Upload 4 files

functions/df_update.py

@@ -0,0 +1,472 @@
+import polars as pl
+import numpy as np
+import joblib
+
+loaded_model = joblib.load('joblib_model/barrel_model.joblib')
+in_zone_model = joblib.load('joblib_model/in_zone_model_knn_20240410.joblib')
+attack_zone_model = joblib.load('joblib_model/model_attack_zone.joblib')
+xwoba_model = joblib.load('joblib_model/xwoba_model.joblib')
+px_model = joblib.load('joblib_model/linear_reg_model_x.joblib')
+pz_model = joblib.load('joblib_model/linear_reg_model_z.joblib')
+
+
+class df_update:
+    def __init__(self):
+        pass
+
+    def update(self, df_clone: pl.DataFrame):
+
+        df = df_clone.clone()
+        # Assuming px_model is defined and df is your DataFrame
+        hit_codes = ['single',
+            'double','home_run', 'triple']
+
+        ab_codes = ['single', 'strikeout', 'field_out',
+            'grounded_into_double_play', 'fielders_choice', 'force_out',
+            'double', 'field_error', 'home_run', 'triple',
+            'double_play',
+            'fielders_choice_out', 'strikeout_double_play',
+            'other_out','triple_play']
+
+
+        obp_true_codes = ['single', 'walk',
+            'double','home_run', 'triple',
+            'hit_by_pitch', 'intent_walk']
+
+        obp_codes = ['single', 'strikeout', 'walk', 'field_out',
+            'grounded_into_double_play', 'fielders_choice', 'force_out',
+            'double', 'sac_fly', 'field_error', 'home_run', 'triple',
+            'hit_by_pitch', 'double_play', 'intent_walk',
+            'fielders_choice_out', 'strikeout_double_play',
+            'sac_fly_double_play',
+            'other_out','triple_play']
+
+
+        contact_codes = ['In play, no out',
+                'Foul', 'In play, out(s)',
+            'In play, run(s)',
+            'Foul Bunt']
+
+        bip_codes = ['In play, no out', 'In play, run(s)','In play, out(s)']
+
+
+        conditions_barrel = [
+            df['launch_speed'].is_null(),
+            (df['launch_speed'] * 1.5 - df['launch_angle'] >= 117) & 
+            (df['launch_speed'] + df['launch_angle'] >= 124) & 
+            (df['launch_speed'] >= 98) & 
+            (df['launch_angle'] >= 4) & (df['launch_angle'] <= 50)
+        ]
+        choices_barrel = [False, True]
+
+        conditions_tb = [
+            (df['event_type'] == 'single'),
+            (df['event_type'] == 'double'),
+            (df['event_type'] == 'triple'),
+            (df['event_type'] == 'home_run')
+        ]
+        choices_tb = [1, 2, 3, 4]
+
+
+        conditions_woba = [
+            df['event_type'].is_in(['strikeout', 'field_out', 'sac_fly', 'force_out', 'grounded_into_double_play', 'fielders_choice', 'field_error', 'sac_bunt', 'double_play', 'fielders_choice_out', 'strikeout_double_play', 'sac_fly_double_play', 'other_out']),
+            df['event_type'] == 'walk',
+            df['event_type'] == 'hit_by_pitch',
+            df['event_type'] == 'single',
+            df['event_type'] == 'double',
+            df['event_type'] == 'triple',
+            df['event_type'] == 'home_run'
+        ]
+        choices_woba = [0, 0.689, 0.720, 0.881, 1.254, 1.589, 2.048]
+
+        woba_codes = ['strikeout', 'field_out', 'single', 'walk', 'hit_by_pitch', 'double', 'sac_fly', 'force_out', 'home_run', 'grounded_into_double_play', 'fielders_choice', 'field_error', 'triple', 'sac_bunt', 'double_play', 'fielders_choice_out', 'strikeout_double_play', 'sac_fly_double_play', 'other_out']
+
+        pitch_cat = {'FA': 'Fastball',
+                    'FF': 'Fastball',
+                    'FT': 'Fastball',
+                    'FC': 'Fastball',
+                    'FS': 'Off-Speed',
+                    'FO': 'Off-Speed',
+                    'SI': 'Fastball',
+                    'ST': 'Breaking',
+                    'SL': 'Breaking',
+                    'CU': 'Breaking',
+                    'KC': 'Breaking',
+                    'SC': 'Off-Speed',
+                    'GY': 'Off-Speed',
+                    'SV': 'Breaking',
+                    'CS': 'Breaking',
+                    'CH': 'Off-Speed',
+                    'KN': 'Off-Speed',
+                    'EP': 'Breaking',
+                    'UN': None,
+                    'IN': None,
+                    'PO': None,
+                    'AB': None,
+                    'AS': None,
+                    'NP': None}
+
+
+        df = df.with_columns([
+            pl.when(df['type_ab'].is_not_null()).then(1).otherwise(0).alias('pa'),
+            pl.when(df['is_pitch']).then(1).otherwise(0).alias('pitches'),
+            pl.when(df['sz_top'] == 0).then(None).otherwise(df['sz_top']).alias('sz_top'),
+            pl.when(df['sz_bot'] == 0).then(None).otherwise(df['sz_bot']).alias('sz_bot'),
+            pl.when(df['zone'] > 0).then(df['zone'] < 10).otherwise(None).alias('in_zone'),
+            pl.Series(px_model.predict(df[['x']].fill_null(0).to_numpy())[:, 0]).alias('px_predict'),
+            pl.Series(pz_model.predict(df[['y']].fill_null(0).to_numpy())[:, 0] + 3.2).alias('pz_predict'),
+            pl.Series(in_zone_model.predict(df[['px','pz','sz_top','sz_bot']].fill_null(0).to_numpy())[:]).alias('in_zone_predict'),
+            pl.Series(attack_zone_model.predict(df[['px','pz','sz_top','sz_bot']].fill_null(0).to_numpy())[:]).alias('attack_zone_predict'),
+            pl.when(df['event_type'].is_in(hit_codes)).then(True).otherwise(False).alias('hits'),
+            pl.when(df['event_type'].is_in(ab_codes)).then(True).otherwise(False).alias('ab'),
+            pl.when(df['event_type'].is_in(obp_true_codes)).then(True).otherwise(False).alias('on_base'),
+            pl.when(df['event_type'].is_in(obp_codes)).then(True).otherwise(False).alias('obp'),
+            pl.when(df['play_description'].is_in(bip_codes)).then(True).otherwise(False).alias('bip'),
+            pl.when(conditions_barrel[0]).then(choices_barrel[0]).when(conditions_barrel[1]).then(choices_barrel[1]).otherwise(None).alias('barrel'),
+            pl.when(df['launch_angle'].is_null()).then(False).when((df['launch_angle'] >= 8) & (df['launch_angle'] <= 32)).then(True).otherwise(None).alias('sweet_spot'),
+            pl.when(df['launch_speed'].is_null()).then(False).when(df['launch_speed'] >= 94.5).then(True).otherwise(None).alias('hard_hit'),
+            pl.when(conditions_tb[0]).then(choices_tb[0]).when(conditions_tb[1]).then(choices_tb[1]).when(conditions_tb[2]).then(choices_tb[2]).when(conditions_tb[3]).then(choices_tb[3]).otherwise(None).alias('tb'),
+            pl.when(conditions_woba[0]).then(choices_woba[0]).when(conditions_woba[1]).then(choices_woba[1]).when(conditions_woba[2]).then(choices_woba[2]).when(conditions_woba[3]).then(choices_woba[3]).when(conditions_woba[4]).then(choices_woba[4]).when(conditions_woba[5]).then(choices_woba[5]).when(conditions_woba[6]).then(choices_woba[6]).otherwise(None).alias('woba'),
+            pl.when((df['play_code'] == 'S') | (df['play_code'] == 'W') | (df['play_code'] == 'T')).then(1).otherwise(0).alias('whiffs'),
+            pl.when((df['play_code'] == 'S') | (df['play_code'] == 'W') | (df['play_code'] == 'T') | (df['play_code'] == 'C')).then(1).otherwise(0).alias('csw'),
+            pl.when(pl.col('is_swing').cast(pl.Boolean)).then(1).otherwise(0).alias('swings'),
+            pl.col('event_type').is_in(['strikeout','strikeout_double_play']).alias('k'),
+            pl.col('event_type').is_in(['walk', 'intent_walk']).alias('bb'),
+            pl.lit(None).alias('attack_zone'),
+            pl.lit(None).alias('woba_pred'),
+            pl.lit(None).alias('woba_pred_contact')
+
+        ])
+
+        df = df.with_columns([
+            pl.when(df['event_type'].is_in(woba_codes)).then(1).otherwise(None).alias('woba_codes'),
+            pl.when(df['event_type'].is_in(woba_codes)).then(1).otherwise(None).alias('xwoba_codes'),
+            pl.when((pl.col('tb') >= 0)).then(df['woba']).otherwise(None).alias('woba_contact'),
+            pl.when(pl.col('px').is_null()).then(pl.col('px_predict')).otherwise(pl.col('px')).alias('px'),
+            pl.when(pl.col('pz').is_null()).then(pl.col('pz_predict')).otherwise(pl.col('pz')).alias('pz'),
+            pl.when(pl.col('in_zone').is_null()).then(pl.col('in_zone_predict')).otherwise(pl.col('in_zone')).alias('in_zone'),
+            pl.when(df['launch_speed'].is_null()).then(None).otherwise(df['barrel']).alias('barrel'),
+            pl.lit('average').alias('average'),
+                pl.when(pl.col('in_zone') == False).then(True).otherwise(False).alias('out_zone'),
+            pl.when((pl.col('in_zone') == True) & (pl.col('swings') == 1)).then(True).otherwise(False).alias('zone_swing'),
+            pl.when((pl.col('in_zone') == True) & (pl.col('swings') == 1) & (pl.col('whiffs') == 0)).then(True).otherwise(False).alias('zone_contact'),
+            pl.when((pl.col('in_zone') == False) & (pl.col('swings') == 1)).then(True).otherwise(False).alias('ozone_swing'),
+            pl.when((pl.col('in_zone') == False) & (pl.col('swings') == 1) & (pl.col('whiffs') == 0)).then(True).otherwise(False).alias('ozone_contact'),
+            pl.when(pl.col('event_type').str.contains('strikeout')).then(True).otherwise(False).alias('k'),
+            pl.when(pl.col('event_type').is_in(['walk', 'intent_walk'])).then(True).otherwise(False).alias('bb'),
+            pl.when(pl.col('attack_zone').is_null()).then(pl.col('attack_zone_predict')).otherwise(pl.col('attack_zone')).alias('attack_zone'),
+            
+
+        ])
+
+        df = df.with_columns([
+            (df['k'].cast(pl.Float32) - df['bb'].cast(pl.Float32)).alias('k_minus_bb'),
+            (df['bb'].cast(pl.Float32) - df['k'].cast(pl.Float32)).alias('bb_minus_k'),
+            (df['launch_speed'] > 0).alias('bip_div'),
+            (df['attack_zone'] == 0).alias('heart'),
+            (df['attack_zone'] == 1).alias('shadow'),
+            (df['attack_zone'] == 2).alias('chase'),
+            (df['attack_zone'] == 3).alias('waste'),
+            ((df['attack_zone'] == 0) & (df['swings'] == 1)).alias('heart_swing'),
+            ((df['attack_zone'] == 1) & (df['swings'] == 1)).alias('shadow_swing'),
+            ((df['attack_zone'] == 2) & (df['swings'] == 1)).alias('chase_swing'),
+            ((df['attack_zone'] == 3) & (df['swings'] == 1)).alias('waste_swing'),
+            ((df['attack_zone'] == 0) & (df['whiffs'] == 1)).alias('heart_whiff'),
+            ((df['attack_zone'] == 1) & (df['whiffs'] == 1)).alias('shadow_whiff'),
+            ((df['attack_zone'] == 2) & (df['whiffs'] == 1)).alias('chase_whiff'),
+            ((df['attack_zone'] == 3) & (df['whiffs'] == 1)).alias('waste_whiff')
+        ])
+
+
+        [0, 0.689, 0.720, 0.881, 1.254, 1.589, 2.048]
+
+        df = df.with_columns([
+                pl.Series(
+                    [sum(x) for x in xwoba_model.predict_proba(df[['launch_angle', 'launch_speed']].fill_null(0).to_numpy()[:]) * ([0, 0.881, 1.254, 1.589, 2.048])]
+                ).alias('woba_pred_predict')
+            ])
+
+        df = df.with_columns([
+            pl.when(pl.col('event_type').is_in(['walk'])).then(0.689)
+            .when(pl.col('event_type').is_in(['hit_by_pitch'])).then(0.720)
+            .when(pl.col('event_type').is_in(['strikeout', 'strikeout_double_play'])).then(0)
+            .otherwise(pl.col('woba_pred_predict')).alias('woba_pred_predict')
+        ])
+
+        df = df.with_columns([
+            pl.when(pl.col('woba_codes').is_null()).then(None).otherwise(pl.col('woba_pred_predict')).alias('woba_pred'),
+            pl.when(pl.col('bip')!=1).then(None).otherwise(pl.col('woba_pred_predict')).alias('woba_pred_contact'),
+        ])
+
+        df = df.with_columns([
+            pl.when(pl.col('trajectory').is_in(['bunt_popup'])).then(pl.lit('popup'))
+            .when(pl.col('trajectory').is_in(['bunt_grounder'])).then(pl.lit('ground_ball'))
+            .when(pl.col('trajectory').is_in(['bunt_line_drive'])).then(pl.lit('line_drive'))
+            .when(pl.col('trajectory').is_in([''])).then(pl.lit(None))
+            .otherwise(pl.col('trajectory')).alias('trajectory')
+        ])
+
+
+        # Create one-hot encoded columns for the trajectory column
+        dummy_df = df.select(pl.col('trajectory')).to_dummies()
+
+        # Rename the one-hot encoded columns
+        dummy_df = dummy_df.rename({
+            'trajectory_fly_ball': 'trajectory_fly_ball',
+            'trajectory_ground_ball': 'trajectory_ground_ball',
+            'trajectory_line_drive': 'trajectory_line_drive',
+            'trajectory_popup': 'trajectory_popup'
+        })
+
+        # Ensure the columns are present in the DataFrame
+        for col in ['trajectory_fly_ball', 'trajectory_ground_ball', 'trajectory_line_drive', 'trajectory_popup']:
+            if col not in dummy_df.columns:
+                dummy_df = dummy_df.with_columns(pl.lit(0).alias(col))
+
+        # Join the one-hot encoded columns back to the original DataFrame
+        df = df.hstack(dummy_df)
+
+        # Check if 'trajectory_null' column exists and drop it
+        if 'trajectory_null' in df.columns:
+            df = df.drop('trajectory_null')
+            
+        return df
+
+    # Assuming df is your Polars DataFrame
+    def update_summary(self, df: pl.DataFrame, pitcher: bool = True) -> pl.DataFrame:
+        """
+        Update summary statistics for pitchers or batters.
+
+        Parameters:
+        df (pl.DataFrame): The input Polars DataFrame containing player statistics.
+        pitcher (bool): A flag indicating whether to calculate statistics for pitchers (True) or batters (False).
+
+        Returns:
+        pl.DataFrame: A Polars DataFrame with aggregated and calculated summary statistics.
+        """
+
+        # Determine the position based on the pitcher flag
+        if pitcher:
+            position = 'pitcher'
+        else:
+            position = 'batter'
+
+        # Group by position_id and position_name, then aggregate various statistics
+        df_summ = df.group_by([f'{position}_id', f'{position}_name']).agg([
+            pl.col('pa').sum().alias('pa'),
+            pl.col('ab').sum().alias('ab'),
+            pl.col('obp').sum().alias('obp_pa'),
+            pl.col('hits').sum().alias('hits'),
+            pl.col('on_base').sum().alias('on_base'),
+            pl.col('k').sum().alias('k'),
+            pl.col('bb').sum().alias('bb'),
+            pl.col('bb_minus_k').sum().alias('bb_minus_k'),
+            pl.col('csw').sum().alias('csw'),
+            pl.col('bip').sum().alias('bip'),
+            pl.col('bip_div').sum().alias('bip_div'),
+            pl.col('tb').sum().alias('tb'),
+            pl.col('woba').sum().alias('woba'),
+            pl.col('woba_contact').sum().alias('woba_contact'),
+            pl.col('woba_pred').sum().alias('xwoba'),
+            pl.col('woba_pred_contact').sum().alias('xwoba_contact'),
+            pl.col('woba_codes').sum().alias('woba_codes'),
+            pl.col('xwoba_codes').sum().alias('xwoba_codes'),
+            pl.col('hard_hit').sum().alias('hard_hit'),
+            pl.col('barrel').sum().alias('barrel'),
+            pl.col('sweet_spot').sum().alias('sweet_spot'),
+            pl.col('launch_speed').max().alias('max_launch_speed'),
+            pl.col('launch_speed').quantile(0.90).alias('launch_speed_90'),
+            pl.col('launch_speed').mean().alias('launch_speed'),
+            pl.col('launch_angle').mean().alias('launch_angle'),
+            pl.col('is_pitch').sum().alias('pitches'),
+            pl.col('swings').sum().alias('swings'),
+            pl.col('in_zone').sum().alias('in_zone'),
+            pl.col('out_zone').sum().alias('out_zone'),
+            pl.col('whiffs').sum().alias('whiffs'),
+            pl.col('zone_swing').sum().alias('zone_swing'),
+            pl.col('zone_contact').sum().alias('zone_contact'),
+            pl.col('ozone_swing').sum().alias('ozone_swing'),
+            pl.col('ozone_contact').sum().alias('ozone_contact'),
+            pl.col('trajectory_ground_ball').sum().alias('ground_ball'),
+            pl.col('trajectory_line_drive').sum().alias('line_drive'),
+            pl.col('trajectory_fly_ball').sum().alias('fly_ball'),
+            pl.col('trajectory_popup').sum().alias('pop_up'),
+            pl.col('attack_zone').count().alias('attack_zone'),
+            pl.col('heart').sum().alias('heart'),
+            pl.col('shadow').sum().alias('shadow'),
+            pl.col('chase').sum().alias('chase'),
+            pl.col('waste').sum().alias('waste'),
+            pl.col('heart_swing').sum().alias('heart_swing'),
+            pl.col('shadow_swing').sum().alias('shadow_swing'),
+            pl.col('chase_swing').sum().alias('chase_swing'),
+            pl.col('waste_swing').sum().alias('waste_swing'),
+            pl.col('heart_whiff').sum().alias('heart_whiff'),
+            pl.col('shadow_whiff').sum().alias('shadow_whiff'),
+            pl.col('chase_whiff').sum().alias('chase_whiff'),
+            pl.col('waste_whiff').sum().alias('waste_whiff')
+        ])
+
+        # Add calculated columns to the summary DataFrame
+        df_summ = df_summ.with_columns([
+            (pl.col('hits') / pl.col('ab')).alias('avg'),
+            (pl.col('on_base') / pl.col('obp_pa')).alias('obp'),
+            (pl.col('tb') / pl.col('ab')).alias('slg'),
+            (pl.col('on_base') / pl.col('obp_pa') + pl.col('tb') / pl.col('ab')).alias('ops'),
+            (pl.col('k') / pl.col('pa')).alias('k_percent'),
+            (pl.col('bb') / pl.col('pa')).alias('bb_percent'),
+            (pl.col('bb_minus_k') / pl.col('pa')).alias('bb_minus_k_percent'),
+            (pl.col('bb') / pl.col('k')).alias('bb_over_k_percent'),
+            (pl.col('csw') / pl.col('pitches')).alias('csw_percent'),
+            (pl.col('sweet_spot') / pl.col('bip_div')).alias('sweet_spot_percent'),
+            (pl.col('woba') / pl.col('woba_codes')).alias('woba_percent'),
+            (pl.col('woba_contact') / pl.col('bip')).alias('woba_percent_contact'),
+            (pl.col('hard_hit') / pl.col('bip_div')).alias('hard_hit_percent'),
+            (pl.col('barrel') / pl.col('bip_div')).alias('barrel_percent'),
+            (pl.col('zone_contact') / pl.col('zone_swing')).alias('zone_contact_percent'),
+            (pl.col('zone_swing') / pl.col('in_zone')).alias('zone_swing_percent'),
+            (pl.col('in_zone') / pl.col('pitches')).alias('zone_percent'),
+            (pl.col('ozone_swing') / (pl.col('pitches') - pl.col('in_zone'))).alias('chase_percent'),
+            (pl.col('ozone_contact') / pl.col('ozone_swing')).alias('chase_contact'),
+            (pl.col('swings') / pl.col('pitches')).alias('swing_percent'),
+            (pl.col('whiffs') / pl.col('swings')).alias('whiff_rate'),
+            (pl.col('whiffs') / pl.col('pitches')).alias('swstr_rate'),
+            (pl.col('ground_ball') / pl.col('bip')).alias('ground_ball_percent'),
+            (pl.col('line_drive') / pl.col('bip')).alias('line_drive_percent'),
+            (pl.col('fly_ball') / pl.col('bip')).alias('fly_ball_percent'),
+            (pl.col('pop_up') / pl.col('bip')).alias('pop_up_percent'),
+            (pl.col('heart') / pl.col('attack_zone')).alias('heart_zone_percent'),
+            (pl.col('shadow') / pl.col('attack_zone')).alias('shadow_zone_percent'),
+            (pl.col('chase') / pl.col('attack_zone')).alias('chase_zone_percent'),
+            (pl.col('waste') / pl.col('attack_zone')).alias('waste_zone_percent'),
+            (pl.col('heart_swing') / pl.col('heart')).alias('heart_zone_swing_percent'),
+            (pl.col('shadow_swing') / pl.col('shadow')).alias('shadow_zone_swing_percent'),
+            (pl.col('chase_swing') / pl.col('chase')).alias('chase_zone_swing_percent'),
+            (pl.col('waste_swing') / pl.col('waste')).alias('waste_zone_swing_percent'),
+            (pl.col('heart_whiff') / pl.col('heart_swing')).alias('heart_zone_whiff_percent'),
+            (pl.col('shadow_whiff') / pl.col('shadow_swing')).alias('shadow_zone_whiff_percent'),
+            (pl.col('chase_whiff') / pl.col('chase_swing')).alias('chase_zone_whiff_percent'),
+            (pl.col('waste_whiff') / pl.col('waste_swing')).alias('waste_zone_whiff_percent'),
+            (pl.col('xwoba') / pl.col('xwoba_codes')).alias('xwoba_percent'),
+            (pl.col('xwoba_contact') / pl.col('bip')).alias('xwoba_percent_contact')
+        ])
+
+        return df_summ
+    
+
+
+
+
+    
+    # Assuming df is your Polars DataFrame
+    def update_summary_select(self, df: pl.DataFrame, selection: list) -> pl.DataFrame:
+        """
+        Update summary statistics for pitchers or batters.
+
+        Parameters:
+        df (pl.DataFrame): The input Polars DataFrame containing player statistics.
+        pitcher (bool): A flag indicating whether to calculate statistics for pitchers (True) or batters (False).
+
+        Returns:
+        pl.DataFrame: A Polars DataFrame with aggregated and calculated summary statistics.
+        """
+
+        # Group by position_id and position_name, then aggregate various statistics
+        df_summ = df.group_by(selection).agg([
+            pl.col('pa').sum().alias('pa'),
+            pl.col('ab').sum().alias('ab'),
+            pl.col('obp').sum().alias('obp_pa'),
+            pl.col('hits').sum().alias('hits'),
+            pl.col('on_base').sum().alias('on_base'),
+            pl.col('k').sum().alias('k'),
+            pl.col('bb').sum().alias('bb'),
+            pl.col('bb_minus_k').sum().alias('bb_minus_k'),
+            pl.col('csw').sum().alias('csw'),
+            pl.col('bip').sum().alias('bip'),
+            pl.col('bip_div').sum().alias('bip_div'),
+            pl.col('tb').sum().alias('tb'),
+            pl.col('woba').sum().alias('woba'),
+            pl.col('woba_contact').sum().alias('woba_contact'),
+            pl.col('woba_pred').sum().alias('xwoba'),
+            pl.col('woba_pred_contact').sum().alias('xwoba_contact'),
+            pl.col('woba_codes').sum().alias('woba_codes'),
+            pl.col('xwoba_codes').sum().alias('xwoba_codes'),
+            pl.col('hard_hit').sum().alias('hard_hit'),
+            pl.col('barrel').sum().alias('barrel'),
+            pl.col('sweet_spot').sum().alias('sweet_spot'),
+            pl.col('launch_speed').max().alias('max_launch_speed'),
+            pl.col('launch_speed').quantile(0.90).alias('launch_speed_90'),
+            pl.col('launch_speed').mean().alias('launch_speed'),
+            pl.col('launch_angle').mean().alias('launch_angle'),
+            pl.col('is_pitch').sum().alias('pitches'),
+            pl.col('swings').sum().alias('swings'),
+            pl.col('in_zone').sum().alias('in_zone'),
+            pl.col('out_zone').sum().alias('out_zone'),
+            pl.col('whiffs').sum().alias('whiffs'),
+            pl.col('zone_swing').sum().alias('zone_swing'),
+            pl.col('zone_contact').sum().alias('zone_contact'),
+            pl.col('ozone_swing').sum().alias('ozone_swing'),
+            pl.col('ozone_contact').sum().alias('ozone_contact'),
+            pl.col('trajectory_ground_ball').sum().alias('ground_ball'),
+            pl.col('trajectory_line_drive').sum().alias('line_drive'),
+            pl.col('trajectory_fly_ball').sum().alias('fly_ball'),
+            pl.col('trajectory_popup').sum().alias('pop_up'),
+            pl.col('attack_zone').count().alias('attack_zone'),
+            pl.col('heart').sum().alias('heart'),
+            pl.col('shadow').sum().alias('shadow'),
+            pl.col('chase').sum().alias('chase'),
+            pl.col('waste').sum().alias('waste'),
+            pl.col('heart_swing').sum().alias('heart_swing'),
+            pl.col('shadow_swing').sum().alias('shadow_swing'),
+            pl.col('chase_swing').sum().alias('chase_swing'),
+            pl.col('waste_swing').sum().alias('waste_swing'),
+            pl.col('heart_whiff').sum().alias('heart_whiff'),
+            pl.col('shadow_whiff').sum().alias('shadow_whiff'),
+            pl.col('chase_whiff').sum().alias('chase_whiff'),
+            pl.col('waste_whiff').sum().alias('waste_whiff')
+        ])
+
+        # Add calculated columns to the summary DataFrame
+        df_summ = df_summ.with_columns([
+            (pl.col('hits') / pl.col('ab')).alias('avg'),
+            (pl.col('on_base') / pl.col('obp_pa')).alias('obp'),
+            (pl.col('tb') / pl.col('ab')).alias('slg'),
+            (pl.col('on_base') / pl.col('obp_pa') + pl.col('tb') / pl.col('ab')).alias('ops'),
+            (pl.col('k') / pl.col('pa')).alias('k_percent'),
+            (pl.col('bb') / pl.col('pa')).alias('bb_percent'),
+            (pl.col('bb_minus_k') / pl.col('pa')).alias('bb_minus_k_percent'),
+            (pl.col('bb') / pl.col('k')).alias('bb_over_k_percent'),
+            (pl.col('csw') / pl.col('pitches')).alias('csw_percent'),
+            (pl.col('sweet_spot') / pl.col('bip_div')).alias('sweet_spot_percent'),
+            (pl.col('woba') / pl.col('woba_codes')).alias('woba_percent'),
+            (pl.col('woba_contact') / pl.col('bip')).alias('woba_percent_contact'),
+            (pl.col('hard_hit') / pl.col('bip_div')).alias('hard_hit_percent'),
+            (pl.col('barrel') / pl.col('bip_div')).alias('barrel_percent'),
+            (pl.col('zone_contact') / pl.col('zone_swing')).alias('zone_contact_percent'),
+            (pl.col('zone_swing') / pl.col('in_zone')).alias('zone_swing_percent'),
+            (pl.col('in_zone') / pl.col('pitches')).alias('zone_percent'),
+            (pl.col('ozone_swing') / (pl.col('pitches') - pl.col('in_zone'))).alias('chase_percent'),
+            (pl.col('ozone_contact') / pl.col('ozone_swing')).alias('chase_contact'),
+            (pl.col('swings') / pl.col('pitches')).alias('swing_percent'),
+            (pl.col('whiffs') / pl.col('swings')).alias('whiff_rate'),
+            (pl.col('whiffs') / pl.col('pitches')).alias('swstr_rate'),
+            (pl.col('ground_ball') / pl.col('bip')).alias('ground_ball_percent'),
+            (pl.col('line_drive') / pl.col('bip')).alias('line_drive_percent'),
+            (pl.col('fly_ball') / pl.col('bip')).alias('fly_ball_percent'),
+            (pl.col('pop_up') / pl.col('bip')).alias('pop_up_percent'),
+            (pl.col('heart') / pl.col('attack_zone')).alias('heart_zone_percent'),
+            (pl.col('shadow') / pl.col('attack_zone')).alias('shadow_zone_percent'),
+            (pl.col('chase') / pl.col('attack_zone')).alias('chase_zone_percent'),
+            (pl.col('waste') / pl.col('attack_zone')).alias('waste_zone_percent'),
+            (pl.col('heart_swing') / pl.col('heart')).alias('heart_zone_swing_percent'),
+            (pl.col('shadow_swing') / pl.col('shadow')).alias('shadow_zone_swing_percent'),
+            (pl.col('chase_swing') / pl.col('chase')).alias('chase_zone_swing_percent'),
+            (pl.col('waste_swing') / pl.col('waste')).alias('waste_zone_swing_percent'),
+            (pl.col('heart_whiff') / pl.col('heart_swing')).alias('heart_zone_whiff_percent'),
+            (pl.col('shadow_whiff') / pl.col('shadow_swing')).alias('shadow_zone_whiff_percent'),
+            (pl.col('chase_whiff') / pl.col('chase_swing')).alias('chase_zone_whiff_percent'),
+            (pl.col('waste_whiff') / pl.col('waste_swing')).alias('waste_zone_whiff_percent'),
+            (pl.col('xwoba') / pl.col('xwoba_codes')).alias('xwoba_percent'),
+            (pl.col('xwoba_contact') / pl.col('bip')).alias('xwoba_percent_contact')
+        ])
+
+        return df_summ

functions/rolling_batter_functions.py

@@ -0,0 +1,338 @@
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import seaborn as sns
+import numpy as np
+from scipy.stats import gaussian_kde
+import matplotlib
+from matplotlib.ticker import MaxNLocator
+from matplotlib.gridspec import GridSpec
+from scipy.stats import zscore
+import math
+import matplotlib
+from adjustText import adjust_text
+import matplotlib.ticker as mtick
+import pandas as pd
+from matplotlib.pyplot import text
+import inflect
+
+
+colour_palette = ['#FFB000','#648FFF','#785EF0',
+                  '#DC267F','#FE6100','#3D1EB2','#894D80','#16AA02','#B5592B','#A3C1ED']
+
+plot_dict = {
+'k':{'x_axis':'Plate Appearances','y_axis':'K%','title':'K%','x_value':'k','x_range':[0.0,0.1,0.2,0.3,0.4],'percent':True,'percentile_label':'k_percent','flip_p':True,'percentile':False,'avg_adjust':False},
+'bb':{'x_axis':'Plate Appearances','y_axis':'BB%','title':'BB%','x_value':'bb','x_range':[0.0,0.1,0.2,0.3],'percent':True,'percentile_label':'bb_percent','flip_p':False,'percentile':False,'avg_adjust':False},
+'bb_minus_k':{'x_axis':'Plate Appearances','y_axis':'BB-K%','title':'BB-K%','x_value':'bb_minus_k','x_range':[-0.3,-0.2,-0.1,0,0.1,0.2],'percent':True,'percentile_label':'bb_minus_k_percent','flip_p':False,'percentile':False,'avg_adjust':False},
+'csw':{'x_axis':'Pitches','y_axis':'CSW%','title':'CSW%','x_value':'csw','x_range':[.2,.25,.3,.35,.4],'percent':True,'percentile_label':'csw_percent','flip_p':True,'percentile':False,'avg_adjust':False},
+'woba':{'x_axis':'wOBA PA','y_axis':'wOBA','title':'wOBA','x_value':'woba','x_range':[.20,.30,.40,.50],'percent':False,'percentile_label':'woba_percent','flip_p':False,'percentile':False,'avg_adjust':True},
+'launch_speed':{'x_axis':'Balls In Play','y_axis':'Exit Velocity','title':'Exit Velocity','x_value':'launch_speed','x_range':[85,90,95,100],'percent':False,'percentile_label':'launch_speed','flip_p':False,'percentile':False,'avg_adjust':False},
+'launch_speed_90':{'x_axis':'Balls In Play','y_axis':'90th Percentile Exit Velocity','title':'90th Percentile Exit Velocity','x_value':'launch_speed','x_range':[95,100,105,110,115],'percent':False,'percentile_label':'launch_speed_90','flip_p':False,'percentile':True,'avg_adjust':False},              
+'hard_hit':{'x_axis':'Balls In Play','y_axis':'HardHit%','title':'HardHit%','x_value':'hard_hit','x_range':[0.2,0.3,0.4,0.5,0.6,0.7],'percent':True,'percentile_label':'hard_hit_percent','flip_p':False,'percentile':False,'avg_adjust':False},
+'sweet_spot':{'x_axis':'Balls In Play','y_axis':'SweetSpot%','title':'SweetSpot%','x_value':'sweet_spot','x_range':[0.2,0.3,0.4,0.5],'percent':True,'percentile_label':'sweet_spot_percent','flip_p':False,'percentile':False,'avg_adjust':False},
+'launch_angle':{'x_axis':'Balls In Play','y_axis':'Launch Angle','title':'Launch Angle','x_value':'launch_angle','x_range':[-20,-10,0,10,20],'percent':False,'percentile_label':'launch_angle','flip_p':False,'percentile':False,'avg_adjust':False},
+'barrel':{'x_axis':'Balls In Play','y_axis':'Barrel%','title':'Barrel%','x_value':'barrel','x_range':[0,0.05,0.10,.15,.20],'percent':True,'percentile_label':'barrel_percent','flip_p':False,'percentile':False,'avg_adjust':False},
+'zone_percent':{'x_axis':'Pitches','y_axis':'Zone%','title':'Zone%','x_value':'in_zone','x_range':[0.3,0.4,0.5,0.6,0.7],'percent':True,'percentile_label':'zone_percent','flip_p':False,'percentile':False,'avg_adjust':False},
+'swing_percent':{'x_axis':'Pitches','y_axis':'Swing%','title':'Swing%','x_value':'swings','x_range':[0.2,0.3,0.4,0.5,0.6,0.7,0.8],'percent':True,'percentile_label':'swing_percent','flip_p':False,'percentile':False,'avg_adjust':False},
+'whiff_percent':{'x_axis':'Swings','y_axis':'Whiff%','title':'Whiff%','x_value':'whiffs','x_range':[0.0,0.1,0.2,0.3,0.4,0.5],'percent':True,'percentile_label':'whiff_rate','flip_p':True,'percentile':False,'avg_adjust':False},
+'sw_str':{'x_axis':'Pitches','y_axis':'SwStr%','title':'SwStr%','x_value':'whiffs','x_range':[0.0,0.05,0.1,0.15,0.2,0.25],'percent':True,'percentile_label':'swstr_rate','flip_p':True,'percentile':False,'avg_adjust':False},
+'zone_swing':{'x_axis':'In-Zone Pitches','y_axis':'Z-Swing%','title':'Z-Swing%','x_value':'zone_swing','x_range':[0.3,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1],'percent':True,'percentile_label':'zone_swing_percent','flip_p':False,'percentile':False,'avg_adjust':False},
+'zone_contact':{'x_axis':'In-Zone Swings','y_axis':'Z-Contact%','title':'Z-Contact%','x_value':'zone_contact','x_range':[0.5,0.6,0.7,0.8,0.9,1],'percent':True,'percentile_label':'zone_contact_percent','flip_p':False,'percentile':False,'avg_adjust':False},
+'chase_percent':{'x_axis':'Out-of-Zone Pitches','y_axis':'O-Swing%','title':'O-Swing%','x_value':'ozone_swing','x_range':[0.0,0.1,0.2,0.3,0.4,0.5],'percent':True,'percentile_label':'chase_percent','flip_p':True,'percentile':False,'avg_adjust':False},
+'chase_contact':{'x_axis':'Out-of-Zone Swings','y_axis':'O-Contact%','title':'O-Contact%','x_value':'ozone_contact','x_range':[0.2,0.3,0.4,0.5,0.6,0.7,0.8],'percent':True,'percentile_label':'chase_contact','flip_p':False,'percentile':False,'avg_adjust':False},}
+
+
+level_dict = {'MLB':'MLB','AAA':'AAA','AA':'AA','A+':'A+','A':'A'}
+
+
+woba_list = ['woba']
+pa_list = ['k','bb','bb_minus_k']
+balls_in_play_list = ['hard_hit','launch_speed','launch_speed_90','launch_angle','barrel','sweet_spot']
+pitches_list = ['zone_percent','swing_percent','sw_str','csw']
+swings_list = ['whiff_percent']
+in_zone_pitches_list = ['zone_swing']
+in_zone_swings_list = ['zone_contact']
+out_zone_pitches_list = ['chase_percent']
+out_zone_swings_list = ['chase_contact']
+
+plot_dict_small = {
+'k':'K%',
+'bb':'BB%',
+'bb_minus_k':'BB-K%',
+'csw':'CSW%',
+'woba':'wOBA',
+'launch_speed':'Exit Velocity',
+'launch_speed_90':'90th Percentile Exit Velocity',
+'hard_hit':'HardHit%',
+'sweet_spot':'SweetSpot%',
+'launch_angle':'Launch Angle',
+'zone_percent':'Zone%',
+'barrel':'Barrel%',
+'swing_percent':'Swing%',
+'whiff_percent':'Whiff%',
+'sw_str':'SwStr%',
+'zone_swing':'Z-Swing%',
+'zone_contact':'Z-Contact%',
+'chase_percent':'O-Swing%',
+'chase_contact':'O-Contact%',}
+
+
+
+def rolling_plot(df,df_summ,player_id,stat_id,batter_dict,window_select,level_id):
+        season_title = df['game_date'].str[0:4].values[0]
+        sns.set_theme(style="whitegrid", palette="pastel")
+        if player_id == "":
+            fig = plt.figure(figsize=(12, 12))
+            fig.text(s='Please Select a Pitcher',x=0.5,y=0.5)
+            return
+        
+        
+
+        swing_min = int(window_select)
+        fig, ax = plt.subplots(1, 1, figsize=(10, 10))
+
+        fig.set_facecolor('white')
+        #ax.set_facecolor('white')
+        #fig.patch.set_facecolor('lightblue')
+
+        print(stat_id)
+
+        if stat_id in pa_list:
+            print('we hAVE MADE IT TO THIS PART OF THE CODE')
+
+
+        if stat_id in pa_list:
+            elly_zone_df = df[(df.pa==1)&(df.batter_id == int(player_id))&(df.level==level_id)]
+            divisor_x = 'pa'
+            print('this is short')
+            print(elly_zone_df)
+
+
+        if stat_id in balls_in_play_list:
+            elly_zone_df = df[(df.bip)&(df.batter_id == int(player_id))&(df.level==level_id)]
+            divisor_x = 'bip'
+            #print('this is short')
+
+        if stat_id in balls_in_play_list:
+            elly_zone_df = df[(df.bip)&(df.batter_id == int(player_id))&(df.level==level_id)]
+            divisor_x = 'bip'
+            print('this is short')
+
+        if stat_id in pitches_list:
+            elly_zone_df = df[(df.pitches == 1)&(df.batter_id == int(player_id))&(df.level==level_id)]
+            divisor_x = 'pitches'
+
+        if stat_id in swings_list:
+            elly_zone_df = df[(df.swings == 1)&(df.batter_id == int(player_id))&(df.level==level_id)]
+            divisor_x = 'swings'
+
+
+        if stat_id in in_zone_pitches_list:
+            elly_zone_df = df[(df.in_zone)&(df.batter_id == int(player_id))&(df.level==level_id)]
+            divisor_x = 'in_zone'
+
+
+        if stat_id in in_zone_swings_list:
+            elly_zone_df = df[(df.zone_swing)&(df.batter_id == int(player_id))&(df.level==level_id)]
+            divisor_x = 'zone_swing'
+
+
+        if stat_id in out_zone_pitches_list:
+            elly_zone_df = df[(df.in_zone == False)&(df.batter_id == int(player_id))&(df.level==level_id)]
+            divisor_x = 'out_zone'
+
+
+        if stat_id in out_zone_swings_list:
+            elly_zone_df = df[(df.ozone_swing)&(df.batter_id == int(player_id))&(df.level==level_id)]
+            divisor_x = 'ozone_swing'
+
+        if stat_id in woba_list:
+            elly_zone_df = df[(df.woba_codes==1)&(df.batter_id == int(player_id))&(df.level==level_id)]
+            divisor_x = 'woba_codes'            
+            
+        # penguins = sns.load_dataset("penguins")
+        # sns.histplot(data=penguins, x="flipper_length_mm")
+        # print('we made it here:')
+        # print(int(player_id))
+        # print(stat_id)
+        # print(level_id)
+        # print(df[(df.batter_id == int(player_id))&(df.level==level_id)])
+        # print(df.columns)
+        # print(elly_zone_df[plot_dict[stat_id]["x_value"]].sum())
+
+        df_summ_new = df_summ.copy()
+        df_summ_new = df_summ_new.set_index('batter_id','batter_name','level')
+        df_summ_new = df_summ_new[df_summ_new[divisor_x] >= int(window_select)]
+        df_summ_new = df_summ_new[df_summ_new.level==level_id]
+
+        df_summ_rank = df_summ_new.rank(method='max',ascending=False)
+        df_summ_rank.columns = df_summ_rank.columns+['_rank']
+
+        df_summ_rank_percent =  df_summ_new.rank(pct=True)
+        df_summ_rank_percent.columns = df_summ_rank_percent.columns+['_percent']
+        
+
+        df_summ_new = df_summ_new.reset_index()
+        df_summ_rank = df_summ_rank.reset_index()
+        df_summ_rank_percent = df_summ_rank_percent.reset_index()
+        print('Table columns:')
+        
+        df_summ_new.batter_id = df_summ_new.batter_id.astype(int)
+        df_summ_rank.batter_id = df_summ_rank.batter_id.astype(int)
+        df_summ_rank_percent.batter_id = df_summ_rank_percent.batter_id.astype(int)
+
+        print('Table columns2:')
+        df_summ_new = df_summ_new.merge(df_summ_rank,left_on=['batter_id'],right_on=['batter_id'],how='left',suffixes=['','_rank'])
+
+        df_summ_new = df_summ_new.merge(df_summ_rank_percent,left_on=['batter_id'],right_on=['batter_id'],how='left',suffixes=['','_percent'])
+
+
+        print(df_summ_new)
+        print(df_summ_rank)
+        print(df_summ_rank_percent)
+        
+
+        
+
+        #sns.scatterplot(x=data_df.launch_speed_90,y=data_df.zone_contact,color=colour_palette[0],s=75,label=int(player_id))
+
+        df_summ_new_select = df_summ_new[df_summ_new.batter_id == int(player_id)].reset_index(drop=True)
+        print('whiffing')
+        print(df)
+        print('Player _df:')
+        print(df_summ_new_select)
+
+        if len(df_summ_new_select) < 1:
+            ax.text(x=0.5,y=0.5,s='Please Select Different Parameters to Produce a plot',fontsize=18,ha='center')
+            return 
+
+        p = inflect.engine()
+
+        df_summ_new_select = df_summ_new_select.loc[:,~df_summ_new_select.columns.duplicated(keep='last')].copy()
+        print('Table for the player:')
+        print(list(df_summ_new_select.columns))
+        print(plot_dict[stat_id]["percentile_label"])
+        print(plot_dict[stat_id]["percentile_label"]+'_percent')
+        print(df_summ_new_select)
+        print(1*plot_dict[stat_id]["flip_p"])
+        print(round(df_summ_new_select[plot_dict[stat_id]["percentile_label"]+"_percent"][0],2))
+        print((1*plot_dict[stat_id]["flip_p"]-round(df_summ_new_select[plot_dict[stat_id]["percentile_label"]+"_percent"][0],2))*100)
+
+        # print(df_summ_new_select[plot_dict[stat_id]["percentile_label"]+'_percent'])
+
+        if plot_dict[stat_id]['percent']:
+            label_1=f'{level_id} Average {df[df.level == level_id][plot_dict[stat_id]["x_value"]].sum()/df[df.level == level_id][divisor_x].sum():.1%}'
+            label_2=f'{batter_dict[int(player_id)]} Average {elly_zone_df[plot_dict[stat_id]["x_value"]].sum()/elly_zone_df[divisor_x].sum():.1%} ({p.ordinal(abs(int((1*plot_dict[stat_id]["flip_p"]-round(df_summ_new_select[plot_dict[stat_id]["percentile_label"]+"_percent"][0],2))*100)))} Percentile)'
+            #label_2=f'{batter_dict[int(player_id)]} Average {elly_zone_df[plot_dict[stat_id]["x_value"]].sum()/elly_zone_df[divisor_x].sum():.1%}'
+            ax.yaxis.set_major_formatter(mtick.PercentFormatter(1))
+
+        else:
+            label_1=f'{level_id} Average {df[df.level == level_id][plot_dict[stat_id]["x_value"]].sum()/df[df.level == level_id][divisor_x].sum():.1f}'
+            label_2=f'{batter_dict[int(player_id)]} Average {elly_zone_df[plot_dict[stat_id]["x_value"]].sum()/elly_zone_df[divisor_x].sum():.1f} ({p.ordinal(abs(int((1*plot_dict[stat_id]["flip_p"]-round(df_summ_new_select[plot_dict[stat_id]["percentile_label"]+"_percent"][0],2))*100)))} Percentile)'
+            #label_2=f'{batter_dict[int(player_id)]} Average {elly_zone_df[plot_dict[stat_id]["x_value"]].sum()/elly_zone_df[divisor_x].sum():.1f}'
+            #ax.yaxis.set_major_formatter(mtick.int)
+        
+
+        if plot_dict[stat_id]['percentile']:
+            label_1=f'{level_id} Average {df[df.level == level_id][plot_dict[stat_id]["x_value"]].quantile(0.9):.1f}'
+            label_2=f'{batter_dict[int(player_id)]} Average {elly_zone_df[plot_dict[stat_id]["x_value"]].quantile(0.9):.1f} ({p.ordinal(abs(int((1*plot_dict[stat_id]["flip_p"]-round(df_summ_new_select[plot_dict[stat_id]["percentile_label"]+"_percent"][0],2))*100)))} Percentile)'
+            #label_2=f'{batter_dict[int(player_id)]} Average {elly_zone_df[plot_dict[stat_id]["x_value"]].sum()/elly_zone_df[divisor_x].sum():.1%}'
+            #ax.yaxis.set_major_formatter(mtick.int)
+
+        if plot_dict[stat_id]['avg_adjust']:
+            label_1=f'{level_id} Average {df[df.level == level_id][plot_dict[stat_id]["x_value"]].sum()/df[df.level == level_id][divisor_x].sum():.3f}'
+            label_2=f'{batter_dict[int(player_id)]} Average {elly_zone_df[plot_dict[stat_id]["x_value"]].sum()/elly_zone_df[divisor_x].sum():.3f} ({p.ordinal(abs(int((1*plot_dict[stat_id]["flip_p"]-round(df_summ_new_select[plot_dict[stat_id]["percentile_label"]+"_percent"][0],2))*100)))} Percentile)'
+            #label_2=f'{batter_dict[int(player_id)]} Average {elly_zone_df[plot_dict[stat_id]["x_value"]].sum()/elly_zone_df[divisor_x].sum():.1%}'
+            #ax.yaxis.set_major_formatter(mtick.int)
+
+        print(plot_dict[stat_id]["x_value"])
+        print(divisor_x)
+
+        # df_summ_new = df_summ.copy()
+        # df_summ_new = df_summ_new[df_summ_new.balls_in_play >= int(window_select)]
+        # df_summ_new = df_summ_new[df_summ_new.level==level_id]
+
+        
+        print('this is here:')
+        print(df_summ_new.head())
+        print(df_summ_new.columns)
+
+
+        if plot_dict[stat_id]["flip_p"] == False:
+            ax.hlines(y=(df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.9),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[2],linestyle='dotted',alpha=0.5)
+            ax.hlines(y=(df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.75),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[3],linestyle='dotted',alpha=0.5)
+            ax.hlines(y=(df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.25),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[4],linestyle='dotted',alpha=0.5)
+            ax.hlines(y=(df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.1),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[5],linestyle='dotted',alpha=0.5)
+
+
+            hard_hit_dates = [(df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.9),
+                              (df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.75),
+                              (df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.25),
+                              (df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.1)]
+            hard_hit_text = ['90th %','75th %','25th %','10th %']
+            for i, x in enumerate(hard_hit_dates):
+                text(min(window_select+window_select/100,+window_select+1), x ,hard_hit_text[i], rotation=0, ha='left',
+                    bbox=dict(facecolor='white',alpha=0.5, edgecolor=colour_palette[2+i], pad=2))
+
+
+
+        if plot_dict[stat_id]["flip_p"] == True:
+            ax.hlines(y=(df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.1),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[2],linestyle='dotted',alpha=0.5)
+            ax.hlines(y=(df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.25),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[3],linestyle='dotted',alpha=0.5)
+            ax.hlines(y=(df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.75),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[4],linestyle='dotted',alpha=0.5)
+            ax.hlines(y=(df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.9),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[5],linestyle='dotted',alpha=0.5)
+
+            hard_hit_dates = [(df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.9),
+                              (df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.75),
+                              (df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.25),
+                              (df_summ_new[plot_dict[stat_id]["percentile_label"]]).quantile(0.1)]
+            hard_hit_text = ['10th %','25th %','75th %','90th %']
+            for i, x in enumerate(hard_hit_dates):
+                text(min(window_select+window_select/100,window_select+window_select+3), x ,hard_hit_text[i], rotation=0, ha='left',
+                    bbox=dict(facecolor='white',alpha=0.5, edgecolor=colour_palette[2+i], pad=2))
+
+
+        
+
+
+
+        if plot_dict[stat_id]["percentile"] == False:
+            ax.hlines(y=df[df.level == level_id][plot_dict[stat_id]["x_value"]].sum()/df[df.level == level_id][divisor_x].sum(),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[1],linestyle='-.',label=label_1)
+
+            ax.hlines(y=elly_zone_df[plot_dict[stat_id]["x_value"]].sum()/elly_zone_df[divisor_x].sum(),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[0],linestyle='--',label=label_2)
+
+            sns.lineplot(x=range(1,len(elly_zone_df)+1),y=elly_zone_df[plot_dict[stat_id]["x_value"]].fillna(0).rolling(window=swing_min).sum()/swing_min,color=colour_palette[0],linewidth=3,ax=ax)
+
+
+
+        if plot_dict[stat_id]["percentile"] == True:
+
+            ax.hlines(y=df[df.level == level_id][plot_dict[stat_id]["x_value"]].quantile(0.9),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[1],linestyle='-.',label=label_1)
+
+            ax.hlines(y=elly_zone_df[plot_dict[stat_id]["x_value"]].fillna(0).quantile(0.9),xmin=swing_min,xmax=len(elly_zone_df),color=colour_palette[0],linestyle='--',label=label_2)
+
+            sns.lineplot(x=range(1,len(elly_zone_df)+1),y=elly_zone_df[plot_dict[stat_id]["x_value"]].fillna(0).rolling(window=swing_min).quantile(0.9),color=colour_palette[0],linewidth=3,ax=ax)
+
+
+        #ax.set_xlim(window_select,exit_velo_df_small.pitch.max())
+        #plt.yticks([0,0.2,0.4,0.6,0.8,1])
+        #ax.set_ylim(math.floor((min(df_summ.zone_contact)/5)*100)*5/100,1)
+        ax.set_xlim(math.floor(swing_min),len(elly_zone_df))
+        ax.set_title(f'{batter_dict[int(player_id)]} - {season_title} - {level_id} - {swing_min} {plot_dict[stat_id]["x_axis"]} Rolling {plot_dict[stat_id]["title"]}', fontsize=16,fontname='Century Gothic',)
+        #vals = ax.get_yticks()
+        ax.set_xlabel(plot_dict[stat_id]['x_axis'], fontsize=16,fontname='Century Gothic')
+        ax.set_ylabel(plot_dict[stat_id]['y_axis'], fontsize=16,fontname='Century Gothic')
+
+        #fig.axes[0].invert_yaxis()
+        
+        #fig.subplots_adjust(wspace=.02, hspace=.02)
+        #ax.xaxis.set_major_formatter(FuncFormatter(lambda x, _: int(x)))
+        ax.set_yticks(plot_dict[stat_id]["x_range"])
+        #fig.colorbar(plot_dist, ax=ax)
+        #fig.colorbar(plot_dist)
+        #fig.axes[0].invert_yaxis()
+        ax.legend(fontsize='16')
+        fig.text(x=0.03,y=0.02,s='By: @TJStats',fontname='Century Gothic')
+        fig.text(x=1-0.03,y=0.02,s='Data: MLB',ha='right',fontname='Century Gothic')
+        fig.tight_layout()
+        return