notebook.community

Edit and run 





In [1]:



    


import pandas as pd
import numpy as np
import matplotlib

from scipy import stats, integrate

import matplotlib.pyplot as plt
%matplotlib inline
import cufflinks as cf

import plotly
plotly.offline.init_notebook_mode()
import plotly.offline as py
import plotly.graph_objs as go
from plotly.graph_objs import *

import seaborn as sns
sns.set(color_codes=True)

print(pd.__version__)



    


















    
















    






0.19.2

















In [6]:



    


df = pd.read_csv('/Users/DanMoeller/git/ncaa-bball-attendance/data/big_12/big_12_2014_2016.csv',sep=",",header='infer')



    











In [7]:



    


df.head(5)



    


















    
Out[7]:










  
    

      
      away_wins
      attendance
      capacity
      home_id
      home_opener
      home_losses
      away_win_pct
      away_losses
      home_win_pct
      home_score
      ...
      game_id
      line
      day
      away_id
      home_rank
      home_wins
      scoring_line
      time
      away_score
      tv_coverage
    

  
  
    

      0
      1
      5706
      10284.0
      239
      0
      0
      1.000
      0
      1.000
      66
      ...
      400506825
      0.0
      tuesday
      2579
      0
      1
      NaN
      20:00
      64
      espn
    

    

      1
      2
      5290
      10284.0
      239
      0
      0
      0.667
      1
      1.000
      87
      ...
      400506826
      0.0
      sunday
      309
      0
      2
      NaN
      22:00
      68
      NaN
    

    

      2
      2
      5185
      10284.0
      239
      0
      0
      0.500
      2
      1.000
      69
      ...
      400506827
      0.0
      thursday
      2127
      0
      3
      NaN
      0:00
      64
      NaN
    

    

      3
      4
      5667
      10284.0
      239
      0
      1
      0.444
      5
      0.889
      91
      ...
      400506831
      0.0
      thursday
      2466
      0
      8
      NaN
      2:30
      84
      NaN
    

    

      4
      3
      5562
      10284.0
      239
      0
      1
      0.273
      8
      0.900
      81
      ...
      400506832
      0.0
      sunday
      2582
      0
      9
      NaN
      22:00
      56
      NaN
    

  



5 rows × 22 columns



















In [8]:



    


df.tail(5)



    


















    
Out[8]:










  
    

      
      away_wins
      attendance
      capacity
      home_id
      home_opener
      home_losses
      away_win_pct
      away_losses
      home_win_pct
      home_score
      ...
      game_id
      line
      day
      away_id
      home_rank
      home_wins
      scoring_line
      time
      away_score
      tv_coverage
    

  
  
    

      534
      2
      8731
      14000.0
      277
      0
      0
      0.222
      7
      1.000
      87
      ...
      400813814
      0.0
      saturday
      338
      0
      6
      NaN
      17:30
      54
      NaN
    

    

      535
      3
      8323
      14000.0
      277
      0
      0
      0.375
      5
      1.000
      100
      ...
      400813815
      -21.5
      sunday
      2433
      0
      8
      133.0
      22:00
      58
      NaN
    

    

      536
      8
      8121
      14000.0
      277
      0
      1
      0.615
      5
      0.900
      84
      ...
      400813816
      -26.0
      tuesday
      2198
      0
      9
      162.0
      0:00
      59
      espnu
    

    

      537
      10
      9567
      9900.0
      259
      0
      4
      0.909
      1
      0.667
      63
      ...
      400813817
      8.0
      wednesday
      277
      0
      8
      148.0
      17:00
      88
      espnu
    

    

      538
      17
      11611
      10151.0
      57
      0
      7
      0.850
      3
      0.650
      88
      ...
      400809416
      1.0
      saturday
      277
      0
      13
      144.0
      17:00
      71
      espn
    

  



5 rows × 22 columns



















In [9]:



    


df.describe()



    


















    
Out[9]:










  
    

      
      away_wins
      attendance
      capacity
      home_id
      home_opener
      home_losses
      away_win_pct
      away_losses
      home_win_pct
      home_score
      away_rank
      game_id
      line
      away_id
      home_rank
      home_wins
      scoring_line
      away_score
    

  
  
    

      count
      539.000000
      539.000000
      504.000000
      539.000000
      539.000000
      539.000000
      539.000000
      539.000000
      539.000000
      539.000000
      539.000000
      5.390000e+02
      539.000000
      539.000000
      539.000000
      539.000000
      142.000000
      539.000000
    

    

      mean
      9.231911
      10169.486085
      12986.073413
      1073.256030
      0.038961
      4.035250
      0.617479
      4.755102
      0.719631
      75.764378
      8.129870
      4.006388e+08
      -1.441558
      1261.717996
      2.230056
      9.831169
      144.507042
      65.528757
    

    

      std
      6.559424
      4033.831734
      3183.263947
      1115.437195
      0.193682
      4.028784
      0.270841
      3.825097
      0.231882
      12.679864
      8.428456
      1.383145e+05
      5.198818
      1129.874449
      5.901527
      5.963900
      8.613651
      12.107197
    

    

      min
      0.000000
      1365.000000
      2500.000000
      2.000000
      0.000000
      0.000000
      0.000000
      0.000000
      0.000000
      42.000000
      0.000000
      4.004969e+08
      -29.500000
      2.000000
      0.000000
      0.000000
      126.000000
      27.000000
    

    

      25%
      3.000000
      6177.000000
      11562.000000
      201.000000
      0.000000
      1.000000
      0.500000
      2.000000
      0.625000
      67.000000
      0.000000
      4.004984e+08
      0.000000
      201.000000
      0.000000
      5.000000
      138.250000
      57.000000
    

    

      50%
      9.000000
      10597.000000
      13611.000000
      251.000000
      0.000000
      3.000000
      0.667000
      4.000000
      0.759000
      76.000000
      6.000000
      4.005858e+08
      0.000000
      299.000000
      0.000000
      10.000000
      144.500000
      65.000000
    

    

      75%
      14.000000
      13239.500000
      15098.000000
      2306.000000
      0.000000
      6.000000
      0.805000
      7.000000
      0.867000
      85.000000
      16.000000
      4.008170e+08
      0.000000
      2435.000000
      0.000000
      14.000000
      149.750000
      73.000000
    

    

      max
      25.000000
      24340.000000
      21750.000000
      2641.000000
      1.000000
      20.000000
      1.000000
      19.000000
      1.000000
      116.000000
      25.000000
      4.008582e+08
      13.000000
      3084.000000
      25.000000
      26.000000
      170.000000
      106.000000
    

  




















In [10]:



    


df.dtypes



    


















    
Out[10]:








away_wins         int64
attendance        int64
capacity        float64
home_id           int64
home_opener       int64
home_losses       int64
away_win_pct    float64
away_losses       int64
home_win_pct    float64
home_score        int64
away_rank         int64
date             object
game_id           int64
line            float64
day              object
away_id           int64
home_rank         int64
home_wins         int64
scoring_line    float64
time             object
away_score        int64
tv_coverage      object
dtype: object

















In [11]:



    


df.columns



    


















    
Out[11]:








Index([u'away_wins', u'attendance', u'capacity', u'home_id', u'home_opener',
       u'home_losses', u'away_win_pct', u'away_losses', u'home_win_pct',
       u'home_score', u'away_rank', u'date', u'game_id', u'line', u'day',
       u'away_id', u'home_rank', u'home_wins', u'scoring_line', u'time',
       u'away_score', u'tv_coverage'],
      dtype='object')

















In [12]:



    


attend = df.attendance.values
cap = df.capacity.values



    











In [13]:



    


# Every Big East games 2014-16 season
plt.plot(cap,attend,"bo")
plt.xlabel("Stadium Capacity")
plt.ylabel("Game Attendance")
plt.show()

# The vertical lines are different teams home games.



    


















    
























In [15]:



    


# only include Big East Conference games
big_12_teams = [239, 66, 2305, 2306, 201, 197, 2628, 251, 2641, 277]
df_conf = df[df.home_id.isin(big_12_teams) & df.away_id.isin(big_12_teams)]



    











In [27]:



    


# Dataframe for each team
# df_baylor = df_conf[df_conf.home_id == 239]
# df_iowa_state = df_conf[df_conf.home_id == 66]
# df_kansas = df_conf[df_conf.home_id == 2305]
# df_kansas_state = df_conf[df_conf.home_id == 2306]
# df_oklahoma = df_conf[df_conf.home_id == 201]
# df_ok_state = df_conf[df_conf.home_id == 197]
# df_tcu = df_conf[df_conf.home_id == 2628]
# df_texas = df_conf[df_conf.home_id == 251]
# df_texas_tech = df_conf[df_conf.home_id == 2641]
# df_west_virginia = df_conf[df_conf.home_id == 277]

df_baylor = df[df.home_id == 239]
df_iowa_state = df[df.home_id == 66]
df_kansas = df[df.home_id == 2305]
df_kansas_state = df[df.home_id == 2306]
df_oklahoma = df[df.home_id == 201]
df_ok_state = df[df.home_id == 197]
df_tcu = df[df.home_id == 2628]
df_texas = df[df.home_id == 251]
df_texas_tech = df[df.home_id == 2641]
df_west_virginia = df[df.home_id == 277]



    











In [28]:



    


attend = df_conf.attendance.values
cap = df_conf.capacity.values
game = df_conf.game_id.values
pct_full = attend / cap
line = df_conf.line.values



    











In [29]:



    


plt.plot(df_baylor.line.values, (df_baylor.attendance.values / df_baylor.capacity.values),"o",color='#00558c',label='Baylor')
plt.plot(df_iowa_state.line.values, (df_iowa_state.attendance.values / df_iowa_state.capacity.values),"o",color='#13294b',label='Iowa State')
plt.plot(df_kansas.line.values, (df_kansas.attendance.values / df_kansas.capacity.values),"o",color='#001F5B',label='Kansas')
plt.plot(df_kansas_state.line.values, (df_kansas_state.attendance.values / df_kansas_state.capacity.values),"o",color='#011e41',label='Kansas St')
plt.plot(df_oklahoma.line.values, (df_oklahoma.attendance.values / df_oklahoma.capacity.values),"ko",label='Oklahoma')
plt.plot(df_ok_state.line.values, (df_ok_state.attendance.values / df_ok_state.capacity.values),"o",color='#FFCC00',label='Ok State')
plt.plot(df_tcu.line.values, (df_tcu.attendance.values / df_tcu.capacity.values),"o",color='#004488',label='TCU')
plt.plot(df_texas.line.values, (df_texas.attendance.values / df_texas.capacity.values),"o",color='#002857',label='Texas')
plt.plot(df_texas_tech.line.values, (df_texas_tech.attendance.values / df_texas_tech.capacity.values),"o",color='#CF102D',label='Texas Tech')
plt.plot(df_west_virginia.line.values, (df_west_virginia.attendance.values / df_west_virginia.capacity.values),"o",color='#3c4982',label='West Virginia')
plt.xlabel("Betting Line")
plt.ylabel("Percent Full")
plt.show()



    


















    
























In [30]:



    


plt.plot(df_baylor.away_win_pct.values, (df_baylor.attendance.values / df_baylor.capacity.values),"o",color='#00558c',label='Baylor')
plt.plot(df_iowa_state.away_win_pct.values, (df_iowa_state.attendance.values / df_iowa_state.capacity.values),"o",color='#13294b',label='Iowa State')
plt.plot(df_kansas.away_win_pct.values, (df_kansas.attendance.values / df_kansas.capacity.values),"o",color='#001F5B',label='Kansas')
plt.plot(df_kansas_state.away_win_pct.values, (df_kansas_state.attendance.values / df_kansas_state.capacity.values),"o",color='#011e41',label='Kansas St')
plt.plot(df_oklahoma.away_win_pct.values, (df_oklahoma.attendance.values / df_oklahoma.capacity.values),"ko",label='Oklahoma')
plt.plot(df_ok_state.away_win_pct.values, (df_ok_state.attendance.values / df_ok_state.capacity.values),"o",color='#FFCC00',label='Ok State')
plt.plot(df_tcu.away_win_pct.values, (df_tcu.attendance.values / df_tcu.capacity.values),"o",color='#004488',label='TCU')
plt.plot(df_texas.away_win_pct.values, (df_texas.attendance.values / df_texas.capacity.values),"o",color='#002857',label='Texas')
plt.plot(df_texas_tech.away_win_pct.values, (df_texas_tech.attendance.values / df_texas_tech.capacity.values),"o",color='#CF102D',label='Texas Tech')
plt.plot(df_west_virginia.away_win_pct.values, (df_west_virginia.attendance.values / df_west_virginia.capacity.values),"o",color='#3c4982',label='West Virginia')
plt.xlabel("Away Win Percentage")
plt.ylabel("Percent Full")
plt.show()



    


















    
























In [31]:



    


plt.plot(df_baylor.home_win_pct.values, (df_baylor.attendance.values / df_baylor.capacity.values),"o",color='#00558c',label='Baylor')
plt.plot(df_iowa_state.home_win_pct.values, (df_iowa_state.attendance.values / df_iowa_state.capacity.values),"o",color='#13294b',label='Iowa State')
plt.plot(df_kansas.home_win_pct.values, (df_kansas.attendance.values / df_kansas.capacity.values),"o",color='#001F5B',label='Kansas')
plt.plot(df_kansas_state.home_win_pct.values, (df_kansas_state.attendance.values / df_kansas_state.capacity.values),"o",color='#011e41',label='Kansas St')
plt.plot(df_oklahoma.home_win_pct.values, (df_oklahoma.attendance.values / df_oklahoma.capacity.values),"ko",label='Oklahoma')
plt.plot(df_ok_state.home_win_pct.values, (df_ok_state.attendance.values / df_ok_state.capacity.values),"o",color='#FFCC00',label='Ok State')
plt.plot(df_tcu.home_win_pct.values, (df_tcu.attendance.values / df_tcu.capacity.values),"o",color='#004488',label='TCU')
plt.plot(df_texas.home_win_pct.values, (df_texas.attendance.values / df_texas.capacity.values),"o",color='#002857',label='Texas')
plt.plot(df_texas_tech.home_win_pct.values, (df_texas_tech.attendance.values / df_texas_tech.capacity.values),"o",color='#CF102D',label='Texas Tech')
plt.plot(df_west_virginia.home_win_pct.values, (df_west_virginia.attendance.values / df_west_virginia.capacity.values),"o",color='#3c4982',label='West Virginia')
plt.xlabel("Home Win Percentage")
plt.ylabel("Percent Full")
plt.show()



    


















    
























In [32]:



    


plt.plot((df_baylor.home_win_pct.values + df_baylor.away_win_pct.values)/2, (df_baylor.attendance.values / df_baylor.capacity.values),"o",color='#00558c',label='Baylor')
plt.plot((df_iowa_state.home_win_pct.values + df_iowa_state.away_win_pct.values)/2, (df_iowa_state.attendance.values / df_iowa_state.capacity.values),"o",color='#13294b',label='Iowa State')
plt.plot((df_kansas.home_win_pct.values + df_kansas.away_win_pct.values)/2, (df_kansas.attendance.values / df_kansas.capacity.values),"o",color='#001F5B',label='Kansas')
plt.plot((df_kansas_state.home_win_pct.values + df_kansas_state.away_win_pct.values)/2, (df_kansas_state.attendance.values / df_kansas_state.capacity.values),"o",color='#011e41',label='Kansas St')
plt.plot((df_oklahoma.home_win_pct.values + df_oklahoma.away_win_pct.values)/2, (df_oklahoma.attendance.values / df_oklahoma.capacity.values),"ko",label='Oklahoma')
plt.plot((df_ok_state.home_win_pct.values + df_ok_state.away_win_pct.values)/2, (df_ok_state.attendance.values / df_ok_state.capacity.values),"o",color='#FFCC00',label='Ok State')
plt.plot((df_tcu.home_win_pct.values + df_tcu.away_win_pct.values)/2, (df_tcu.attendance.values / df_tcu.capacity.values),"o",color='#004488',label='TCU')
plt.plot((df_texas.home_win_pct.values + df_texas.away_win_pct.values)/2, (df_texas.attendance.values / df_texas.capacity.values),"o",color='#002857',label='Texas')
plt.plot((df_texas_tech.home_win_pct.values + df_texas_tech.away_win_pct.values)/2, (df_texas_tech.attendance.values / df_texas_tech.capacity.values),"o",color='#CF102D',label='Texas Tech')
plt.plot((df_west_virginia.home_win_pct.values + df_west_virginia.away_win_pct.values)/2, (df_west_virginia.attendance.values / df_west_virginia.capacity.values),"o",color='#3c4982',label='West Virginia')
plt.xlabel("Average Win Percentage")
plt.ylabel("Percent Full")
plt.show()



    


















    
























In [33]:



    


plt.plot(df_baylor.away_rank.values, (df_baylor.attendance.values / df_baylor.capacity.values),"o",color='#00558c',label='Baylor')
plt.plot(df_iowa_state.away_rank.values, (df_iowa_state.attendance.values / df_iowa_state.capacity.values),"o",color='#13294b',label='Iowa State')
plt.plot(df_kansas.away_rank.values, (df_kansas.attendance.values / df_kansas.capacity.values),"o",color='#001F5B',label='Kansas')
plt.plot(df_kansas_state.away_rank.values, (df_kansas_state.attendance.values / df_kansas_state.capacity.values),"o",color='#011e41',label='Kansas St')
plt.plot(df_oklahoma.away_rank.values, (df_oklahoma.attendance.values / df_oklahoma.capacity.values),"ko",label='Oklahoma')
plt.plot(df_ok_state.away_rank.values, (df_ok_state.attendance.values / df_ok_state.capacity.values),"o",color='#FFCC00',label='Ok State')
plt.plot(df_tcu.away_rank.values, (df_tcu.attendance.values / df_tcu.capacity.values),"o",color='#004488',label='TCU')
plt.plot(df_texas.away_rank.values, (df_texas.attendance.values / df_texas.capacity.values),"o",color='#002857',label='Texas')
plt.plot(df_texas_tech.away_rank.values, (df_texas_tech.attendance.values / df_texas_tech.capacity.values),"o",color='#CF102D',label='Texas Tech')
plt.plot(df_west_virginia.away_rank.values, (df_west_virginia.attendance.values / df_west_virginia.capacity.values),"o",color='#3c4982',label='West Virginia')
plt.xlabel("Away Rank")
plt.ylabel("Percent Full")
plt.show()



    


















    
























In [34]:



    


plt.plot(df_baylor.home_rank.values, (df_baylor.attendance.values / df_baylor.capacity.values),"o",color='#00558c',label='Baylor')
plt.plot(df_iowa_state.home_rank.values, (df_iowa_state.attendance.values / df_iowa_state.capacity.values),"o",color='#13294b',label='Iowa State')
plt.plot(df_kansas.home_rank.values, (df_kansas.attendance.values / df_kansas.capacity.values),"o",color='#001F5B',label='Kansas')
plt.plot(df_kansas_state.home_rank.values, (df_kansas_state.attendance.values / df_kansas_state.capacity.values),"o",color='#011e41',label='Kansas St')
plt.plot(df_oklahoma.home_rank.values, (df_oklahoma.attendance.values / df_oklahoma.capacity.values),"ko",label='Oklahoma')
plt.plot(df_ok_state.home_rank.values, (df_ok_state.attendance.values / df_ok_state.capacity.values),"o",color='#FFCC00',label='Ok State')
plt.plot(df_tcu.home_rank.values, (df_tcu.attendance.values / df_tcu.capacity.values),"o",color='#004488',label='TCU')
plt.plot(df_texas.home_rank.values, (df_texas.attendance.values / df_texas.capacity.values),"o",color='#002857',label='Texas')
plt.plot(df_texas_tech.home_rank.values, (df_texas_tech.attendance.values / df_texas_tech.capacity.values),"o",color='#CF102D',label='Texas Tech')
plt.plot(df_west_virginia.home_rank.values, (df_west_virginia.attendance.values / df_west_virginia.capacity.values),"o",color='#3c4982',label='West Virginia')
plt.xlabel("Home Rank")
plt.ylabel("Percent Full")
plt.show()



    


















    
























In [35]:



    


plt.plot(df_baylor.scoring_line.values, (df_baylor.attendance.values / df_baylor.capacity.values),"o",color='#00558c',label='Baylor')
plt.plot(df_iowa_state.scoring_line.values, (df_iowa_state.attendance.values / df_iowa_state.capacity.values),"o",color='#13294b',label='Iowa State')
plt.plot(df_kansas.scoring_line.values, (df_kansas.attendance.values / df_kansas.capacity.values),"o",color='#001F5B',label='Kansas')
plt.plot(df_kansas_state.scoring_line.values, (df_kansas_state.attendance.values / df_kansas_state.capacity.values),"o",color='#011e41',label='Kansas St')
plt.plot(df_oklahoma.scoring_line.values, (df_oklahoma.attendance.values / df_oklahoma.capacity.values),"ko",label='Oklahoma')
plt.plot(df_ok_state.scoring_line.values, (df_ok_state.attendance.values / df_ok_state.capacity.values),"o",color='#FFCC00',label='Ok State')
plt.plot(df_tcu.scoring_line.values, (df_tcu.attendance.values / df_tcu.capacity.values),"o",color='#004488',label='TCU')
plt.plot(df_texas.scoring_line.values, (df_texas.attendance.values / df_texas.capacity.values),"o",color='#002857',label='Texas')
plt.plot(df_texas_tech.scoring_line.values, (df_texas_tech.attendance.values / df_texas_tech.capacity.values),"o",color='#CF102D',label='Texas Tech')
plt.plot(df_west_virginia.scoring_line.values, (df_west_virginia.attendance.values / df_west_virginia.capacity.values),"o",color='#3c4982',label='West Virginia')
plt.xlabel("Scoring Line")
plt.ylabel("Percent Full")
plt.show()



    


















    
























In [ ]:



    






    











In [ ]:

Content source: danmoeller/ncaa-bball-attendance

Similar notebooks:

notebook.community | gallery | about