groupby()グルーピングした値の平均値を算出し、最下部に列として追加して表示したい

Question

前提・実現したいこと

groupby()グルーピングした値の平均値を算出し、その値を最下列にインデックス名を”Average”として追加したいです。
イメージは、各Playerの平均値やmax値などを.agg()を用いて算出したものの下に、各項目のチームアベレージを追記する感じです。

ご教授の程よろしくお願い致します。

発生している問題・エラーメッセージ

#データフレームを二つに分けて算出し結合してみたのですが、平均値の方、カラムとインデックスが入れ替わっておりきれいに結合できませんでした。

                    (Player, count)  (Speed, mean)  (Speed, 90th)  \
(Angle, AA416)                  NaN            NaN            NaN   
(Angle, mean)                   NaN            NaN            NaN   
(Angle, std)                    NaN            NaN            NaN   
(Efficiency, mean)              NaN            NaN            NaN   
(Player, count)                 NaN            NaN            NaN   
(Speed, 90th)                   NaN            NaN            NaN   
(Speed, amax)                   NaN            NaN            NaN   
(Speed, mean)                   NaN            NaN            NaN   
(Speed, std)                    NaN            NaN            NaN   
Player1                        47.0           53.5           87.9   
Player2                        37.0           50.2           84.7   
Player3                        42.0           52.5           90.5   
Player4                        43.0           49.1           90.5   
Player5                        31.0           57.0           94.6   

                    (Speed, amax)  (Speed, std)  (Angle, mean)  (Angle, std)  \
(Angle, AA416)                NaN           NaN            NaN           NaN   
(Angle, mean)                 NaN           NaN            NaN           NaN   
(Angle, std)                  NaN           NaN            NaN           NaN   
(Efficiency, mean)            NaN           NaN            NaN           NaN   
(Player, count)               NaN           NaN            NaN           NaN   
(Speed, 90th)                 NaN           NaN            NaN           NaN   
(Speed, amax)                 NaN           NaN            NaN           NaN   
(Speed, mean)                 NaN           NaN            NaN           NaN   
(Speed, std)                  NaN           NaN            NaN           NaN   
Player1                      93.3          26.3            0.0          12.1   
Player2                      93.4          26.5           -1.3          12.4   
Player3                      98.1          26.9           -3.0          10.5   
Player4                      96.9          27.9           -0.4          11.6   
Player5                      99.8          29.0            1.3          11.3   

                    (Angle, AA416)  (Efficiency, mean)    0  
(Angle, AA416)                 NaN                 NaN 29.0  
(Angle, mean)                  NaN                 NaN -0.7  
(Angle, std)                   NaN                 NaN 11.6  
(Efficiency, mean)             NaN                 NaN 50.1  
(Player, count)                NaN                 NaN 40.0  
(Speed, 90th)                  NaN                 NaN 89.6  
(Speed, amax)                  NaN                 NaN 96.3  
(Speed, mean)                  NaN                 NaN 52.5  
(Speed, std)                   NaN                 NaN 27.3  
Player1                       23.4                51.4  NaN  
Player2                       27.0                52.8  NaN  
Player3                       21.4                48.9  NaN  
Player4                       27.9                49.9  NaN  
Player5                       45.2                47.5  NaN  

該当のソースコード

import pandas as pd
import numpy as np
pd.options.display.float_format="{:.1f}".format
start,end = "2021/4/1","2021/4/30"
dates = pd.date_range(start=start,end=end,freq="D")
players = [f"Player{i}"for i in range(1,6)]

N = 200
dates = np.random.choice(dates,size=N)
dates.sort()
players=np.random.choice(players,size=N)

df = pd.DataFrame({
        "Date": dates,
      "Player": players,
       "Speed": np.random.sample(N) * 100.0,
       "Angle": np.random.sample(N) * 40.0 - 20.0,
  "Efficiency": np.random.sample(N) * 100.0,
})

def percentile(n):
    def percentile_(x):
        return np.percentile(x, n)
    percentile_.__name__ = '%sth' % n
    return percentile_

def AA416(angle):
    return 100*angle[(angle>=4.0)&(angle<16.0)].count()/angle.count()

dfx = df.groupby("Player").agg({
               "Player":"count",
     "Speed":[np.mean,percentile(90),np.max,np.std],
   "Angle":[np.mean,np.std,AA416],
   "Efficiency":[np.mean],
})
dfy = dfx.mean()
dfz = pd.concat([dfx,dfy],axis=1)
print(dfz)

試したこと

データフレームを二つに分けて結合してみました。

補足情報（FW/ツールのバージョンなど）

python3.9.4,vscode

Comments

[melian]

どうもイメージが掴みづらいのですが、もしかして、
dfx.loc['Average'] = dfx.mean()
ということをしたいのでしょうか？

[8960]

これです…。難しく説明してしまってましたね。
基礎的なところかもしれませんが、いつもありがとうございます。
改めて回答に記載いただけると幸いです。

Answer 1

dfx.loc['Average'] = dfx.mean()

print(dfx)

#
        Player Speed                Angle            Efficiency
         count  mean 90th amax  std  mean  std AA416       mean
Player                                                         
Player1   36.0  51.5 88.9 93.9 30.2  -0.6 11.4  36.1       51.8
Player2   38.0  49.4 83.2 98.5 26.5  -0.1 12.5  23.7       55.7
Player3   40.0  51.3 91.8 98.3 29.6   2.9 11.9  40.0       52.4
Player4   38.0  48.7 90.8 99.8 30.9  -0.2 10.3  21.1       45.5
Player5   48.0  47.7 90.2 96.2 31.6   0.8 12.1  33.3       54.0
Average   40.0  49.7 89.0 97.3 29.8   0.5 11.6  30.8       51.9