Enhance app functionality by adding calculations for the highest owned teams and pitchers based on projections, improving user insights into team ownership trends. Refactor small_field_preset function to ensure consistent DataFrame structure by sorting working portfolio by median before returning results.

app.py

@@ -940,6 +940,8 @@ with tab2:
                     axis=1
                 )
             st.session_state['working_frame'] = predict_dupes(st.session_state['working_frame'], st.session_state['map_dict'], site_var, type_var, Contest_Size, strength_var, sport_var)
+            st.session_state['highest_owned_teams'] = st.session_state['projections_df'][~st.session_state['projections_df']['position'].isin(['P', 'SP'])].groupby('team')['ownership'].sum().sort_values(ascending=False).head(3).index.tolist()
+            st.session_state['highest_owned_pitchers'] = st.session_state['projections_df'][st.session_state['projections_df']['position'].isin(['P', 'SP'])]['player_names'].sort_values(by='ownership', ascending=False).head(3).tolist()
             if 'info_columns_dict' not in st.session_state:
                 st.session_state['info_columns_dict'] = {
                     'Dupes': st.session_state['working_frame']['Dupes'],

global_func/small_field_preset.py

@@ -17,6 +17,6 @@ def small_field_preset(portfolio: pd.DataFrame, lineup_target: int):
 
         working_portfolio = working_portfolio.drop(rows_to_drop).reset_index(drop=True)
         if len(working_portfolio) >= lineup_target:
-            return working_portfolio.sort_values(by='Median', ascending=False).head(lineup_target)
+            return working_portfolio.sort_values(by='median', ascending=False).head(lineup_target)
         
-    return working_portfolio.sort_values(by='Median', ascending=False)
+    return working_portfolio.sort_values(by='median', ascending=False)