Upload 13 files

stuff_model/calculate_arm_angles.py

@@ -1,112 +1,112 @@
-import polars as pl
-import numpy as np
-import requests
-from io import StringIO
-
-def calculate_arm_angles(df: pl.DataFrame, pitcher_id: int) -> pl.DataFrame:
-    def fetch_arm_angle_data(url: str):
-        headers = {
-            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3'
-        }
-        try:
-            response = requests.get(url, headers=headers, timeout=5)
-            if not response.ok or "<html" in response.text.lower():
-                return None
-            return pl.read_csv(StringIO(response.text), truncate_ragged_lines=True)
-        except requests.RequestException:
-            return None
-
-    date_start = df['game_date'][0]
-    date_end = df['game_date'][-1]
-    season = int(date_start[:4])
-    daily_check = date_start == date_end
-
-    # Try fetching current-season arm angle data
-    url = (
-        f"https://baseballsavant.mlb.com/leaderboard/pitcher-arm-angles"
-        f"?batSide=&dateStart={date_start}&dateEnd={date_end}&gameType=R&groupBy=&min=1"
-        f"&minGroupPitches=1&perspective=back&pitchHand=&pitchType=&season={season}"
-        f"&size=small&sort=ascending&team=&csv=true"
-    )
-    df_arm_angle = fetch_arm_angle_data(url)
-    print("ARM ANGLE",df_arm_angle)
-
-    old_data = False
-    if df_arm_angle is None or pitcher_id not in df_arm_angle["pitcher"]:
-        old_data = True
-
-        # Fallback to saved CSVs if 2025 data isn't fetched or pitcher not found
-        try:
-            df_arm_angle_2025 = pl.read_csv("stuff_model/pitcher_arm_angles_2025.csv", truncate_ragged_lines=True)
-        except Exception as e:
-            raise RuntimeError("Failed to load fallback 2025 arm angle CSV.") from e
-
-        try:
-            df_arm_angle_2024 = pl.read_csv("stuff_model/pitcher_arm_angles_2024.csv", truncate_ragged_lines=True)
-            df_arm_angle_2024 = df_arm_angle_2024.cast(df_arm_angle_2025.schema)
-        except Exception as e:
-            raise RuntimeError("Failed to load or cast 2024 arm angle CSV.") from e
-
-        df_arm_angle = pl.concat([df_arm_angle_2025, df_arm_angle_2024]).unique(subset=["pitcher"], keep="first")
-
-    # Filter your tracking data
-    df_filter = df.filter(pl.col("pitcher_id") == pitcher_id).drop_nulls(subset=["release_pos_x", "release_pos_z"])
-
-    if pitcher_id not in df_arm_angle["pitcher"]:
-        data = requests.get(f'https://statsapi.mlb.com/api/v1/people?personIds={pitcher_id}').json()
-        height_in = data['people'][0]['height']
-        height = int(height_in.split("'")[0]) * 12 + int(height_in.split("'")[1].split('"')[0])
-        
-        df_filter = (
-            df_filter.with_columns(
-                (pl.col("release_pos_x") * 12).alias("release_pos_x"),
-                (pl.col("release_pos_z") * 12).alias("release_pos_z"),
-                (pl.lit(height * 0.70)).alias("shoulder_pos"),
-            )
-            .with_columns(
-                (pl.col("release_pos_z") - pl.col("shoulder_pos")).alias("Opp"),
-                pl.col("release_pos_x").abs().alias("Adj"),
-            )
-            .with_columns(
-                pl.struct(["Opp", "Adj"]).map_elements(lambda x: np.arctan2(x["Opp"], x["Adj"])).alias("arm_angle_rad")
-            )
-            .with_columns(
-                pl.col("arm_angle_rad").degrees().alias("arm_angle")
-            )
-        )
-
-    else:
-        row = df_arm_angle.filter(pl.col("pitcher") == pitcher_id).select([
-            "relative_shoulder_x", "shoulder_z", "relative_release_ball_x", "release_ball_z", "ball_angle"
-        ]).row(0)
-        shoulder_x, shoulder_z, rel_x, rel_z, ball_angle = row
-        hyp = np.sqrt((rel_x - shoulder_x)**2 + (rel_z - shoulder_z)**2)
-
-        df_filter = (
-            df_filter.with_columns(
-                (pl.col("release_pos_z") - shoulder_z).alias("Opp"),
-                pl.lit(hyp).alias("Hyp"),
-            )
-            .with_columns(
-                pl.struct(["Opp", "Hyp"]).map_elements(lambda x: np.arcsin(x["Opp"] / x["Hyp"])).alias("arm_angle_rad")
-            )
-            .with_columns(
-                pl.col("arm_angle_rad").degrees().alias("arm_angle")
-            )
-        )
-
-        # Adjust based on data source freshness
-        if old_data:
-            df_filter = df_filter.with_columns(((pl.col("arm_angle") * 0.5) + (ball_angle * 0.5)).alias("arm_angle"))
-        elif daily_check:
-            df_filter = df_filter.with_columns(((pl.col("arm_angle") * 0.25) + (ball_angle * 0.75)).alias("arm_angle"))
-        else:
-            df_filter = df_filter.with_columns(((pl.col("arm_angle") * 0.0) + (ball_angle * 1)).alias("arm_angle"))
-
-        # Fill missing arm_angle values with mean
-        valid_mean = df_filter["arm_angle"].fill_nan(None).drop_nulls().mean()
-        df_filter = df_filter.with_columns(
-            df_filter["arm_angle"].fill_nan(None).fill_null(valid_mean)
-        )
-
-    return df_filter
+import polars as pl
+import numpy as np
+import requests
+from io import StringIO
+
+def calculate_arm_angles(df: pl.DataFrame, pitcher_id: int) -> pl.DataFrame:
+    def fetch_arm_angle_data(url: str):
+        headers = {
+            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3'
+        }
+        try:
+            response = requests.get(url, headers=headers, timeout=5)
+            if not response.ok or "<html" in response.text.lower():
+                return None
+            return pl.read_csv(StringIO(response.text), truncate_ragged_lines=True)
+        except requests.RequestException:
+            return None
+
+    date_start = df['game_date'][0]
+    date_end = df['game_date'][-1]
+    season = int(date_start[:4])
+    daily_check = date_start == date_end
+
+    # Try fetching current-season arm angle data
+    url = (
+        f"https://baseballsavant.mlb.com/leaderboard/pitcher-arm-angles"
+        f"?batSide=&dateStart={date_start}&dateEnd={date_end}&gameType=R&groupBy=&min=1"
+        f"&minGroupPitches=1&perspective=back&pitchHand=&pitchType=&season={season}"
+        f"&size=small&sort=ascending&team=&csv=true"
+    )
+    df_arm_angle = fetch_arm_angle_data(url)
+    print("ARM ANGLE",df_arm_angle)
+
+    old_data = False
+    if df_arm_angle is None or pitcher_id not in df_arm_angle["pitcher"]:
+        old_data = True
+
+        # Fallback to saved CSVs if 2025 data isn't fetched or pitcher not found
+        try:
+            df_arm_angle_2025 = pl.read_csv("stuff_model/pitcher_arm_angles_2025.csv", truncate_ragged_lines=True)
+        except Exception as e:
+            raise RuntimeError("Failed to load fallback 2025 arm angle CSV.") from e
+
+        try:
+            df_arm_angle_2024 = pl.read_csv("stuff_model/pitcher_arm_angles_2024.csv", truncate_ragged_lines=True)
+            df_arm_angle_2024 = df_arm_angle_2024.cast(df_arm_angle_2025.schema)
+        except Exception as e:
+            raise RuntimeError("Failed to load or cast 2024 arm angle CSV.") from e
+
+        df_arm_angle = pl.concat([df_arm_angle_2025, df_arm_angle_2024]).unique(subset=["pitcher"], keep="first")
+
+    # Filter your tracking data
+    df_filter = df.filter(pl.col("pitcher_id") == pitcher_id).drop_nulls(subset=["release_pos_x", "release_pos_z"])
+
+    if pitcher_id not in df_arm_angle["pitcher"]:
+        data = requests.get(f'https://statsapi.mlb.com/api/v1/people?personIds={pitcher_id}').json()
+        height_in = data['people'][0]['height']
+        height = int(height_in.split("'")[0]) * 12 + int(height_in.split("'")[1].split('"')[0])
+        
+        df_filter = (
+            df_filter.with_columns(
+                (pl.col("release_pos_x") * 12).alias("release_pos_x"),
+                (pl.col("release_pos_z") * 12).alias("release_pos_z"),
+                (pl.lit(height * 0.70)).alias("shoulder_pos"),
+            )
+            .with_columns(
+                (pl.col("release_pos_z") - pl.col("shoulder_pos")).alias("Opp"),
+                pl.col("release_pos_x").abs().alias("Adj"),
+            )
+            .with_columns(
+                pl.struct(["Opp", "Adj"]).map_elements(lambda x: np.arctan2(x["Opp"], x["Adj"])).alias("arm_angle_rad")
+            )
+            .with_columns(
+                pl.col("arm_angle_rad").degrees().alias("arm_angle")
+            )
+        )
+
+    else:
+        row = df_arm_angle.filter(pl.col("pitcher") == pitcher_id).select([
+            "relative_shoulder_x", "shoulder_z", "relative_release_ball_x", "release_ball_z", "ball_angle"
+        ]).row(0)
+        shoulder_x, shoulder_z, rel_x, rel_z, ball_angle = row
+        hyp = np.sqrt((rel_x - shoulder_x)**2 + (rel_z - shoulder_z)**2)
+
+        df_filter = (
+            df_filter.with_columns(
+                (pl.col("release_pos_z") - shoulder_z).alias("Opp"),
+                pl.lit(hyp).alias("Hyp"),
+            )
+            .with_columns(
+                pl.struct(["Opp", "Hyp"]).map_elements(lambda x: np.arcsin(x["Opp"] / x["Hyp"])).alias("arm_angle_rad")
+            )
+            .with_columns(
+                pl.col("arm_angle_rad").degrees().alias("arm_angle")
+            )
+        )
+
+        # Adjust based on data source freshness
+        if old_data:
+            df_filter = df_filter.with_columns(((pl.col("arm_angle") * 0.5) + (ball_angle * 0.5)).alias("arm_angle"))
+        elif daily_check:
+            df_filter = df_filter.with_columns(((pl.col("arm_angle") * 0.25) + (ball_angle * 0.75)).alias("arm_angle"))
+        else:
+            df_filter = df_filter.with_columns(((pl.col("arm_angle") * 0.0) + (ball_angle * 1)).alias("arm_angle"))
+
+        # Fill missing arm_angle values with mean
+        valid_mean = df_filter["arm_angle"].fill_nan(None).drop_nulls().mean()
+        df_filter = df_filter.with_columns(
+            df_filter["arm_angle"].fill_nan(None).fill_null(valid_mean)
+        )
+
+    return df_filter

stuff_model/df_update.py

@@ -1,650 +1,650 @@
-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]).alias('pz_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'),
-
-    
-        ])
-        
-        df = df.with_columns([
-        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'),
-        
-        ])
-
-        df = df.with_columns([
-        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_final'),
-        ])
-
-        df = df.with_columns([
-            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_final') == False).then(True).otherwise(False).alias('out_zone'),
-            pl.when((pl.col('in_zone_final') == True) & (pl.col('swings') == 1)).then(True).otherwise(False).alias('zone_swing'),
-            pl.when((pl.col('in_zone_final') == True) & (pl.col('swings') == 1) & (pl.col('whiffs') == 0)).then(True).otherwise(False).alias('zone_contact'),
-            pl.when((pl.col('in_zone_final') == False) & (pl.col('swings') == 1)).then(True).otherwise(False).alias('ozone_swing'),
-            pl.when((pl.col('in_zone_final') == 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_final'),
-            
-
-        ])
-
-        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_final'] == 0).alias('heart'),
-            (df['attack_zone_final'] == 1).alias('shadow'),
-            (df['attack_zone_final'] == 2).alias('chase'),
-            (df['attack_zone_final'] == 3).alias('waste'),
-            ((df['attack_zone_final'] == 0) & (df['swings'] == 1)).alias('heart_swing'),
-            ((df['attack_zone_final'] == 1) & (df['swings'] == 1)).alias('shadow_swing'),
-            ((df['attack_zone_final'] == 2) & (df['swings'] == 1)).alias('chase_swing'),
-            ((df['attack_zone_final'] == 3) & (df['swings'] == 1)).alias('waste_swing'),
-            ((df['attack_zone_final'] == 0) & (df['whiffs'] == 1)).alias('heart_whiff'),
-            ((df['attack_zone_final'] == 1) & (df['whiffs'] == 1)).alias('shadow_whiff'),
-            ((df['attack_zone_final'] == 2) & (df['whiffs'] == 1)).alias('chase_whiff'),
-            ((df['attack_zone_final'] == 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')
-
-
-        pitch_cat = {'FA': None,
-                            '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(
-            df["pitch_type"]
-            .replace(pitch_cat)
-            .fill_null("Unknown")
-            .alias("pitch_group")
-        )
-
-        df = df.with_columns([
-            
-            (-(pl.col('vy0')**2 - (2 * pl.col('ay') * (pl.col('y0') - 17/12)))**0.5).alias('vy_f'),
-            ])
-
-        df = df.with_columns([
-            ((pl.col('vy_f') - pl.col('vy0')) / pl.col('ay')).alias('t'),
-            ])
-
-        df = df.with_columns([
-            (pl.col('vz0') + (pl.col('az') * pl.col('t'))).alias('vz_f'),
-            (pl.col('vx0') + (pl.col('ax') * pl.col('t'))).alias('vx_f')
-            ])
-
-        df = df.with_columns([
-                (-np.arctan(pl.col('vz_f') / pl.col('vy_f')) * (180 / np.pi)).alias('vaa'),
-                (-np.arctan(pl.col('vx_f') / pl.col('vy_f')) * (180 / np.pi)).alias('haa')
-            ])
-
-        # Mirror horizontal break for left-handed pitchers
-        df = df.with_columns(
-            pl.when(pl.col('pitcher_hand') == 'L')
-            .then(-pl.col('ax'))
-            .otherwise(pl.col('ax'))
-            .alias('ax')
-        )
-
-        # Mirror horizontal break for left-handed pitchers
-        df = df.with_columns(
-            pl.when(pl.col('pitcher_hand') == 'L')
-            .then(-pl.col('hb'))
-            .otherwise(pl.col('hb'))
-            .alias('hb')
-        )
-
-        # Mirror horizontal release point for left-handed pitchers
-        df = df.with_columns(
-            pl.when(pl.col('pitcher_hand') == 'L')
-            .then(pl.col('x0'))
-            .otherwise(-pl.col('x0'))
-            .alias('x0')
-        )
-
-        df = df.with_columns([
-            pl.when(df['swings'].is_null()).then(None).otherwise(df['swings']).alias('is_swing'),
-            pl.when(df['bip'].is_null()).then(None).otherwise(df['bip']).alias('is_bip')])
-
-
-
-
-        df = df.with_columns([
-            (np.arctan((pl.col("hit_x")*-1 + 125.42) / (198.27 - pl.col("hit_y"))) * 180 / np.pi * 0.75).alias("spray_angle")
-        ])
-
-        df = df.with_columns([
-            pl.when(pl.col("batter_hand") == "L")
-                .then(-pl.col("spray_angle"))
-                .otherwise(pl.col("spray_angle"))
-                .alias("adj_spray_angle")
-        ]).drop("spray_angle")
-
-
-        df = df.with_columns([
-            pl.when(pl.col("adj_spray_angle").is_not_null() & (pl.col("adj_spray_angle") < -15))
-            .then(pl.lit("oppo"))
-            .when(pl.col("adj_spray_angle").is_not_null() & (pl.col("adj_spray_angle") > 15))
-            .then(pl.lit("pull"))
-            .when(pl.col("adj_spray_angle").is_not_null())
-            .then(pl.lit("straight"))
-            .otherwise(None)  # Keep null if adj_spray_angle is null
-            .alias("hit_direction")
-        ])
-
-        df = df.with_columns([
-        pl.when(pl.col("hit_direction") == "oppo").then(1).otherwise(None).alias("oppo"),
-        pl.when(pl.col("hit_direction") == "pull").then(1).otherwise(None).alias("pull"),
-        pl.when(pl.col("hit_direction") == "straight").then(1).otherwise(None).alias("straight")
-        ])
-
-        df = df.with_columns([
-            pl.when(pl.col("event_type") == "single").then(1).otherwise(0).alias("single"),
-            pl.when(pl.col("event_type") == "double").then(1).otherwise(0).alias("double"),
-            pl.when(pl.col("event_type") == "triple").then(1).otherwise(0).alias("triple"),
-            pl.when(pl.col("event_type") == "home_run").then(1).otherwise(0).alias("home_run")
-        ])
-
-
-
-
-
-        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('k_minus_bb').sum().alias('k_minus_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'),
-            pl.col('pull').sum().alias('pull'),
-            pl.col('straight').sum().alias('straight'),
-            pl.col('oppo').sum().alias('oppo'),
-            ((pl.col('trajectory_fly_ball') == 1) | (pl.col('trajectory_line_drive') == 1)).sum().alias('fly_line_bip'),
-            (pl.col('pull') & ((pl.col('trajectory_fly_ball') == 1) | (pl.col('trajectory_line_drive') == 1))).sum().alias('pull_fly_ball'),
-            pl.col('single').sum().alias('single'),
-            pl.col('double').sum().alias('double'),
-            pl.col('triple').sum().alias('triple'),
-            pl.col('home_run').sum().alias('home_run'),
-            (pl.col('extension').mean()).alias('extension'),
-            (pl.col('start_speed').filter(pl.col('pitch_type').is_in(['FF','SI'])).mean().alias('avg_start_speed_ff')),
-
-
-
-        ])
-
-        # 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('k_minus_bb') / pl.col('pa')).alias('k_minus_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('out_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'),
-            (pl.col('pull') / pl.col('bip')).alias('pull_percent'),
-            (pl.col('straight') / pl.col('bip')).alias('straight_percent'),
-            (pl.col('oppo') / pl.col('bip')).alias('oppo_percent'),
-            (pl.col('pull_fly_ball') / pl.col('fly_line_bip')).alias('pulled_fly_ball_percent'),
-
-        ])
-
-        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('k_minus_bb').sum().alias('k_minus_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'),
-            pl.col('pull').sum().alias('pull'),
-            pl.col('straight').sum().alias('straight'),
-            pl.col('oppo').sum().alias('oppo'),
-            ((pl.col('trajectory_fly_ball') == 1) | (pl.col('trajectory_line_drive') == 1)).sum().alias('fly_line_bip'),
-            (pl.col('pull') & ((pl.col('trajectory_fly_ball') == 1) | (pl.col('trajectory_line_drive') == 1))).sum().alias('pull_fly_ball'),
-            pl.col('single').sum().alias('single'),
-            pl.col('double').sum().alias('double'),
-            pl.col('triple').sum().alias('triple'),
-            pl.col('home_run').sum().alias('home_run'),
-            (pl.col('extension').mean()).alias('extension'),
-            (pl.col('start_speed').filter(pl.col('pitch_type').is_in(['FF','SI','FC'])).mean().alias('avg_start_speed_ff')),
-
-            
-        ])
-
-        # 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('k_minus_bb') / pl.col('pa')).alias('k_minus_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'),
-            (pl.col('pull') / pl.col('bip')).alias('pull_percent'),
-            (pl.col('straight') / pl.col('bip')).alias('straight_percent'),
-            (pl.col('oppo') / pl.col('bip')).alias('oppo_percent'),
-            (pl.col('pull_fly_ball') / pl.col('fly_line_bip')).alias('pulled_fly_ball_percent'),
-        ])
-
+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]).alias('pz_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'),
+
+    
+        ])
+        
+        df = df.with_columns([
+        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'),
+        
+        ])
+
+        df = df.with_columns([
+        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_final'),
+        ])
+
+        df = df.with_columns([
+            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_final') == False).then(True).otherwise(False).alias('out_zone'),
+            pl.when((pl.col('in_zone_final') == True) & (pl.col('swings') == 1)).then(True).otherwise(False).alias('zone_swing'),
+            pl.when((pl.col('in_zone_final') == True) & (pl.col('swings') == 1) & (pl.col('whiffs') == 0)).then(True).otherwise(False).alias('zone_contact'),
+            pl.when((pl.col('in_zone_final') == False) & (pl.col('swings') == 1)).then(True).otherwise(False).alias('ozone_swing'),
+            pl.when((pl.col('in_zone_final') == 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_final'),
+            
+
+        ])
+
+        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_final'] == 0).alias('heart'),
+            (df['attack_zone_final'] == 1).alias('shadow'),
+            (df['attack_zone_final'] == 2).alias('chase'),
+            (df['attack_zone_final'] == 3).alias('waste'),
+            ((df['attack_zone_final'] == 0) & (df['swings'] == 1)).alias('heart_swing'),
+            ((df['attack_zone_final'] == 1) & (df['swings'] == 1)).alias('shadow_swing'),
+            ((df['attack_zone_final'] == 2) & (df['swings'] == 1)).alias('chase_swing'),
+            ((df['attack_zone_final'] == 3) & (df['swings'] == 1)).alias('waste_swing'),
+            ((df['attack_zone_final'] == 0) & (df['whiffs'] == 1)).alias('heart_whiff'),
+            ((df['attack_zone_final'] == 1) & (df['whiffs'] == 1)).alias('shadow_whiff'),
+            ((df['attack_zone_final'] == 2) & (df['whiffs'] == 1)).alias('chase_whiff'),
+            ((df['attack_zone_final'] == 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')
+
+
+        pitch_cat = {'FA': None,
+                            '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(
+            df["pitch_type"]
+            .replace(pitch_cat)
+            .fill_null("Unknown")
+            .alias("pitch_group")
+        )
+
+        df = df.with_columns([
+            
+            (-(pl.col('vy0')**2 - (2 * pl.col('ay') * (pl.col('y0') - 17/12)))**0.5).alias('vy_f'),
+            ])
+
+        df = df.with_columns([
+            ((pl.col('vy_f') - pl.col('vy0')) / pl.col('ay')).alias('t'),
+            ])
+
+        df = df.with_columns([
+            (pl.col('vz0') + (pl.col('az') * pl.col('t'))).alias('vz_f'),
+            (pl.col('vx0') + (pl.col('ax') * pl.col('t'))).alias('vx_f')
+            ])
+
+        df = df.with_columns([
+                (-np.arctan(pl.col('vz_f') / pl.col('vy_f')) * (180 / np.pi)).alias('vaa'),
+                (-np.arctan(pl.col('vx_f') / pl.col('vy_f')) * (180 / np.pi)).alias('haa')
+            ])
+
+        # Mirror horizontal break for left-handed pitchers
+        df = df.with_columns(
+            pl.when(pl.col('pitcher_hand') == 'L')
+            .then(-pl.col('ax'))
+            .otherwise(pl.col('ax'))
+            .alias('ax')
+        )
+
+        # Mirror horizontal break for left-handed pitchers
+        df = df.with_columns(
+            pl.when(pl.col('pitcher_hand') == 'L')
+            .then(-pl.col('hb'))
+            .otherwise(pl.col('hb'))
+            .alias('hb')
+        )
+
+        # Mirror horizontal release point for left-handed pitchers
+        df = df.with_columns(
+            pl.when(pl.col('pitcher_hand') == 'L')
+            .then(pl.col('x0'))
+            .otherwise(-pl.col('x0'))
+            .alias('x0')
+        )
+
+        df = df.with_columns([
+            pl.when(df['swings'].is_null()).then(None).otherwise(df['swings']).alias('is_swing'),
+            pl.when(df['bip'].is_null()).then(None).otherwise(df['bip']).alias('is_bip')])
+
+
+
+
+        df = df.with_columns([
+            (np.arctan((pl.col("hit_x")*-1 + 125.42) / (198.27 - pl.col("hit_y"))) * 180 / np.pi * 0.75).alias("spray_angle")
+        ])
+
+        df = df.with_columns([
+            pl.when(pl.col("batter_hand") == "L")
+                .then(-pl.col("spray_angle"))
+                .otherwise(pl.col("spray_angle"))
+                .alias("adj_spray_angle")
+        ]).drop("spray_angle")
+
+
+        df = df.with_columns([
+            pl.when(pl.col("adj_spray_angle").is_not_null() & (pl.col("adj_spray_angle") < -15))
+            .then(pl.lit("oppo"))
+            .when(pl.col("adj_spray_angle").is_not_null() & (pl.col("adj_spray_angle") > 15))
+            .then(pl.lit("pull"))
+            .when(pl.col("adj_spray_angle").is_not_null())
+            .then(pl.lit("straight"))
+            .otherwise(None)  # Keep null if adj_spray_angle is null
+            .alias("hit_direction")
+        ])
+
+        df = df.with_columns([
+        pl.when(pl.col("hit_direction") == "oppo").then(1).otherwise(None).alias("oppo"),
+        pl.when(pl.col("hit_direction") == "pull").then(1).otherwise(None).alias("pull"),
+        pl.when(pl.col("hit_direction") == "straight").then(1).otherwise(None).alias("straight")
+        ])
+
+        df = df.with_columns([
+            pl.when(pl.col("event_type") == "single").then(1).otherwise(0).alias("single"),
+            pl.when(pl.col("event_type") == "double").then(1).otherwise(0).alias("double"),
+            pl.when(pl.col("event_type") == "triple").then(1).otherwise(0).alias("triple"),
+            pl.when(pl.col("event_type") == "home_run").then(1).otherwise(0).alias("home_run")
+        ])
+
+
+
+
+
+        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('k_minus_bb').sum().alias('k_minus_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'),
+            pl.col('pull').sum().alias('pull'),
+            pl.col('straight').sum().alias('straight'),
+            pl.col('oppo').sum().alias('oppo'),
+            ((pl.col('trajectory_fly_ball') == 1) | (pl.col('trajectory_line_drive') == 1)).sum().alias('fly_line_bip'),
+            (pl.col('pull') & ((pl.col('trajectory_fly_ball') == 1) | (pl.col('trajectory_line_drive') == 1))).sum().alias('pull_fly_ball'),
+            pl.col('single').sum().alias('single'),
+            pl.col('double').sum().alias('double'),
+            pl.col('triple').sum().alias('triple'),
+            pl.col('home_run').sum().alias('home_run'),
+            (pl.col('extension').mean()).alias('extension'),
+            (pl.col('start_speed').filter(pl.col('pitch_type').is_in(['FF','SI'])).mean().alias('avg_start_speed_ff')),
+
+
+
+        ])
+
+        # 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('k_minus_bb') / pl.col('pa')).alias('k_minus_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('out_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'),
+            (pl.col('pull') / pl.col('bip')).alias('pull_percent'),
+            (pl.col('straight') / pl.col('bip')).alias('straight_percent'),
+            (pl.col('oppo') / pl.col('bip')).alias('oppo_percent'),
+            (pl.col('pull_fly_ball') / pl.col('fly_line_bip')).alias('pulled_fly_ball_percent'),
+
+        ])
+
+        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('k_minus_bb').sum().alias('k_minus_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'),
+            pl.col('pull').sum().alias('pull'),
+            pl.col('straight').sum().alias('straight'),
+            pl.col('oppo').sum().alias('oppo'),
+            ((pl.col('trajectory_fly_ball') == 1) | (pl.col('trajectory_line_drive') == 1)).sum().alias('fly_line_bip'),
+            (pl.col('pull') & ((pl.col('trajectory_fly_ball') == 1) | (pl.col('trajectory_line_drive') == 1))).sum().alias('pull_fly_ball'),
+            pl.col('single').sum().alias('single'),
+            pl.col('double').sum().alias('double'),
+            pl.col('triple').sum().alias('triple'),
+            pl.col('home_run').sum().alias('home_run'),
+            (pl.col('extension').mean()).alias('extension'),
+            (pl.col('start_speed').filter(pl.col('pitch_type').is_in(['FF','SI','FC'])).mean().alias('avg_start_speed_ff')),
+
+            
+        ])
+
+        # 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('k_minus_bb') / pl.col('pa')).alias('k_minus_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'),
+            (pl.col('pull') / pl.col('bip')).alias('pull_percent'),
+            (pl.col('straight') / pl.col('bip')).alias('straight_percent'),
+            (pl.col('oppo') / pl.col('bip')).alias('oppo_percent'),
+            (pl.col('pull_fly_ball') / pl.col('fly_line_bip')).alias('pulled_fly_ball_percent'),
+        ])
+
         return df_summ

stuff_model/feature_engineering.py

@@ -82,8 +82,8 @@ def feature_engineering(df: pl.DataFrame) -> pl.DataFrame:
     # Group by pitcher_id and year, then aggregate to calculate average speed and usage percentage
     df_agg = df_filtered.group_by(['pitcher_id', 'year', 'pitch_type']).agg([
         pl.col('start_speed').mean().alias('avg_fastball_speed'),
-        pl.col('az').mean().alias('avg_fastball_az'),
-        pl.col('ax').mean().alias('avg_fastball_ax'),
+        pl.col('ivb').mean().alias('avg_fastball_ivb'),
+        pl.col('hb').mean().alias('avg_fastball_hb'),
         pl.len().alias('count')
     ])
 
@@ -104,25 +104,25 @@ def feature_engineering(df: pl.DataFrame) -> pl.DataFrame:
 
     # If no fastball, use the fastest pitch for avg_fastball_az
     df = df.with_columns(
-        pl.when(pl.col('avg_fastball_az').is_null())
-        .then(pl.col('az').max().over('pitcher_id'))
-        .otherwise(pl.col('avg_fastball_az'))
-        .alias('avg_fastball_az')
+        pl.when(pl.col('avg_fastball_ivb').is_null())
+        .then(pl.col('ivb').max().over('pitcher_id'))
+        .otherwise(pl.col('avg_fastball_ivb'))
+        .alias('avg_fastball_ivb')
     )
 
     # If no fastball, use the fastest pitch for avg_fastball_ax
     df = df.with_columns(
-        pl.when(pl.col('avg_fastball_ax').is_null())
-        .then(pl.col('ax').max().over('pitcher_id'))
-        .otherwise(pl.col('avg_fastball_ax'))
-        .alias('avg_fastball_ax')
+        pl.when(pl.col('avg_fastball_hb').is_null())
+        .then(pl.col('hb').max().over('pitcher_id'))
+        .otherwise(pl.col('avg_fastball_hb'))
+        .alias('avg_fastball_hb')
     )
 
     # Calculate pitch differentials
     df = df.with_columns(
         (pl.col('start_speed') - pl.col('avg_fastball_speed')).alias('speed_diff'),
-        (pl.col('az') - pl.col('avg_fastball_az')).alias('az_diff'),
-        (pl.col('ax') - pl.col('avg_fastball_ax')).abs().alias('ax_diff')
+        (pl.col('ivb') - pl.col('avg_fastball_ivb')).alias('ivb_diff'),
+        (pl.col('hb') - pl.col('avg_fastball_hb')).abs().alias('hb_diff')
     )
 
     # Cast the year column to integer type

stuff_model/stuff_apply.py

@@ -1,57 +1,57 @@
-import polars as pl
-import joblib
-
-model = joblib.load('stuff_model/lgbm_model_2020_2024.joblib')
-# Read the values from the text file
-with open('stuff_model/target_stats.txt', 'r') as file:
-    lines = file.readlines()
-    target_mean = float(lines[0].strip())
-    target_std = float(lines[1].strip())
-    
-# Define the features to be used for training
-features = ['start_speed',
-            'spin_rate',
-            'extension',
-            'az',
-            'ax',
-            'x0',
-            'z0',
-            'speed_diff',
-            'az_diff',
-            'ax_diff']
-
-
-def stuff_apply(df:pl.DataFrame) -> pl.DataFrame:
-    # Filter the dataframe to include only the rows for the year 2024 and drop rows with null values in the specified features and target column
-    # df_test = df.drop_nulls(subset=features)
-    df_test = df.clone()
-
-    # Predict the target values for the 2024 data using the trained model
-    df_test = df_test.with_columns(
-        pl.Series(name="target", values=model.predict(df_test[features].to_numpy()))
-    )
-    # Standardize the target column to create a z-score
-    df_test = df_test.with_columns(
-        ((pl.col('target') - target_mean) / target_std).alias('target_zscore')
-    )
-
-    # Convert the z-score to tj_stuff_plus
-    df_test = df_test.with_columns(
-        (100 - (pl.col('target_zscore') * 10)).alias('tj_stuff_plus')
-    )
-
-    df_pitch_types = pl.read_csv('stuff_model/tj_stuff_plus_pitch.csv')
-
-    # Join the pitch type statistics with the main DataFrame based on pitch_type
-    df_pitch_all = df_test.join(df_pitch_types, left_on='pitch_type', right_on='pitch_type')
-
-    # Normalize pitch_grade values to a range between -0.5 and 0.5 based on the percentiles
-    df_pitch_all = df_pitch_all.with_columns(
-        ((pl.col('tj_stuff_plus') - pl.col('mean')) / pl.col('std')).alias('pitch_grade')
-    )
-
-    # Scale the pitch_grade values to a range between 20 and 80
-    df_pitch_all = df_pitch_all.with_columns(
-        (pl.col('pitch_grade') * 10 + 50).clip(20, 80)
-    )
+import polars as pl
+import joblib
+
+model = joblib.load('stuff_model/stuff_model.joblib')
+# Read the values from the text file
+with open('stuff_model/target_stats.txt', 'r') as file:
+    lines = file.readlines()
+    target_mean = float(lines[0].strip())
+    target_std = float(lines[1].strip())
+    
+# Define the features to be used for training
+features = ['start_speed',
+            'spin_rate',
+            'extension',
+            'ivb',
+            'hb',
+            'x0',
+            'z0',
+            'speed_diff',
+            'ivb_diff',
+            'hb_diff']
+
+
+def stuff_apply(df:pl.DataFrame) -> pl.DataFrame:
+    # Filter the dataframe to include only the rows for the year 2024 and drop rows with null values in the specified features and target column
+    # df_test = df.drop_nulls(subset=features)
+    df_test = df.clone()
+
+    # Predict the target values for the 2024 data using the trained model
+    df_test = df_test.with_columns(
+        pl.Series(name="target", values=model.predict(df_test[features].to_numpy()))
+    )
+    # Standardize the target column to create a z-score
+    df_test = df_test.with_columns(
+        ((pl.col('target') - target_mean) / target_std).alias('target_zscore')
+    )
+
+    # Convert the z-score to tj_stuff_plus
+    df_test = df_test.with_columns(
+        (100 - (pl.col('target_zscore') * 10)).alias('tj_stuff_plus')
+    )
+
+    df_pitch_types = pl.read_csv('stuff_model/tj_stuff_plus_pitch.csv')
+
+    # Join the pitch type statistics with the main DataFrame based on pitch_type
+    df_pitch_all = df_test.join(df_pitch_types, left_on='pitch_type', right_on='pitch_type')
+
+    # Normalize pitch_grade values to a range between -0.5 and 0.5 based on the percentiles
+    df_pitch_all = df_pitch_all.with_columns(
+        ((pl.col('tj_stuff_plus') - pl.col('mean')) / pl.col('std')).alias('pitch_grade')
+    )
+
+    # Scale the pitch_grade values to a range between 20 and 80
+    df_pitch_all = df_pitch_all.with_columns(
+        (pl.col('pitch_grade') * 10 + 50).clip(20, 80)
+    )
     return df_pitch_all

stuff_model/stuff_model.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a13ce41054747c57d91a025e4ee2375a806f01431e686e0f30691fc0244ebae9
+size 3104666

stuff_model/target_stats.txt

@@ -1,2 +1,2 @@
-0.0036819646648982335
-0.006743548730130907
+0.0027508874807158695
+0.0060355533615784425