Add velo stats to pitch leaderboard

pitch_leaderboard.py

@@ -10,9 +10,9 @@ from stats import compute_pitch_stats, filter_data_by_date_and_game_kind
 from convert import ball_kind, ball_kind_to_color, get_text_color_from_color, team_names_short_to_color, get_text_color_from_team
 from plotting import stat_cmap
 
-STATS = ['Count', 'Usage', 'Swing%', 'Z-Swing%', 'Chase%', 'Contact%', 'Z-Contact%', 'O-Contact%', 'SwStr%', 'Whiff%', 'CSW%', 'GB%', 'FB%', 'LD%', 'Zone%', 'Arm%', 'Glove%', 'High%', 'Low%', 'MM%']
+STATS = ['Count', 'Usage', 'Avg Velo', 'Max Velo', 'Swing%', 'Z-Swing%', 'Chase%', 'Contact%', 'Z-Contact%', 'O-Contact%', 'SwStr%', 'Whiff%', 'CSW%', 'GB%', 'FB%', 'LD%', 'Zone%', 'Arm%', 'Glove%', 'High%', 'Low%', 'MM%']
 PCT_STATS = ['Usage', 'Swing%', 'Z-Swing%', 'Chase%', 'Contact%', 'Z-Contact%', 'O-Contact%', 'SwStr%', 'Whiff%', 'CSW%', 'GB%', 'FB%', 'LD%', 'Zone%', 'Arm%', 'Glove%', 'High%', 'Low%', 'MM%']
-STATS_WITH_PCTLS = ['Swing%', 'Z-Swing%', 'Chase%', 'Contact%', 'Z-Contact%', 'O-Contact%', 'SwStr%', 'Whiff%', 'CSW%', 'GB%', 'FB%', 'LD%', 'Zone%']
+STATS_WITH_PCTLS = ['Avg Velo', 'Max Velo', 'Swing%', 'Z-Swing%', 'Chase%', 'Contact%', 'Z-Contact%', 'O-Contact%', 'SwStr%', 'Whiff%', 'CSW%', 'GB%', 'FB%', 'LD%', 'Zone%']
 COLUMNS = ['Pitcher', 'Team', 'Pitch', 'Pitch (General)'] + STATS
 
 PITCH_TYPES = [pitch_type for pitch_type in ball_kind.values() if pitch_type != '-']
@@ -69,7 +69,18 @@ def gr_create_pitch_leaderboard(start_date, end_date, min_pitches, pitcher_lr='B
         compute_pitch_stats(data, player_type='pitcher', min_pitches=min_pitches, pitch_class_type='specific')
         .filter(pl.col('qualified') & (pl.col('ballKind').is_in(include_pitches)))
         .drop('pitId', 'ballKind_code', 'qualified')
-        .rename({'pitcher_name': 'Pitcher', 'pitcher_team_name_short': 'Team', 'count': 'Count', 'usage': 'Usage', 'ballKind': 'Pitch', 'general_ballKind': 'Pitch (General)'})
+        .rename({
+          'pitcher_name': 'Pitcher', 
+          'pitcher_team_name_short': 'Team', 
+          'count': 'Count', 
+          'usage': 'Usage', 
+          'ballKind': 'Pitch', 
+          'general_ballKind': 'Pitch (General)', 
+          'Avg MPH': 'Avg Velo', 
+          'Max MPH': 'Max Velo',
+          'Avg MPH_pctl': 'Avg Velo_pctl', 
+          'Max MPH_pctl': 'Max Velo_pctl'
+        })
         .with_columns(
           pl.col(stat).mul(100)
           for stat in PCT_STATS
@@ -85,7 +96,7 @@ def gr_create_pitch_leaderboard(start_date, end_date, min_pitches, pitcher_lr='B
     styling_row = []
     for col, item in zip(pitch_stats[COLUMNS].columns, row):
       _styling = 'font-size: 0.75em; '
-      if f'{col}_pctl' in pitch_stats:
+      if col in STATS_WITH_PCTLS:
         r, g, b = (stat_cmap([pitch_stats[f'{col}_pctl'][i]])[0, :3]*255).astype(np.uint8)
         styling_row.append(f'background-color: rgba({r}, {g}, {b})')
       elif col == 'Team':
@@ -100,9 +111,11 @@ def gr_create_pitch_leaderboard(start_date, end_date, min_pitches, pitcher_lr='B
   display_value = []
   for row in pitch_stats[COLUMNS].iter_rows():
     display_value_row = []
-    for item in row:
-      if isinstance(item, float):
+    for col, item in zip(COLUMNS, row):
+      if col in PCT_STATS:
         display_value_row.append(f'{item:.1f}%')
+      elif isinstance(item, float):
+        display_value_row.append(f'{item:.1f}')
       else:
         display_value_row.append(item)
     display_value.append(display_value_row)

plotting.py

@@ -58,8 +58,8 @@ y_range = np.arange(0, 250+1)
 X, Y = np.meshgrid(x_range, y_range)
 
 
-def fit_pred_kde(data):
-  kde = gaussian_kde(data)
+def fit_pred_kde(data, **kwargs):
+  kde = gaussian_kde(data, **kwargs)
   Z = kde(np.concat((X, Y)).reshape(2, -1)).reshape(*X.shape)
   return Z
 
@@ -169,7 +169,6 @@ font = load_google_font('Saira Extra Condensed', weight='medium')
 
 def create_pitcher_overview_card(id, season, dpi=300):
   data = get_card_data(id, start_date=date(season, 1, 1), end_date=date(season, 12, 31), game_kind='Regular Season', min_pitches=100, pitch_class_type='general')
-  print(data)
 
   fig = plt.figure(figsize=(1080/300, 1350/300), dpi=dpi)
   gs = fig.add_gridspec(8, 6, height_ratios=[1, 1, 1.5, 6, 1, 3, 1, 0.5])

stats.py

@@ -72,12 +72,17 @@ def compute_pitch_stats(data, player_type, pitch_class_type, min_pitches=1):
   pitch_name_col = 'ballKind' if pitch_class_type == 'specific' else 'general_ballKind'
   pitch_stats = (
       data
+      .with_columns((pl.col('ballSpeed') / 1.609).round(1).alias('mph'))
       .group_by(id_col, pitch_col, 'pitcher_team_name_short')
       .agg(
           pl.first(name_col),
           *([pl.first('general_ballKind')] if pitch_class_type == 'specific' else []),
           pl.first(pitch_name_col),
           pl.len().alias('count'),
+          pl.when(pl.col('x').is_not_null() & pl.col('y').is_not_null() & (pl.col('ballSpeed') > 0)).then('ballSpeed').mean().alias('Avg KPH'),
+          pl.col('ballSpeed').max().alias('Max KPH'),
+          pl.when(pl.col('x').is_not_null() & pl.col('y').is_not_null() & (pl.col('ballSpeed') > 0)).then('mph').mean().round(1).alias('Avg MPH'),
+          pl.col('mph').max().alias('Max MPH'),
           pl.col('aux_bresult').struct.field('batType').drop_nulls().value_counts(normalize=True),
           (pl.col('swing').sum() / pl.col('pitch').sum()).alias('Swing%'),
           ((pl.col('swing') & pl.col('zone')).sum() / pl.col('pitch').sum()).alias('Z-Swing%'),
@@ -97,7 +102,7 @@ def compute_pitch_stats(data, player_type, pitch_class_type, min_pitches=1):
       )
       .with_columns(
           (pl.col('count')/pl.sum('count').over('pitId')).alias('usage'),
-          (pl.col('count') >= min_pitches).alias('qualified')
+          (pl.col('count') >= min_pitches).alias('qualified'),
       )
       .explode('batType')
       .unnest('batType')
@@ -111,7 +116,7 @@ def compute_pitch_stats(data, player_type, pitch_class_type, min_pitches=1):
       .drop('G', 'F', 'B', 'P', 'L', 'null')
       .with_columns(
           (pl.when(pl.col('qualified')).then(pl.col(stat)).rank(descending=((stat in ['FB%', 'LD%'] or 'Contact%' in stat)))/pl.when(pl.col('qualified')).then(pl.col(stat)).count()).alias(f'{stat}_pctl')
-          for stat in ['Swing%', 'Z-Swing%', 'Chase%', 'Contact%', 'Z-Contact%', 'O-Contact%', 'SwStr%', 'Whiff%', 'CSW%', 'GB%', 'FB%', 'LD%', 'Zone%']
+          for stat in ['Avg KPH', 'Max KPH', 'Avg MPH', 'Max MPH', 'Swing%', 'Z-Swing%', 'Chase%', 'Contact%', 'Z-Contact%', 'O-Contact%', 'SwStr%', 'Whiff%', 'CSW%', 'GB%', 'FB%', 'LD%', 'Zone%']
       )
       .rename({pitch_col: 'ballKind_code', pitch_name_col: 'ballKind'} if pitch_class_type == 'general' else {})
       .sort(id_col, 'count', descending=[False, True])