app.py

import streamlit as st
st.set_page_config(layout="wide")

for name in dir():
    if not name.startswith('_'):
        del globals()[name]

import numpy as np
import pandas as pd
import streamlit as st
import gspread
from database import gcservice_account, NHL_data


percentages_format = {'Shots': '{:.2%}', 'HDCF': '{:.2%}', 'Goals': '{:.2%}', 'Assists': '{:.2%}', 'Blocks': '{:.2%}',
                      'L14_Shots': '{:.2%}', 'L14_HDCF': '{:.2%}', 'L14_Goals': '{:.2%}', 'L14_Assists': '{:.2%}',
                      'L14_Blocks': '{:.2%}', 'Max Goal%': '{:.2%}', 'L14 Max Goal%': '{:.2%}'}

matchups_format = {'HDCF%': '{:.2%}', 'o_HDCA%': '{:.2%}', 'HDCF_m%': '{:.2%}'}

@st.cache_resource(ttl = 599)
def init_baselines():
    parse_hold = pd.DataFrame(columns=['Line', 'SK1', 'SK2', 'SK3', 'Cost', 'Team Total', 'Shots', 'HDCF', 'Goals', 'Assists', 'Blocks',
                              'L14_Shots', 'L14_HDCF', 'L14_Goals', 'L14_Assists', 'L14_Blocks', 'Max Goal%'])
    
    sh = gcservice_account.open_by_url(NHL_data)
    
    worksheet = sh.worksheet('Player_Level_ROO')
    raw_display = pd.DataFrame(worksheet.get_values())
    raw_display.columns = raw_display.iloc[0]
    raw_display = raw_display[1:]
    raw_display = raw_display.reset_index(drop=True)
    raw_display = raw_display[raw_display['Opp'] != ""]
    team_frame = raw_display[['Team', 'Opp']]
    team_list = team_frame['Team'].unique()
    team_dict = dict(zip(team_frame['Team'], team_frame['Opp']))
    
    worksheet = sh.worksheet('Matchups')
    raw_display = pd.DataFrame(worksheet.get_values())
    raw_display.columns = raw_display.iloc[0]
    raw_display = raw_display[1:]
    raw_display = raw_display.reset_index(drop=True)
    raw_display = raw_display[raw_display['Opp'] != ""]
    matchups = raw_display[['Team', 'Opp', 'FL1$', 'FL2$', 'FL3$', 'Team Total', 'Game Pace', 'SF', 'o_SA', 'SF_m', 'HDCF',
                              'o_HDCA', 'HDCF_m', 'HDCF%', 'o_HDCA%', 'HDCF_m%', 'HDSF+']]
    data_cols = matchups.columns.drop(['Team', 'Opp'])
    matchups[data_cols] = matchups[data_cols].apply(pd.to_numeric, errors='coerce')
    matchups = matchups.dropna(subset='HDSF+')
    matchups = matchups.sort_values(by='HDCF_m', ascending=False)
    
    worksheet = sh.worksheet('Marketshares')
    raw_display = pd.DataFrame(worksheet.get_values())
    raw_display.columns = raw_display.iloc[0]
    raw_display = raw_display[1:]
    raw_display = raw_display.reset_index(drop=True)
    # raw_display = raw_display[raw_display['Line'] != ""]
    overall_ms = raw_display[['Line', 'SK1', 'SK2', 'SK3', 'Cost', 'Team Total', 'Shots', 'HDCF', 'Goals', 'Assists', 'Blocks',
                              'L14_Shots', 'L14_HDCF', 'L14_Goals', 'L14_Assists', 'L14_Blocks']]
    pat = '|'.join(team_list)
    s = overall_ms['Line'].str.extract('('+ pat + ')', expand=False)
    overall_ms['Max Goal%'] = overall_ms.groupby(s)['Goals'].transform('max')
    overall_ms['L14 Max Goal%'] = overall_ms.groupby(s)['L14_Goals'].transform('max')
    
    data_cols = overall_ms.columns.drop(['Line', 'SK1', 'SK2', 'SK3'])
    overall_ms[data_cols] = overall_ms[data_cols].apply(pd.to_numeric, errors='coerce')
    overall_ms['Proj Goal'] = overall_ms['Goals'] * overall_ms['Team Total']
    overall_ms['L14 Proj Goal'] = overall_ms['L14_Goals'] * overall_ms['Team Total']
    overall_ms = overall_ms[['Line', 'SK1', 'SK2', 'SK3', 'Cost', 'Team Total', 'Shots', 'HDCF', 'Goals', 'Max Goal%', 'Proj Goal',
                             'Assists', 'Blocks', 'L14_Shots', 'L14_HDCF', 'L14_Goals', 'L14 Max Goal%', 'L14 Proj Goal', 'L14_Assists', 'L14_Blocks']]
    overall_ms = overall_ms.sort_values(by='Shots', ascending=False)
        
    return matchups, overall_ms, team_frame, team_list, team_dict

def convert_df_to_csv(df):
    return df.to_csv().encode('utf-8')

matchups, overall_ms, team_frame, team_list, team_dict = init_baselines()

col1, col2 = st.columns([1, 9])
with col1:
    if st.button("Reset Data", key='reset1'):
              st.cache_data.clear()
              matchups, overall_ms, team_frame, team_list, team_dict = init_baselines()
    split_var1 = st.radio("View matchups or line marketshares?", ('Slate Matchups', 'Line Marketshares'), key='split_var1')
    if split_var1 == "Line Marketshares":
        team_var = st.radio("View all teams or specific teams?", ('All Teams', 'Specific Teams'), key='team_var')
        if team_var == "All Teams":
            team_split = team_frame.Team.values.tolist()
        elif team_var == "Specific Teams":
            team_split = st.multiselect('Which teams would you like to include in the tables?', options = team_frame['Team'].unique(), key='team_var1')
    
with col2:
    if split_var1 == 'Slate Matchups':
        display_table = matchups
        display_table = display_table.set_index('Team')
        st.dataframe(display_table.style.background_gradient(axis=0).background_gradient(cmap='RdYlGn').format(matchups_format, precision=2), height=500, use_container_width = True)
    elif split_var1 == 'Line Marketshares':
        display_table = overall_ms
        display_parsed = display_table[display_table['Line'].str.contains('|'.join(team_split))]
        # display_parsed = display_parsed.set_index('Line')
        st.dataframe(display_parsed.style.background_gradient(axis=0).background_gradient(cmap='RdYlGn').format(percentages_format, precision=2), height=500, use_container_width = True)

if split_var1 == 'Line Marketshares':
    st.download_button(
        label="Export Marketshares (CSV)",
        data=convert_df_to_csv(display_table),
        file_name='Marketshares_export.csv',
        mime='text/csv',
    )
elif split_var1 == 'Slate Matchups':
    st.download_button(
        label="Export Matchups (CSV)",
        data=convert_df_to_csv(display_table),
        file_name='Matchups_export.csv',
        mime='text/csv',
    )