y="MPG",data=data_team2)

# Draw a picture of the violin 
sns.set_style("whitegrid")
plt.figure(figsize=(20,10))
plt.subplot(3,1,1)
sns.violinplot(x="TEAM",y="3P%",data=data_team2)

plt.subplot(3,1,2)
sns.violinplot(x="TEAM",y="eFG%",data=data_team2)

plt.subplot(3,1,3)
sns.violinplot(x="TEAM",y="POINTS",data=data_team2)