Make pitch locations represent top 40% of pitches instead of 68%

data.py

@@ -311,6 +311,7 @@ data_df = (
     .with_columns((pl.col('x').is_between(-40, 40) & pl.col('y').is_between(75, 75+100)).alias('heart'))
     .with_columns((pl.col('x').is_between(-80, 80) & pl.col('y').is_between(25, 25+200)  & ~pl.col('heart')).alias('shadow'))
     .with_columns((pl.col('x').is_between(-100, 101) & pl.col('y').is_between(0, 0+251) & ~pl.col('heart') & ~pl.col('shadow')).alias('chase'))
+    .filter(pl.col('ballKind_code') != '-')
 )
 
 if __name__ == '__main__':

plotting.py

@@ -84,7 +84,7 @@ def plot_loc(ax, locs):
     sorted_Z = np.sort(Z_flat)
     sorted_Z_idxs = np.argsort(Z_flat)
     Z_cumsum = (sorted_Z).cumsum()
-    t = Z_flat[sorted_Z_idxs[np.argmin(np.abs(Z_cumsum - (1-0.68)))]]
+    t = Z_flat[sorted_Z_idxs[np.argmin(np.abs(Z_cumsum - (1-0.4)))]]
 
     ax.contourf(X, Y, Z, levels=[t, 1], colors=ball_kind_code_to_color[pitch], alpha=0.5)
     ax.contour(X, Y, Z, levels=t.reshape(1), colors=ball_kind_code_to_color[pitch], alpha=0.75)

stats.py

@@ -164,7 +164,7 @@ def compute_pitcher_stats(data, min_ip='qualified'):
   
 
 def get_pitcher_stats(id, lr=None, game_kind=None, start_date=None, end_date=None, min_ip=1, min_pitches=1, pitch_class_type='specific'):
-  source_data = data_df.filter(pl.col('ballKind_code') != '-')
+  # source_data = data_df.filter(pl.col('ballKind_code') != '-')
 
   # if start_date is not None:
     # source_data = source_data.filter(pl.col('date') >= start_date)
@@ -173,6 +173,7 @@ def get_pitcher_stats(id, lr=None, game_kind=None, start_date=None, end_date=Non
 # 
   # if game_kind is not None:
     # source_data = source_data.filter(pl.col('coarse_game_kind') == game_kind)
+  source_data = data_df
   source_data = filter_data_by_date_and_game_kind(source_data, start_date=start_date, end_date=end_date, game_kind=game_kind)
 
   source_data = (