Update app.py

app.py

@@ -1,17 +1,6 @@
-import pulp
 import numpy as np
 import pandas as pd
-import random
-import sys
-import openpyxl
-import re
-import time
 import streamlit as st
-import matplotlib
-from  matplotlib.colors import LinearSegmentedColormap
-from st_aggrid import GridOptionsBuilder, AgGrid, GridUpdateMode, DataReturnMode
-import json
-import requests
 import gspread
 import plotly.figure_factory as ff
 
@@ -41,91 +30,42 @@ american_format = {'First Inning Lead Percentage': '{:.2%}', 'Fifth Inning Lead
 
 master_hold = 'https://docs.google.com/spreadsheets/d/1f42Ergav8K1VsOLOK9MUn7DM_MLMvv4GR2Fy7EfnZTc/edit#gid=340831852'
 
-@st.cache_data
-def load_pitcher_props():
+@st.cache_resource(ttl = 299)
+def init_baselines():
           sh = gc.open_by_url(master_hold)
           worksheet = sh.worksheet('Pitcher_Stats')
           props_frame_hold = pd.DataFrame(worksheet.get_all_records())
           props_frame_hold.rename(columns={"Names": "Player"}, inplace = True)
           props_frame_hold = props_frame_hold[['Player', 'Team', 'BB', 'Hits', 'HRs', 'ERs', 'Ks', 'Outs', 'Fantasy', 'FD_Fantasy', 'PrizePicks']]
-          props_frame_hold = props_frame_hold.drop_duplicates(subset='Player')
+          pitcher_stats = props_frame_hold.drop_duplicates(subset='Player')
           
-          return props_frame_hold
-
-@st.cache_data
-def load_time():
-          sh = gc.open_by_url(master_hold)
           worksheet = sh.worksheet('Timestamp')
           raw_stamp = worksheet.acell('a1').value
           
           t_stamp = f"Last update was at {raw_stamp}"
           
-          return t_stamp
-
-@st.cache_data
-def load_hitter_props():
-          sh = gc.open_by_url(master_hold)
           worksheet = sh.worksheet('Hitter_Stats')
           props_frame_hold = pd.DataFrame(worksheet.get_all_records())
           props_frame_hold.rename(columns={"Names": "Player"}, inplace = True)
           props_frame_hold = props_frame_hold[['Player', 'Team', 'Walks', 'Steals', 'Hits', 'Singles', 'Doubles', 'HRs', 'RBIs', 'Runs', 'Fantasy', 'FD_Fantasy', 'PrizePicks']]
           props_frame_hold['Total Bases'] = props_frame_hold['Singles'] + (props_frame_hold['Doubles'] * 2) + (props_frame_hold['HRs'] * 4)
           props_frame_hold['Hits + Runs + RBIs'] = props_frame_hold['Hits'] + props_frame_hold['Runs'] + props_frame_hold['RBIs']
-          props_frame_hold = props_frame_hold.drop_duplicates(subset='Player')
+          hitter_stats = props_frame_hold.drop_duplicates(subset='Player')
           
-          return props_frame_hold
-
-@st.cache_data
-def load_team_table():
-          sh = gc.open_by_url(master_hold)
           worksheet = sh.worksheet('Game_Betting_Model')
           team_frame = pd.DataFrame(worksheet.get_all_records())
           team_frame = team_frame.drop_duplicates(subset='Names')
           team_frame['Win Percentage'] = team_frame['Win Percentage'].str.replace('%', '').astype('float')/100
           team_frame['Cover Spread Percentage'] = team_frame['Cover Spread Percentage'].str.replace('%', '').astype('float')/100
           
-          return team_frame
-
-@st.cache_data
-def load_strikeout_props():
-          sh = gc.open_by_url(master_hold)
-          worksheet = sh.worksheet('Strikeout_Props')
-          prop_type_frame = pd.DataFrame(worksheet.get_all_records())
-          prop_type_frame = prop_type_frame.drop_duplicates(subset='Player')
-          
-          return prop_type_frame
-
-@st.cache_data
-def load_total_outs_props():
-          sh = gc.open_by_url(master_hold)
-          worksheet = sh.worksheet('Total_Outs_Props')
-          prop_type_frame = pd.DataFrame(worksheet.get_all_records())
-          prop_type_frame = prop_type_frame.drop_duplicates(subset='Player')
-          
-          return prop_type_frame
-
-@st.cache_data
-def load_total_bases_props():
-          sh = gc.open_by_url(master_hold)
-          worksheet = sh.worksheet('Total_Base_Props')
-          prop_type_frame = pd.DataFrame(worksheet.get_all_records())
-          prop_type_frame = prop_type_frame.drop_duplicates(subset='Player')
-          
-          return prop_type_frame
-
-@st.cache_data
-def load_stolen_bases_props():
-          sh = gc.open_by_url(master_hold)
-          worksheet = sh.worksheet('SB_Props')
-          prop_type_frame = pd.DataFrame(worksheet.get_all_records())
-          prop_type_frame = prop_type_frame.drop_duplicates(subset='Player')
+          worksheet = sh.worksheet('prop_frame')
+          raw_display = pd.DataFrame(worksheet.get_all_records())
+          raw_display.replace('', np.nan, inplace=True)
+          prop_frame = raw_display.dropna(subset='Team')
           
-          return prop_type_frame
+          return pitcher_stats, hitter_stats, team_frame, prop_frame, t_stamp
 
-pitcher_frame_hold = load_pitcher_props()
-hitter_frame_hold = load_hitter_props()
-team_frame_hold = load_team_table()
-t_stamp = load_time()
+pitcher_stats, hitter_stats, team_frame, prop_frame, t_stamp = init_baselines()
 
 tab1, tab2, tab3, tab4, tab5 = st.tabs(["Game Betting Model", "Pitcher Prop Projections", "Hitter Prop Projections", "Player Prop Simulations", "Stat Specific Simulations"])
 
@@ -136,12 +76,8 @@ with tab1:
     st.info(t_stamp)
     if st.button("Reset Data", key='reset1'):
               st.cache_data.clear()
-              pitcher_frame_hold = load_pitcher_props()
-              hitter_frame_hold = load_hitter_props()
-              team_frame_hold = load_team_table()
-              t_stamp = load_time()
+              pitcher_stats, hitter_stats, team_frame, prop_frame, t_stamp = init_baselines()
     line_var1 = st.radio('How would you like to display odds?', options = ['Percentage', 'American'], key='line_var1')
-    team_frame = team_frame_hold
     if line_var1 == 'Percentage':
         team_frame = team_frame[['Names', 'Game', 'Win Percentage', 'Spread', 'Cover Spread Percentage', 'Avg Score', 'Game Total', 'Avg Fifth Inning', 'Fifth Inning Lead Percentage']]
         team_frame = team_frame.set_index('Names')
@@ -164,17 +100,14 @@ with tab2:
     st.info(t_stamp)
     if st.button("Reset Data", key='reset2'):
               st.cache_data.clear()
-              pitcher_frame_hold = load_pitcher_props()
-              hitter_frame_hold = load_hitter_props()
-              team_frame_hold = load_team_table()
-              t_stamp = load_time()
+              pitcher_stats, hitter_stats, team_frame, prop_frame, t_stamp = init_baselines()
     split_var1 = st.radio("Would you like to view all teams or specific ones?", ('All', 'Specific Teams'), key='split_var1')
     if split_var1 == 'Specific Teams':
-        team_var1 = st.multiselect('Which teams would you like to include in the tables?', options = pitcher_frame_hold['Team'].unique(), key='team_var1')
+        team_var1 = st.multiselect('Which teams would you like to include in the tables?', options = pitcher_stats['Team'].unique(), key='team_var1')
     elif split_var1 == 'All':
-        team_var1 = pitcher_frame_hold.Team.values.tolist()
-    pitcher_frame_hold = pitcher_frame_hold[pitcher_frame_hold['Team'].isin(team_var1)]
-    pitcher_frame = pitcher_frame_hold.set_index('Player')
+        team_var1 = pitcher_stats.Team.values.tolist()
+    pitcher_stats = pitcher_stats[pitcher_stats['Team'].isin(team_var1)]
+    pitcher_frame = pitcher_stats.set_index('Player')
     pitcher_frame = pitcher_frame.sort_values(by='Ks', ascending=False)
     st.dataframe(pitcher_frame.style.background_gradient(axis=0).background_gradient(cmap='RdYlGn').format(precision=2), use_container_width = True)
     st.download_button(
@@ -189,17 +122,14 @@ with tab3:
     st.info(t_stamp)
     if st.button("Reset Data", key='reset3'):
               st.cache_data.clear()
-              pitcher_frame_hold = load_pitcher_props()
-              hitter_frame_hold = load_hitter_props()
-              team_frame_hold = load_team_table()
-              t_stamp = load_time()
+              pitcher_stats, hitter_stats, team_frame, prop_frame, t_stamp = init_baselines()
     split_var2 = st.radio("Would you like to view all teams or specific ones?", ('All', 'Specific Teams'), key='split_var2')
     if split_var2 == 'Specific Teams':
-        team_var2 = st.multiselect('Which teams would you like to include in the tables?', options = hitter_frame_hold['Team'].unique(), key='team_var2')
+        team_var2 = st.multiselect('Which teams would you like to include in the tables?', options = hitter_stats['Team'].unique(), key='team_var2')
     elif split_var2 == 'All':
-        team_var2 = hitter_frame_hold.Team.values.tolist()
-    hitter_frame_hold = hitter_frame_hold[hitter_frame_hold['Team'].isin(team_var2)]
-    hitter_frame = hitter_frame_hold.set_index('Player')
+        team_var2 = hitter_stats.Team.values.tolist()
+    hitter_stats = hitter_stats[hitter_stats['Team'].isin(team_var2)]
+    hitter_frame = hitter_stats.set_index('Player')
     hitter_frame = hitter_frame.sort_values(by='Hits + Runs + RBIs', ascending=False)
     st.dataframe(hitter_frame.style.background_gradient(axis=0).background_gradient(cmap='RdYlGn').format(precision=2), use_container_width = True)
     st.download_button(
@@ -214,10 +144,7 @@ with tab4:
     st.info(t_stamp)
     if st.button("Reset Data", key='reset4'):
               st.cache_data.clear()
-              pitcher_frame_hold = load_pitcher_props()
-              hitter_frame_hold = load_hitter_props()
-              team_frame_hold = load_team_table()
-              t_stamp = load_time()
+              pitcher_stats, hitter_stats, team_frame, prop_frame, t_stamp = init_baselines()
     col1, col2 = st.columns([1, 5])
     
     with col2:
@@ -228,10 +155,10 @@ with tab4:
     with col1:
         prop_group_var = st.selectbox('What kind of props are you simulating?', options = ['Pitchers', 'Hitters'])
         if prop_group_var == 'Pitchers':
-            player_check = st.selectbox('Select player to simulate props', options = pitcher_frame_hold['Player'].unique())
+            player_check = st.selectbox('Select player to simulate props', options = pitcher_stats['Player'].unique())
             prop_type_var = st.selectbox('Select type of prop to simulate', options = ['Strikeouts', 'Walks', 'Hits', 'Homeruns', 'Earned Runs', 'Outs', 'Fantasy', 'FD_Fantasy', 'PrizePicks'])
         elif prop_group_var == 'Hitters':
-            player_check = st.selectbox('Select player to simulate props', options = hitter_frame_hold['Player'].unique())
+            player_check = st.selectbox('Select player to simulate props', options = hitter_stats['Player'].unique())
             prop_type_var = st.selectbox('Select type of prop to simulate', options = ['Total Bases', 'Walks', 'Steals', 'Hits', 'Singles', 'Doubles', 'Homeruns', 'RBIs', 'Runs', 'Hits + Runs + RBIs', 'Fantasy', 'FD_Fantasy', 'PrizePicks'])
         
         ou_var = st.selectbox('Select wether it is an over or under', options = ['Over', 'Under'])
@@ -245,9 +172,9 @@ with tab4:
                     with df_hold_container.container():
 
                         if prop_group_var == 'Pitchers':
-                            df = pitcher_frame_hold
+                            df = pitcher_stats
                         elif prop_group_var == 'Hitters':
-                            df = hitter_frame_hold
+                            df = hitter_stats
 
                         total_sims = 1000
 
@@ -384,7 +311,7 @@ with tab5:
               # Clear values from *all* all in-memory and on-disk data caches:
               # i.e. clear values from both square and cube
               st.cache_data.clear()
-              t_stamp = load_time()
+              pitcher_stats, hitter_stats, team_frame, prop_frame, t_stamp = init_baselines()
     col1, col2 = st.columns([1, 5])
     
     with col2:
@@ -402,8 +329,8 @@ with tab5:
                     with df_hold_container.container():
 
                         if prop_type_var == "Strikeouts (Pitchers)":
-                            player_df = pitcher_frame_hold
-                            prop_df = load_strikeout_props()
+                            player_df = pitcher_stats
+                            prop_df = pitcher_stats[pitcher_stats['prop_type'] == 'pitcher_strikeouts']
                             prop_df = prop_df[['Player', 'over_prop', 'over_line', 'under_line']]
                             prop_df.rename(columns={"over_prop": "Prop"}, inplace = True)
                             prop_df = prop_df.loc[prop_df['Prop'] != 0]
@@ -411,8 +338,8 @@ with tab5:
                             prop_df['Under'] = np.where(prop_df['under_line'] < 0, (-(prop_df['under_line'])/((-(prop_df['under_line']))+100)), 100/(prop_df['under_line']+100))
                             df = pd.merge(player_df, prop_df, how='left', left_on=['Player'], right_on = ['Player'])
                         elif prop_type_var == "Total Outs (Pitchers)":
-                            player_df = pitcher_frame_hold
-                            prop_df = load_total_outs_props()
+                            player_df = pitcher_stats
+                            prop_df = pitcher_stats[pitcher_stats['prop_type'] == 'pitcher_strikeouts']
                             prop_df = prop_df[['Player', 'over_prop', 'over_line', 'under_line']]
                             prop_df.rename(columns={"over_prop": "Prop"}, inplace = True)   
                             prop_df = prop_df.loc[prop_df['Prop'] != 0]
@@ -420,8 +347,8 @@ with tab5:
                             prop_df['Under'] = np.where(prop_df['under_line'] < 0, (-(prop_df['under_line'])/((-(prop_df['under_line']))+100)), 100/(prop_df['under_line']+100))
                             df = pd.merge(player_df, prop_df, how='left', left_on=['Player'], right_on = ['Player'])
                         elif prop_type_var == "Total Bases (Hitters)":
-                            player_df = hitter_frame_hold
-                            prop_df = load_total_bases_props()
+                            player_df = hitter_stats
+                            prop_df = hitter_stats[hitter_stats['prop_type'] == 'batter_total_bases']
                             prop_df = prop_df[['Player', 'over_prop', 'over_line', 'under_line']]
                             prop_df.rename(columns={"over_prop": "Prop"}, inplace = True)
                             prop_df = prop_df.loc[prop_df['Prop'] != 0]
@@ -429,8 +356,8 @@ with tab5:
                             prop_df['Under'] = np.where(prop_df['under_line'] < 0, (-(prop_df['under_line'])/((-(prop_df['under_line']))+100)), 100/(prop_df['under_line']+100))
                             df = pd.merge(player_df, prop_df, how='left', left_on=['Player'], right_on = ['Player'])
                         elif prop_type_var == "Stolen Bases (Hitters)":
-                            player_df = hitter_frame_hold
-                            prop_df = load_stolen_base_props()
+                            player_df = hitter_stats
+                            prop_df = hitter_stats[hitter_stats['prop_type'] == 'batter_stolen_bases']
                             prop_df = prop_df[['Player', 'over_prop', 'over_line', 'under_line']]
                             prop_df.rename(columns={"over_prop": "Prop"}, inplace = True)
                             prop_df = prop_df.loc[prop_df['Prop'] != 0]