{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b22c6ee8",
   "metadata": {},
   "source": [
    "player_match_dataset_creation code for past seasons\n",
    "\n",
    "Select the season in season variable, and repeat the code if needed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "8a74364d",
   "metadata": {},
   "outputs": [],
   "source": [
    "season = '2021'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "8a6867a5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\nicol\\AppData\\Local\\Temp\\ipykernel_5084\\813865162.py:21: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  team_data['team_idx'][i] = team_data['team_idx'][i].split(' ')[-1]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>team</th>\n",
       "      <th>team_players_used</th>\n",
       "      <th>team_possession</th>\n",
       "      <th>team_games</th>\n",
       "      <th>team_games_starts</th>\n",
       "      <th>team_minutes</th>\n",
       "      <th>team_goals</th>\n",
       "      <th>team_assists</th>\n",
       "      <th>team_pens_made</th>\n",
       "      <th>team_pens_att</th>\n",
       "      <th>...</th>\n",
       "      <th>vs_team_fouls</th>\n",
       "      <th>vs_team_fouled</th>\n",
       "      <th>vs_team_offsides</th>\n",
       "      <th>vs_team_pens_won</th>\n",
       "      <th>vs_team_pens_conceded</th>\n",
       "      <th>vs_team_own_goals</th>\n",
       "      <th>vs_team_ball_recoveries</th>\n",
       "      <th>vs_team_aerials_won</th>\n",
       "      <th>vs_team_aerials_lost</th>\n",
       "      <th>vs_team_aerials_won_pct</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>team_idx</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Atalanta</th>\n",
       "      <td>Atalanta</td>\n",
       "      <td>30.0</td>\n",
       "      <td>54.6</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>65.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>504.0</td>\n",
       "      <td>548.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2251.0</td>\n",
       "      <td>506.0</td>\n",
       "      <td>633.0</td>\n",
       "      <td>44.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Benevento</th>\n",
       "      <td>Benevento</td>\n",
       "      <td>29.0</td>\n",
       "      <td>42.6</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>621.0</td>\n",
       "      <td>463.0</td>\n",
       "      <td>94.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1854.0</td>\n",
       "      <td>615.0</td>\n",
       "      <td>496.0</td>\n",
       "      <td>55.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bologna</th>\n",
       "      <td>Bologna</td>\n",
       "      <td>37.0</td>\n",
       "      <td>50.8</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>49.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>495.0</td>\n",
       "      <td>495.0</td>\n",
       "      <td>71.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2024.0</td>\n",
       "      <td>594.0</td>\n",
       "      <td>564.0</td>\n",
       "      <td>51.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cagliari</th>\n",
       "      <td>Cagliari</td>\n",
       "      <td>32.0</td>\n",
       "      <td>44.8</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>...</td>\n",
       "      <td>507.0</td>\n",
       "      <td>496.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1920.0</td>\n",
       "      <td>602.0</td>\n",
       "      <td>637.0</td>\n",
       "      <td>48.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Crotone</th>\n",
       "      <td>Crotone</td>\n",
       "      <td>31.0</td>\n",
       "      <td>46.3</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>...</td>\n",
       "      <td>524.0</td>\n",
       "      <td>499.0</td>\n",
       "      <td>103.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2130.0</td>\n",
       "      <td>534.0</td>\n",
       "      <td>472.0</td>\n",
       "      <td>53.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fiorentina</th>\n",
       "      <td>Fiorentina</td>\n",
       "      <td>30.0</td>\n",
       "      <td>46.3</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>545.0</td>\n",
       "      <td>500.0</td>\n",
       "      <td>94.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1915.0</td>\n",
       "      <td>544.0</td>\n",
       "      <td>538.0</td>\n",
       "      <td>50.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Genoa</th>\n",
       "      <td>Genoa</td>\n",
       "      <td>34.0</td>\n",
       "      <td>44.8</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>502.0</td>\n",
       "      <td>501.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2197.0</td>\n",
       "      <td>643.0</td>\n",
       "      <td>505.0</td>\n",
       "      <td>56.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Verona</th>\n",
       "      <td>Hellas Verona</td>\n",
       "      <td>32.0</td>\n",
       "      <td>49.8</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>...</td>\n",
       "      <td>557.0</td>\n",
       "      <td>593.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2102.0</td>\n",
       "      <td>832.0</td>\n",
       "      <td>749.0</td>\n",
       "      <td>52.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Inter</th>\n",
       "      <td>Inter</td>\n",
       "      <td>25.0</td>\n",
       "      <td>52.5</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>84.0</td>\n",
       "      <td>62.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>...</td>\n",
       "      <td>444.0</td>\n",
       "      <td>439.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1690.0</td>\n",
       "      <td>428.0</td>\n",
       "      <td>439.0</td>\n",
       "      <td>49.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Juventus</th>\n",
       "      <td>Juventus</td>\n",
       "      <td>30.0</td>\n",
       "      <td>56.9</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>76.0</td>\n",
       "      <td>59.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>...</td>\n",
       "      <td>490.0</td>\n",
       "      <td>483.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1805.0</td>\n",
       "      <td>378.0</td>\n",
       "      <td>426.0</td>\n",
       "      <td>47.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lazio</th>\n",
       "      <td>Lazio</td>\n",
       "      <td>30.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>59.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>...</td>\n",
       "      <td>454.0</td>\n",
       "      <td>480.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2054.0</td>\n",
       "      <td>510.0</td>\n",
       "      <td>539.0</td>\n",
       "      <td>48.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Milan</th>\n",
       "      <td>Milan</td>\n",
       "      <td>29.0</td>\n",
       "      <td>51.7</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>72.0</td>\n",
       "      <td>47.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>...</td>\n",
       "      <td>502.0</td>\n",
       "      <td>486.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1977.0</td>\n",
       "      <td>539.0</td>\n",
       "      <td>567.0</td>\n",
       "      <td>48.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Napoli</th>\n",
       "      <td>Napoli</td>\n",
       "      <td>25.0</td>\n",
       "      <td>55.1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>56.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>...</td>\n",
       "      <td>580.0</td>\n",
       "      <td>362.0</td>\n",
       "      <td>72.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1811.0</td>\n",
       "      <td>404.0</td>\n",
       "      <td>414.0</td>\n",
       "      <td>49.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Parma</th>\n",
       "      <td>Parma</td>\n",
       "      <td>42.0</td>\n",
       "      <td>47.8</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>...</td>\n",
       "      <td>508.0</td>\n",
       "      <td>486.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1950.0</td>\n",
       "      <td>474.0</td>\n",
       "      <td>630.0</td>\n",
       "      <td>42.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Roma</th>\n",
       "      <td>Roma</td>\n",
       "      <td>31.0</td>\n",
       "      <td>52.2</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>67.0</td>\n",
       "      <td>51.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>...</td>\n",
       "      <td>453.0</td>\n",
       "      <td>536.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1924.0</td>\n",
       "      <td>414.0</td>\n",
       "      <td>428.0</td>\n",
       "      <td>49.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sampdoria</th>\n",
       "      <td>Sampdoria</td>\n",
       "      <td>27.0</td>\n",
       "      <td>45.2</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>52.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>520.0</td>\n",
       "      <td>473.0</td>\n",
       "      <td>111.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2132.0</td>\n",
       "      <td>597.0</td>\n",
       "      <td>628.0</td>\n",
       "      <td>48.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sassuolo</th>\n",
       "      <td>Sassuolo</td>\n",
       "      <td>28.0</td>\n",
       "      <td>60.3</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>36.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>502.0</td>\n",
       "      <td>468.0</td>\n",
       "      <td>91.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1843.0</td>\n",
       "      <td>450.0</td>\n",
       "      <td>408.0</td>\n",
       "      <td>52.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Spezia</th>\n",
       "      <td>Spezia</td>\n",
       "      <td>35.0</td>\n",
       "      <td>51.9</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>52.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>...</td>\n",
       "      <td>556.0</td>\n",
       "      <td>548.0</td>\n",
       "      <td>140.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1985.0</td>\n",
       "      <td>568.0</td>\n",
       "      <td>537.0</td>\n",
       "      <td>51.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Torino</th>\n",
       "      <td>Torino</td>\n",
       "      <td>29.0</td>\n",
       "      <td>47.3</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>49.0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>632.0</td>\n",
       "      <td>544.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1936.0</td>\n",
       "      <td>622.0</td>\n",
       "      <td>595.0</td>\n",
       "      <td>51.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Udinese</th>\n",
       "      <td>Udinese</td>\n",
       "      <td>33.0</td>\n",
       "      <td>46.3</td>\n",
       "      <td>38.0</td>\n",
       "      <td>418.0</td>\n",
       "      <td>3420.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>559.0</td>\n",
       "      <td>474.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1867.0</td>\n",
       "      <td>442.0</td>\n",
       "      <td>491.0</td>\n",
       "      <td>47.4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20 rows × 321 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     team  team_players_used  team_possession  team_games  \\\n",
       "team_idx                                                                    \n",
       "Atalanta         Atalanta               30.0             54.6        38.0   \n",
       "Benevento       Benevento               29.0             42.6        38.0   \n",
       "Bologna           Bologna               37.0             50.8        38.0   \n",
       "Cagliari         Cagliari               32.0             44.8        38.0   \n",
       "Crotone           Crotone               31.0             46.3        38.0   \n",
       "Fiorentina     Fiorentina               30.0             46.3        38.0   \n",
       "Genoa               Genoa               34.0             44.8        38.0   \n",
       "Verona      Hellas Verona               32.0             49.8        38.0   \n",
       "Inter               Inter               25.0             52.5        38.0   \n",
       "Juventus         Juventus               30.0             56.9        38.0   \n",
       "Lazio               Lazio               30.0             53.0        38.0   \n",
       "Milan               Milan               29.0             51.7        38.0   \n",
       "Napoli             Napoli               25.0             55.1        38.0   \n",
       "Parma               Parma               42.0             47.8        38.0   \n",
       "Roma                 Roma               31.0             52.2        38.0   \n",
       "Sampdoria       Sampdoria               27.0             45.2        38.0   \n",
       "Sassuolo         Sassuolo               28.0             60.3        38.0   \n",
       "Spezia             Spezia               35.0             51.9        38.0   \n",
       "Torino             Torino               29.0             47.3        38.0   \n",
       "Udinese           Udinese               33.0             46.3        38.0   \n",
       "\n",
       "            team_games_starts  team_minutes  team_goals  team_assists  \\\n",
       "team_idx                                                                \n",
       "Atalanta                418.0        3420.0        90.0          65.0   \n",
       "Benevento               418.0        3420.0        40.0          26.0   \n",
       "Bologna                 418.0        3420.0        49.0          38.0   \n",
       "Cagliari                418.0        3420.0        43.0          30.0   \n",
       "Crotone                 418.0        3420.0        44.0          27.0   \n",
       "Fiorentina              418.0        3420.0        44.0          31.0   \n",
       "Genoa                   418.0        3420.0        46.0          32.0   \n",
       "Verona                  418.0        3420.0        38.0          30.0   \n",
       "Inter                   418.0        3420.0        84.0          62.0   \n",
       "Juventus                418.0        3420.0        76.0          59.0   \n",
       "Lazio                   418.0        3420.0        59.0          40.0   \n",
       "Milan                   418.0        3420.0        72.0          47.0   \n",
       "Napoli                  418.0        3420.0        83.0          56.0   \n",
       "Parma                   418.0        3420.0        39.0          26.0   \n",
       "Roma                    418.0        3420.0        67.0          51.0   \n",
       "Sampdoria               418.0        3420.0        52.0          37.0   \n",
       "Sassuolo                418.0        3420.0        61.0          36.0   \n",
       "Spezia                  418.0        3420.0        52.0          37.0   \n",
       "Torino                  418.0        3420.0        49.0          28.0   \n",
       "Udinese                 418.0        3420.0        40.0          30.0   \n",
       "\n",
       "            team_pens_made  team_pens_att  ...  vs_team_fouls  vs_team_fouled  \\\n",
       "team_idx                                   ...                                  \n",
       "Atalanta               6.0            7.0  ...          504.0           548.0   \n",
       "Benevento              5.0            7.0  ...          621.0           463.0   \n",
       "Bologna                3.0            4.0  ...          495.0           495.0   \n",
       "Cagliari               4.0            5.0  ...          507.0           496.0   \n",
       "Crotone                9.0            9.0  ...          524.0           499.0   \n",
       "Fiorentina             6.0            6.0  ...          545.0           500.0   \n",
       "Genoa                  4.0            4.0  ...          502.0           501.0   \n",
       "Verona                 3.0            3.0  ...          557.0           593.0   \n",
       "Inter                  8.0            9.0  ...          444.0           439.0   \n",
       "Juventus               8.0           10.0  ...          490.0           483.0   \n",
       "Lazio                  6.0           10.0  ...          454.0           480.0   \n",
       "Milan                 15.0           20.0  ...          502.0           486.0   \n",
       "Napoli                 7.0            8.0  ...          580.0           362.0   \n",
       "Parma                  5.0            5.0  ...          508.0           486.0   \n",
       "Roma                   7.0            8.0  ...          453.0           536.0   \n",
       "Sampdoria              5.0            6.0  ...          520.0           473.0   \n",
       "Sassuolo              13.0           14.0  ...          502.0           468.0   \n",
       "Spezia                 4.0            5.0  ...          556.0           548.0   \n",
       "Torino                 5.0            6.0  ...          632.0           544.0   \n",
       "Udinese                4.0            4.0  ...          559.0           474.0   \n",
       "\n",
       "            vs_team_offsides  vs_team_pens_won  vs_team_pens_conceded  \\\n",
       "team_idx                                                                \n",
       "Atalanta                40.0               5.0                    7.0   \n",
       "Benevento               94.0               4.0                    7.0   \n",
       "Bologna                 71.0               6.0                    4.0   \n",
       "Cagliari                83.0               4.0                    5.0   \n",
       "Crotone                103.0               4.0                    9.0   \n",
       "Fiorentina              94.0               9.0                    6.0   \n",
       "Genoa                   37.0               7.0                    4.0   \n",
       "Verona                  42.0               6.0                    3.0   \n",
       "Inter                   32.0               5.0                    9.0   \n",
       "Juventus                40.0               6.0                   10.0   \n",
       "Lazio                   43.0              10.0                   10.0   \n",
       "Milan                   44.0               5.0                   20.0   \n",
       "Napoli                  72.0               5.0                    8.0   \n",
       "Parma                   64.0               3.0                    5.0   \n",
       "Roma                   125.0               5.0                    8.0   \n",
       "Sampdoria              111.0               5.0                    6.0   \n",
       "Sassuolo                91.0               5.0                   14.0   \n",
       "Spezia                 140.0               7.0                    5.0   \n",
       "Torino                  75.0               8.0                    6.0   \n",
       "Udinese                 42.0               9.0                    4.0   \n",
       "\n",
       "            vs_team_own_goals  vs_team_ball_recoveries  vs_team_aerials_won  \\\n",
       "team_idx                                                                      \n",
       "Atalanta                  0.0                   2251.0                506.0   \n",
       "Benevento                 0.0                   1854.0                615.0   \n",
       "Bologna                   2.0                   2024.0                594.0   \n",
       "Cagliari                  0.0                   1920.0                602.0   \n",
       "Crotone                   1.0                   2130.0                534.0   \n",
       "Fiorentina                3.0                   1915.0                544.0   \n",
       "Genoa                     1.0                   2197.0                643.0   \n",
       "Verona                    5.0                   2102.0                832.0   \n",
       "Inter                     5.0                   1690.0                428.0   \n",
       "Juventus                  1.0                   1805.0                378.0   \n",
       "Lazio                     2.0                   2054.0                510.0   \n",
       "Milan                     2.0                   1977.0                539.0   \n",
       "Napoli                    3.0                   1811.0                404.0   \n",
       "Parma                     0.0                   1950.0                474.0   \n",
       "Roma                      1.0                   1924.0                414.0   \n",
       "Sampdoria                 0.0                   2132.0                597.0   \n",
       "Sassuolo                  3.0                   1843.0                450.0   \n",
       "Spezia                    0.0                   1985.0                568.0   \n",
       "Torino                    1.0                   1936.0                622.0   \n",
       "Udinese                   2.0                   1867.0                442.0   \n",
       "\n",
       "            vs_team_aerials_lost  vs_team_aerials_won_pct  \n",
       "team_idx                                                   \n",
       "Atalanta                   633.0                     44.4  \n",
       "Benevento                  496.0                     55.4  \n",
       "Bologna                    564.0                     51.3  \n",
       "Cagliari                   637.0                     48.6  \n",
       "Crotone                    472.0                     53.1  \n",
       "Fiorentina                 538.0                     50.3  \n",
       "Genoa                      505.0                     56.0  \n",
       "Verona                     749.0                     52.6  \n",
       "Inter                      439.0                     49.4  \n",
       "Juventus                   426.0                     47.0  \n",
       "Lazio                      539.0                     48.6  \n",
       "Milan                      567.0                     48.7  \n",
       "Napoli                     414.0                     49.4  \n",
       "Parma                      630.0                     42.9  \n",
       "Roma                       428.0                     49.2  \n",
       "Sampdoria                  628.0                     48.7  \n",
       "Sassuolo                   408.0                     52.4  \n",
       "Spezia                     537.0                     51.4  \n",
       "Torino                     595.0                     51.1  \n",
       "Udinese                    491.0                     47.4  \n",
       "\n",
       "[20 rows x 321 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "rcsv = pd.read_csv('fbref_data/season' + season + '/teams.csv')   \n",
    "team = pd.DataFrame(rcsv)\n",
    "\n",
    "rcsv = pd.read_csv('fbref_data/season' + season + '/teams_vs.csv')   \n",
    "vs_team = pd.DataFrame(rcsv)\n",
    "\n",
    "for i in range(1, team.shape[1]):\n",
    "    team = team.rename(columns = {team.columns[i] : 'team_' + team.columns[i]})\n",
    "    \n",
    "for i in range(1, vs_team.shape[1]):\n",
    "    vs_team = vs_team.rename(columns = {vs_team.columns[i] : 'vs_team_' + vs_team.columns[i]})\n",
    "    \n",
    "vs_team.pop('team')\n",
    "\n",
    "team_data = pd.concat([team, vs_team], axis = 1)\n",
    "\n",
    "team_data['team_idx'] = team_data['team']\n",
    "for i in range(team_data.shape[0]):\n",
    "    team_data['team_idx'][i] = team_data['team_idx'][i].split(' ')[-1]\n",
    "\n",
    "team_data = team_data.set_index('team_idx')\n",
    "\n",
    "team_data\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "b6a3db98",
   "metadata": {},
   "outputs": [],
   "source": [
    "team_data.to_excel('mid_outputs/season' + season + '/team_data.xlsx')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "c34eb398",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>r</th>\n",
       "      <th>team</th>\n",
       "      <th>surname</th>\n",
       "      <th>initial</th>\n",
       "      <th>fb_ID</th>\n",
       "      <th>age</th>\n",
       "      <th>birth_year</th>\n",
       "      <th>games</th>\n",
       "      <th>...</th>\n",
       "      <th>gk_pct_goal_kicks_launched</th>\n",
       "      <th>gk_goal_kick_length_avg</th>\n",
       "      <th>gk_crosses</th>\n",
       "      <th>gk_crosses_stopped</th>\n",
       "      <th>gk_crosses_stopped_pct</th>\n",
       "      <th>gk_def_actions_outside_pen_area</th>\n",
       "      <th>gk_def_actions_outside_pen_area_per90</th>\n",
       "      <th>gk_avg_distance_def_actions</th>\n",
       "      <th>vote_avg</th>\n",
       "      <th>vote_std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Donnarumma G.</th>\n",
       "      <td>0</td>\n",
       "      <td>350</td>\n",
       "      <td>P</td>\n",
       "      <td>Milan</td>\n",
       "      <td>Donnarumma</td>\n",
       "      <td>G</td>\n",
       "      <td>629</td>\n",
       "      <td>21</td>\n",
       "      <td>1999</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>41.5</td>\n",
       "      <td>33.7</td>\n",
       "      <td>556.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>43.0</td>\n",
       "      <td>1.16</td>\n",
       "      <td>14.8</td>\n",
       "      <td>6.283784</td>\n",
       "      <td>0.394909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Szczesny</th>\n",
       "      <td>1</td>\n",
       "      <td>453</td>\n",
       "      <td>P</td>\n",
       "      <td>Juventus</td>\n",
       "      <td>Szczesny</td>\n",
       "      <td>NaN</td>\n",
       "      <td>664</td>\n",
       "      <td>30</td>\n",
       "      <td>1990</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>10.6</td>\n",
       "      <td>22.0</td>\n",
       "      <td>399.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>1.24</td>\n",
       "      <td>16.2</td>\n",
       "      <td>6.216667</td>\n",
       "      <td>0.441273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Handanovic</th>\n",
       "      <td>2</td>\n",
       "      <td>250</td>\n",
       "      <td>P</td>\n",
       "      <td>Inter</td>\n",
       "      <td>Handanovic</td>\n",
       "      <td>NaN</td>\n",
       "      <td>635</td>\n",
       "      <td>36</td>\n",
       "      <td>1984</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>26.9</td>\n",
       "      <td>30.8</td>\n",
       "      <td>441.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>2.9</td>\n",
       "      <td>22.0</td>\n",
       "      <td>0.61</td>\n",
       "      <td>14.9</td>\n",
       "      <td>6.027027</td>\n",
       "      <td>0.505990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gollini</th>\n",
       "      <td>3</td>\n",
       "      <td>610</td>\n",
       "      <td>P</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Gollini</td>\n",
       "      <td>NaN</td>\n",
       "      <td>634</td>\n",
       "      <td>25</td>\n",
       "      <td>1995</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>63.8</td>\n",
       "      <td>49.7</td>\n",
       "      <td>234.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2.6</td>\n",
       "      <td>39.0</td>\n",
       "      <td>1.63</td>\n",
       "      <td>17.4</td>\n",
       "      <td>6.100000</td>\n",
       "      <td>0.632456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Audero</th>\n",
       "      <td>4</td>\n",
       "      <td>761</td>\n",
       "      <td>P</td>\n",
       "      <td>Sampdoria</td>\n",
       "      <td>Audero</td>\n",
       "      <td>NaN</td>\n",
       "      <td>620</td>\n",
       "      <td>23</td>\n",
       "      <td>1997</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>64.1</td>\n",
       "      <td>45.3</td>\n",
       "      <td>563.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>2.1</td>\n",
       "      <td>34.0</td>\n",
       "      <td>0.92</td>\n",
       "      <td>14.8</td>\n",
       "      <td>6.378378</td>\n",
       "      <td>0.409439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Rabbi</th>\n",
       "      <td>559</td>\n",
       "      <td>5361</td>\n",
       "      <td>A</td>\n",
       "      <td>Bologna</td>\n",
       "      <td>Rabbi</td>\n",
       "      <td>NaN</td>\n",
       "      <td>464</td>\n",
       "      <td>18</td>\n",
       "      <td>2001</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.027304</td>\n",
       "      <td>0.472755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Oddei</th>\n",
       "      <td>560</td>\n",
       "      <td>5378</td>\n",
       "      <td>A</td>\n",
       "      <td>Sassuolo</td>\n",
       "      <td>Oddei</td>\n",
       "      <td>NaN</td>\n",
       "      <td>400</td>\n",
       "      <td>17</td>\n",
       "      <td>2002</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.207968</td>\n",
       "      <td>0.447162</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Karamoko Is.</th>\n",
       "      <td>561</td>\n",
       "      <td>5414</td>\n",
       "      <td>A</td>\n",
       "      <td>Sassuolo</td>\n",
       "      <td>Karamoko</td>\n",
       "      <td>I</td>\n",
       "      <td>267</td>\n",
       "      <td>18</td>\n",
       "      <td>2002</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.186813</td>\n",
       "      <td>0.444240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Correia</th>\n",
       "      <td>562</td>\n",
       "      <td>5419</td>\n",
       "      <td>A</td>\n",
       "      <td>Juventus</td>\n",
       "      <td>Correia</td>\n",
       "      <td>NaN</td>\n",
       "      <td>113</td>\n",
       "      <td>19</td>\n",
       "      <td>2001</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.600386</td>\n",
       "      <td>0.396770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Raimondo</th>\n",
       "      <td>563</td>\n",
       "      <td>5436</td>\n",
       "      <td>A</td>\n",
       "      <td>Bologna</td>\n",
       "      <td>Raimondo</td>\n",
       "      <td>NaN</td>\n",
       "      <td>470</td>\n",
       "      <td>16</td>\n",
       "      <td>2004</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.209072</td>\n",
       "      <td>0.275318</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>564 rows × 169 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               Unnamed: 0    id  r       team     surname initial  fb_ID  age  \\\n",
       "name                                                                            \n",
       "Donnarumma G.           0   350  P      Milan  Donnarumma       G    629   21   \n",
       "Szczesny                1   453  P   Juventus    Szczesny     NaN    664   30   \n",
       "Handanovic              2   250  P      Inter  Handanovic     NaN    635   36   \n",
       "Gollini                 3   610  P   Atalanta     Gollini     NaN    634   25   \n",
       "Audero                  4   761  P  Sampdoria      Audero     NaN    620   23   \n",
       "...                   ...   ... ..        ...         ...     ...    ...  ...   \n",
       "Rabbi                 559  5361  A    Bologna       Rabbi     NaN    464   18   \n",
       "Oddei                 560  5378  A   Sassuolo       Oddei     NaN    400   17   \n",
       "Karamoko Is.          561  5414  A   Sassuolo    Karamoko       I    267   18   \n",
       "Correia               562  5419  A   Juventus     Correia     NaN    113   19   \n",
       "Raimondo              563  5436  A    Bologna    Raimondo     NaN    470   16   \n",
       "\n",
       "               birth_year  games  ...  gk_pct_goal_kicks_launched  \\\n",
       "name                              ...                               \n",
       "Donnarumma G.        1999      0  ...                        41.5   \n",
       "Szczesny             1990      0  ...                        10.6   \n",
       "Handanovic           1984      0  ...                        26.9   \n",
       "Gollini              1995      0  ...                        63.8   \n",
       "Audero               1997      0  ...                        64.1   \n",
       "...                   ...    ...  ...                         ...   \n",
       "Rabbi                2001      4  ...                         0.0   \n",
       "Oddei                2002      5  ...                         0.0   \n",
       "Karamoko Is.         2002      1  ...                         0.0   \n",
       "Correia              2001      1  ...                         0.0   \n",
       "Raimondo             2004      1  ...                         0.0   \n",
       "\n",
       "               gk_goal_kick_length_avg  gk_crosses  gk_crosses_stopped  \\\n",
       "name                                                                     \n",
       "Donnarumma G.                     33.7       556.0                29.0   \n",
       "Szczesny                          22.0       399.0                12.0   \n",
       "Handanovic                        30.8       441.0                13.0   \n",
       "Gollini                           49.7       234.0                 6.0   \n",
       "Audero                            45.3       563.0                12.0   \n",
       "...                                ...         ...                 ...   \n",
       "Rabbi                              0.0         0.0                 0.0   \n",
       "Oddei                              0.0         0.0                 0.0   \n",
       "Karamoko Is.                       0.0         0.0                 0.0   \n",
       "Correia                            0.0         0.0                 0.0   \n",
       "Raimondo                           0.0         0.0                 0.0   \n",
       "\n",
       "               gk_crosses_stopped_pct  gk_def_actions_outside_pen_area  \\\n",
       "name                                                                     \n",
       "Donnarumma G.                     5.2                             43.0   \n",
       "Szczesny                          3.0                             37.0   \n",
       "Handanovic                        2.9                             22.0   \n",
       "Gollini                           2.6                             39.0   \n",
       "Audero                            2.1                             34.0   \n",
       "...                               ...                              ...   \n",
       "Rabbi                             0.0                              0.0   \n",
       "Oddei                             0.0                              0.0   \n",
       "Karamoko Is.                      0.0                              0.0   \n",
       "Correia                           0.0                              0.0   \n",
       "Raimondo                          0.0                              0.0   \n",
       "\n",
       "               gk_def_actions_outside_pen_area_per90  \\\n",
       "name                                                   \n",
       "Donnarumma G.                                   1.16   \n",
       "Szczesny                                        1.24   \n",
       "Handanovic                                      0.61   \n",
       "Gollini                                         1.63   \n",
       "Audero                                          0.92   \n",
       "...                                              ...   \n",
       "Rabbi                                           0.00   \n",
       "Oddei                                           0.00   \n",
       "Karamoko Is.                                    0.00   \n",
       "Correia                                         0.00   \n",
       "Raimondo                                        0.00   \n",
       "\n",
       "               gk_avg_distance_def_actions  vote_avg  vote_std  \n",
       "name                                                            \n",
       "Donnarumma G.                         14.8  6.283784  0.394909  \n",
       "Szczesny                              16.2  6.216667  0.441273  \n",
       "Handanovic                            14.9  6.027027  0.505990  \n",
       "Gollini                               17.4  6.100000  0.632456  \n",
       "Audero                                14.8  6.378378  0.409439  \n",
       "...                                    ...       ...       ...  \n",
       "Rabbi                                  0.0  6.027304  0.472755  \n",
       "Oddei                                  0.0  6.207968  0.447162  \n",
       "Karamoko Is.                           0.0  6.186813  0.444240  \n",
       "Correia                                0.0  6.600386  0.396770  \n",
       "Raimondo                               0.0  6.209072  0.275318  \n",
       "\n",
       "[564 rows x 169 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rx = pd.read_excel('mid_outputs/season' + season + '/players_stats.xlsx', index_col = 3)   \n",
    "players = pd.DataFrame(rx)\n",
    "\n",
    "players\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "f9f7f88c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def player_match_data(player, pteam, oppteam):\n",
    "    if(not(player in players.index)):\n",
    "        return None\n",
    "    \n",
    "    pdata = players.loc[[player]]\n",
    "    \n",
    "    pteam_stats = team_data.loc[[pteam]].rename(index = {pteam : player})\n",
    "    \n",
    "    oppteam_stats = team_data.loc[[oppteam]].rename(index = {oppteam : player})\n",
    "    \n",
    "    oppteam_stats = oppteam_stats.rename(lambda x: 'opp_' + x, axis='columns')\n",
    "    \n",
    "    out = pd.concat([pdata, pteam_stats, oppteam_stats], axis = 1)\n",
    "    \n",
    "    return(out)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "7eb4e666",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>r</th>\n",
       "      <th>team</th>\n",
       "      <th>surname</th>\n",
       "      <th>initial</th>\n",
       "      <th>fb_ID</th>\n",
       "      <th>age</th>\n",
       "      <th>birth_year</th>\n",
       "      <th>games</th>\n",
       "      <th>...</th>\n",
       "      <th>opp_vs_team_fouls</th>\n",
       "      <th>opp_vs_team_fouled</th>\n",
       "      <th>opp_vs_team_offsides</th>\n",
       "      <th>opp_vs_team_pens_won</th>\n",
       "      <th>opp_vs_team_pens_conceded</th>\n",
       "      <th>opp_vs_team_own_goals</th>\n",
       "      <th>opp_vs_team_ball_recoveries</th>\n",
       "      <th>opp_vs_team_aerials_won</th>\n",
       "      <th>opp_vs_team_aerials_lost</th>\n",
       "      <th>opp_vs_team_aerials_won_pct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Immobile</th>\n",
       "      <td>452</td>\n",
       "      <td>785</td>\n",
       "      <td>A</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>Immobile</td>\n",
       "      <td>NaN</td>\n",
       "      <td>250</td>\n",
       "      <td>30</td>\n",
       "      <td>1990</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>504.0</td>\n",
       "      <td>548.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2251.0</td>\n",
       "      <td>506.0</td>\n",
       "      <td>633.0</td>\n",
       "      <td>44.4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 811 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          Unnamed: 0   id  r   team   surname initial  fb_ID  age  birth_year  \\\n",
       "Immobile         452  785  A  Lazio  Immobile     NaN    250   30        1990   \n",
       "\n",
       "          games  ...  opp_vs_team_fouls  opp_vs_team_fouled  \\\n",
       "Immobile     35  ...              504.0               548.0   \n",
       "\n",
       "          opp_vs_team_offsides  opp_vs_team_pens_won  \\\n",
       "Immobile                  40.0                   5.0   \n",
       "\n",
       "          opp_vs_team_pens_conceded  opp_vs_team_own_goals  \\\n",
       "Immobile                        7.0                    0.0   \n",
       "\n",
       "          opp_vs_team_ball_recoveries  opp_vs_team_aerials_won  \\\n",
       "Immobile                       2251.0                    506.0   \n",
       "\n",
       "          opp_vs_team_aerials_lost  opp_vs_team_aerials_won_pct  \n",
       "Immobile                     633.0                         44.4  \n",
       "\n",
       "[1 rows x 811 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test = player_match_data('Immobile', 'Lazio', 'Atalanta')\n",
    "test.to_excel('tmp/playermatchdata.xlsx')\n",
    "\n",
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "6b4c1b00",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Lazio'"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pteam = 'Lazio'\n",
    "pteam_stats = team_data.loc[[pteam]]\n",
    "pteam_stats.index[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "6ea6df1f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['r',\n",
       " 'fb_ID',\n",
       " 'games',\n",
       " 'games_starts',\n",
       " 'minutes',\n",
       " 'goals',\n",
       " 'assists',\n",
       " 'cards_yellow',\n",
       " 'cards_red',\n",
       " 'xg',\n",
       " 'npxg',\n",
       " 'shots_on_target',\n",
       " 'shots_on_target_pct',\n",
       " 'goals_per_shot',\n",
       " 'goals_per_shot_on_target',\n",
       " 'passes_completed',\n",
       " 'passes_pct',\n",
       " 'passes_into_final_third',\n",
       " 'passes_into_penalty_area',\n",
       " 'progressive_passes',\n",
       " 'passes_live',\n",
       " 'passes_dead',\n",
       " 'through_balls',\n",
       " 'passes_switches',\n",
       " 'crosses',\n",
       " 'corner_kicks',\n",
       " 'dribble_tackles',\n",
       " 'dribbles_vs',\n",
       " 'dribble_tackles_pct',\n",
       " 'dribbled_past',\n",
       " 'blocks',\n",
       " 'blocked_shots',\n",
       " 'blocked_passes',\n",
       " 'interceptions',\n",
       " 'clearances',\n",
       " 'errors',\n",
       " 'touches',\n",
       " 'touches_def_pen_area',\n",
       " 'touches_def_3rd',\n",
       " 'touches_mid_3rd',\n",
       " 'touches_att_3rd',\n",
       " 'touches_att_pen_area',\n",
       " 'touches_live_ball',\n",
       " 'dribbles_completed',\n",
       " 'dribbles',\n",
       " 'dribbles_completed_pct',\n",
       " 'carries',\n",
       " 'progressive_carries',\n",
       " 'carries_into_final_third',\n",
       " 'carries_into_penalty_area',\n",
       " 'passes_received',\n",
       " 'miscontrols',\n",
       " 'dispossessed',\n",
       " 'fouls',\n",
       " 'fouled',\n",
       " 'aerials_won',\n",
       " 'aerials_lost',\n",
       " 'aerials_won_pct',\n",
       " 'team_possession',\n",
       " 'team_goals_assists_per90',\n",
       " 'team_goals_pens_per90',\n",
       " 'team_goals_assists_pens_per90',\n",
       " 'team_xg_per90',\n",
       " 'team_gk_goals_against_per90',\n",
       " 'team_gk_save_pct',\n",
       " 'team_gk_clean_sheets_pct',\n",
       " 'team_passes_pct',\n",
       " 'team_passes_pct_medium',\n",
       " 'team_passes_pct_long',\n",
       " 'team_sca_per90',\n",
       " 'team_gca_per90',\n",
       " 'team_dribble_tackles_pct',\n",
       " 'team_aerials_won_pct',\n",
       " 'vs_team_possession',\n",
       " 'vs_team_goals_per90',\n",
       " 'vs_team_assists_per90',\n",
       " 'vs_team_xg_per90',\n",
       " 'vs_team_gk_save_pct',\n",
       " 'vs_team_gk_clean_sheets_pct',\n",
       " 'vs_team_gk_pct_passes_launched',\n",
       " 'vs_team_gk_crosses_stopped_pct',\n",
       " 'vs_team_shots_on_target_per90',\n",
       " 'vs_team_passes_pct',\n",
       " 'vs_team_passes_pct_short',\n",
       " 'vs_team_passes_pct_medium',\n",
       " 'vs_team_passes_pct_long',\n",
       " 'vs_team_sca_per90',\n",
       " 'vs_team_gca_per90',\n",
       " 'vs_team_dribble_tackles_pct',\n",
       " 'vs_team_dribbles_completed_pct',\n",
       " 'vs_team_aerials_won_pct']"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rx = pd.read_excel('tmp/playermatchdata_sel.xlsx', index_col = 0, header = None)   \n",
    "\n",
    "out = list()\n",
    "\n",
    "for i in range(1, rx.shape[1]):\n",
    "        if(rx[i][2] == \"x\"):\n",
    "            out.append(rx[i][0])\n",
    "            \n",
    "out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "47bc651f",
   "metadata": {},
   "outputs": [],
   "source": [
    "features_abs = ['r',\n",
    " 'games',\n",
    " 'games_starts',                \n",
    " 'minutes',\n",
    " 'shots_on_target_pct',\n",
    " 'goals_per_shot',\n",
    " 'goals_per_shot_on_target',\n",
    " 'passes_pct',\n",
    " 'dribble_tackles_pct',\n",
    " 'dribbles_completed_pct',\n",
    " 'aerials_won_pct',\n",
    " 'team_possession',\n",
    " 'team_goals_assists_per90',\n",
    " 'team_goals_pens_per90',\n",
    " 'team_goals_assists_pens_per90',\n",
    " 'team_xg_per90',\n",
    " 'team_gk_goals_against_per90',\n",
    " 'team_gk_save_pct',\n",
    " 'team_gk_clean_sheets_pct',\n",
    " 'team_passes_pct',\n",
    " 'team_passes_pct_medium',\n",
    " 'team_passes_pct_long',\n",
    " 'team_sca_per90',\n",
    " 'team_gca_per90',\n",
    " 'team_dribble_tackles_pct',\n",
    " 'team_aerials_won_pct',\n",
    " 'vs_team_possession',\n",
    " 'vs_team_goals_per90',\n",
    " 'vs_team_assists_per90',\n",
    " 'vs_team_xg_per90',\n",
    " 'vs_team_gk_save_pct',\n",
    " 'vs_team_gk_clean_sheets_pct',\n",
    " 'vs_team_gk_pct_passes_launched',\n",
    " 'vs_team_gk_crosses_stopped_pct',\n",
    " 'vs_team_shots_on_target_per90',\n",
    " 'vs_team_passes_pct',\n",
    " 'vs_team_passes_pct_short',\n",
    " 'vs_team_passes_pct_medium',\n",
    " 'vs_team_passes_pct_long',\n",
    " 'vs_team_sca_per90',\n",
    " 'vs_team_gca_per90',\n",
    " 'vs_team_dribble_tackles_pct',\n",
    " 'vs_team_dribbles_completed_pct',\n",
    " 'vs_team_aerials_won_pct',\n",
    " 'opp_team_possession',\n",
    " 'opp_team_goals_assists_per90',\n",
    " 'opp_team_goals_pens_per90',\n",
    " 'opp_team_goals_assists_pens_per90',\n",
    " 'opp_team_xg_per90',\n",
    " 'opp_team_gk_goals_against_per90',\n",
    " 'opp_team_gk_save_pct',\n",
    " 'opp_team_gk_clean_sheets_pct',\n",
    " 'opp_team_passes_pct',\n",
    " 'opp_team_passes_pct_medium',\n",
    " 'opp_team_passes_pct_long',\n",
    " 'opp_team_sca_per90',\n",
    " 'opp_team_gca_per90',\n",
    " 'opp_team_dribble_tackles_pct',\n",
    " 'opp_team_aerials_won_pct',\n",
    " 'opp_vs_team_possession',\n",
    " 'opp_vs_team_goals_per90',\n",
    " 'opp_vs_team_assists_per90',\n",
    " 'opp_vs_team_xg_per90',\n",
    " 'opp_vs_team_gk_save_pct',\n",
    " 'opp_vs_team_gk_clean_sheets_pct',\n",
    " 'opp_vs_team_gk_pct_passes_launched',\n",
    " 'opp_vs_team_gk_crosses_stopped_pct',\n",
    " 'opp_vs_team_shots_on_target_per90',\n",
    " 'opp_vs_team_passes_pct',\n",
    " 'opp_vs_team_passes_pct_short',\n",
    " 'opp_vs_team_passes_pct_medium',\n",
    " 'opp_vs_team_passes_pct_long',\n",
    " 'opp_vs_team_sca_per90',\n",
    " 'opp_vs_team_gca_per90',\n",
    " 'opp_vs_team_dribble_tackles_pct',\n",
    " 'opp_vs_team_dribbles_completed_pct',\n",
    " 'opp_vs_team_aerials_won_pct',\n",
    "                \n",
    " 'vote_avg',\n",
    " 'vote_std']\n",
    "\n",
    "features_rel = [\n",
    " 'goals',\n",
    " 'assists',\n",
    " 'cards_yellow',\n",
    " 'cards_red',\n",
    " 'xg',\n",
    " 'npxg',\n",
    " 'shots_on_target',\n",
    " 'passes_completed',\n",
    " 'passes_into_final_third',\n",
    " 'passes_into_penalty_area',\n",
    " 'progressive_passes',\n",
    " 'passes_live',\n",
    " 'passes_dead',\n",
    " 'through_balls',\n",
    " 'passes_switches',\n",
    " 'crosses',\n",
    " 'corner_kicks',\n",
    " 'dribble_tackles',\n",
    " 'dribbles_vs',\n",
    " 'dribbled_past',\n",
    " 'blocks',\n",
    " 'blocked_shots',\n",
    " 'blocked_passes',\n",
    " 'interceptions',\n",
    " 'clearances',\n",
    " 'errors',\n",
    " 'touches',\n",
    " 'touches_def_pen_area',\n",
    " 'touches_def_3rd',\n",
    " 'touches_mid_3rd',\n",
    " 'touches_att_3rd',\n",
    " 'touches_att_pen_area',\n",
    " 'touches_live_ball',\n",
    " 'dribbles_completed',\n",
    " 'dribbles',\n",
    " 'passes_received',\n",
    " 'miscontrols',\n",
    " 'dispossessed',\n",
    " 'fouls',\n",
    " 'fouled',\n",
    " 'aerials_won',\n",
    " 'aerials_lost',\n",
    " 'carries',\n",
    " 'progressive_carries',\n",
    " 'carries_into_final_third',\n",
    " 'carries_into_penalty_area']\n",
    "\n",
    "features_rel_gamecorr = [\n",
    "    'goals',\n",
    "    'assists',\n",
    "    'xg',\n",
    "    'npxg',\n",
    "    'cards_yellow',\n",
    "    'cards_red'\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "3f43d509",
   "metadata": {},
   "outputs": [],
   "source": [
    "def player_match_data_ext(player, pteam, oppteam):\n",
    "    pdata = player_match_data(player, pteam, oppteam)\n",
    "    \n",
    "    if(not isinstance(pdata, pd.DataFrame)):\n",
    "        return None\n",
    "    \n",
    "    out = pd.concat([pdata[features_abs], pdata[features_rel]], axis = 1)\n",
    "    \n",
    "    out[features_rel] = out[features_rel] / max(pdata['minutes'][0], 1)\n",
    "    \n",
    "    out[features_rel_gamecorr] = out[features_rel_gamecorr] * (pdata['minutes'][0] / max(pdata['games'][0], 1) / 90)\n",
    "    \n",
    "    return out\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "930d00c7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>r</th>\n",
       "      <th>games</th>\n",
       "      <th>games_starts</th>\n",
       "      <th>minutes</th>\n",
       "      <th>shots_on_target_pct</th>\n",
       "      <th>goals_per_shot</th>\n",
       "      <th>goals_per_shot_on_target</th>\n",
       "      <th>passes_pct</th>\n",
       "      <th>dribble_tackles_pct</th>\n",
       "      <th>dribbles_completed_pct</th>\n",
       "      <th>...</th>\n",
       "      <th>miscontrols</th>\n",
       "      <th>dispossessed</th>\n",
       "      <th>fouls</th>\n",
       "      <th>fouled</th>\n",
       "      <th>aerials_won</th>\n",
       "      <th>aerials_lost</th>\n",
       "      <th>carries</th>\n",
       "      <th>progressive_carries</th>\n",
       "      <th>carries_into_final_third</th>\n",
       "      <th>carries_into_penalty_area</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Immobile</th>\n",
       "      <td>A</td>\n",
       "      <td>35</td>\n",
       "      <td>34</td>\n",
       "      <td>2849</td>\n",
       "      <td>39.3</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.36</td>\n",
       "      <td>72.6</td>\n",
       "      <td>31.8</td>\n",
       "      <td>57.4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.024921</td>\n",
       "      <td>0.011934</td>\n",
       "      <td>0.012987</td>\n",
       "      <td>0.015444</td>\n",
       "      <td>0.009126</td>\n",
       "      <td>0.018954</td>\n",
       "      <td>0.2457</td>\n",
       "      <td>0.025974</td>\n",
       "      <td>0.015093</td>\n",
       "      <td>0.013338</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 125 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          r  games  games_starts  minutes  shots_on_target_pct  \\\n",
       "Immobile  A     35            34     2849                 39.3   \n",
       "\n",
       "          goals_per_shot  goals_per_shot_on_target  passes_pct  \\\n",
       "Immobile            0.14                      0.36        72.6   \n",
       "\n",
       "          dribble_tackles_pct  dribbles_completed_pct  ...  miscontrols  \\\n",
       "Immobile                 31.8                    57.4  ...     0.024921   \n",
       "\n",
       "          dispossessed     fouls    fouled  aerials_won  aerials_lost  \\\n",
       "Immobile      0.011934  0.012987  0.015444     0.009126      0.018954   \n",
       "\n",
       "          carries  progressive_carries  carries_into_final_third  \\\n",
       "Immobile   0.2457             0.025974                  0.015093   \n",
       "\n",
       "          carries_into_penalty_area  \n",
       "Immobile                   0.013338  \n",
       "\n",
       "[1 rows x 125 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test = player_match_data_ext('Immobile', 'Lazio', 'Atalanta')\n",
    "\n",
    "add_columns = test.columns\n",
    "\n",
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "3ae718d2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>matchday</th>\n",
       "      <th>player</th>\n",
       "      <th>team</th>\n",
       "      <th>oppteam</th>\n",
       "      <th>home</th>\n",
       "      <th>vote</th>\n",
       "      <th>goals</th>\n",
       "      <th>assists</th>\n",
       "      <th>cards_malus</th>\n",
       "      <th>fantavote</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Sportiello</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Toloi</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>Djimsiti</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Hateboer</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>7.5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Gosens</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>7.5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10569</th>\n",
       "      <td>38</td>\n",
       "      <td>Bessa</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10570</th>\n",
       "      <td>38</td>\n",
       "      <td>Zaccagni</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10571</th>\n",
       "      <td>38</td>\n",
       "      <td>Ilic</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10572</th>\n",
       "      <td>38</td>\n",
       "      <td>Lasagna</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10573</th>\n",
       "      <td>38</td>\n",
       "      <td>Kalinic</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10574 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       matchday      player      team oppteam  home  vote  goals  assists  \\\n",
       "0             1  Sportiello  Atalanta   Lazio     0   6.5     -1        0   \n",
       "1             1       Toloi  Atalanta   Lazio     0   6.0      0        0   \n",
       "2             1    Djimsiti  Atalanta   Lazio     0   6.5      0        0   \n",
       "3             1    Hateboer  Atalanta   Lazio     0   7.5      1        0   \n",
       "4             1      Gosens  Atalanta   Lazio     0   7.5      1        1   \n",
       "...         ...         ...       ...     ...   ...   ...    ...      ...   \n",
       "10569        38       Bessa    Verona  Napoli     0   6.0      0        0   \n",
       "10570        38    Zaccagni    Verona  Napoli     0   6.5      0        0   \n",
       "10571        38        Ilic    Verona  Napoli     0   6.5      0        0   \n",
       "10572        38     Lasagna    Verona  Napoli     0   5.5      0        0   \n",
       "10573        38     Kalinic    Verona  Napoli     0   5.5      0        0   \n",
       "\n",
       "       cards_malus  fantavote  \n",
       "0              0.0        5.5  \n",
       "1              0.5        5.5  \n",
       "2              0.5        6.0  \n",
       "3              0.0       10.5  \n",
       "4              0.0       11.5  \n",
       "...            ...        ...  \n",
       "10569          0.0        6.0  \n",
       "10570          0.0        6.5  \n",
       "10571          0.5        6.0  \n",
       "10572          0.0        5.5  \n",
       "10573          0.0        5.5  \n",
       "\n",
       "[10574 rows x 10 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "votes = pd.read_excel('mid_outputs/season' + season + '/players_votes.xlsx', index_col = 0)\n",
    "votes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "ec0bc56b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "100\n",
      "200\n",
      "300\n",
      "400\n",
      "500\n",
      "700\n",
      "800\n",
      "900\n",
      "1000\n",
      "1100\n",
      "1300\n",
      "1400\n",
      "1500\n",
      "1600\n",
      "1700\n",
      "1800\n",
      "1900\n",
      "2000\n",
      "2100\n",
      "2200\n",
      "2300\n",
      "2400\n",
      "2500\n",
      "2600\n",
      "2700\n",
      "2800\n",
      "2900\n",
      "3000\n",
      "3100\n",
      "3200\n",
      "3300\n",
      "3400\n",
      "3500\n",
      "3600\n",
      "3700\n",
      "3800\n",
      "3900\n",
      "4000\n",
      "4100\n",
      "4200\n",
      "4300\n",
      "4400\n",
      "4500\n",
      "4600\n",
      "4700\n",
      "4800\n",
      "5000\n",
      "5100\n",
      "5200\n",
      "5300\n",
      "5400\n",
      "5500\n",
      "5600\n",
      "5700\n",
      "5800\n",
      "5900\n",
      "6000\n",
      "6100\n",
      "6200\n",
      "6300\n",
      "6400\n",
      "6500\n",
      "6600\n",
      "6700\n",
      "6800\n",
      "6900\n",
      "7000\n",
      "7100\n",
      "7200\n",
      "7300\n",
      "7400\n",
      "7500\n",
      "7600\n",
      "7700\n",
      "7800\n",
      "7900\n",
      "8000\n",
      "8100\n",
      "8200\n",
      "8300\n",
      "8400\n",
      "8500\n",
      "8600\n",
      "8700\n",
      "8800\n",
      "8900\n",
      "9000\n",
      "9100\n",
      "9200\n",
      "9300\n",
      "9400\n",
      "9500\n",
      "9600\n",
      "9700\n",
      "9800\n",
      "9900\n",
      "10000\n",
      "10100\n",
      "10200\n",
      "10300\n",
      "10400\n",
      "10500\n"
     ]
    }
   ],
   "source": [
    "db = votes.copy()\n",
    "\n",
    "delta = pd.DataFrame(columns = add_columns)\n",
    "\n",
    "for i in range(db.shape[0]):\n",
    "    ext_data = player_match_data_ext(db['player'][i], db['team'][i], db['oppteam'][i])\n",
    "    \n",
    "    if(not isinstance(ext_data, pd.DataFrame)):\n",
    "        continue\n",
    "    \n",
    "    ext_data = ext_data.rename(index = {db['player'][i] : i})\n",
    "    \n",
    "    delta = pd.concat([delta, ext_data], axis = 0)\n",
    "    \n",
    "    if(i % 100 == 0):\n",
    "        print(i)\n",
    "    \n",
    "db = pd.concat([db, delta], axis = 1)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "b292f9f9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>matchday</th>\n",
       "      <th>player</th>\n",
       "      <th>team</th>\n",
       "      <th>oppteam</th>\n",
       "      <th>home</th>\n",
       "      <th>vote</th>\n",
       "      <th>goals</th>\n",
       "      <th>assists</th>\n",
       "      <th>cards_malus</th>\n",
       "      <th>fantavote</th>\n",
       "      <th>...</th>\n",
       "      <th>miscontrols</th>\n",
       "      <th>dispossessed</th>\n",
       "      <th>fouls</th>\n",
       "      <th>fouled</th>\n",
       "      <th>aerials_won</th>\n",
       "      <th>aerials_lost</th>\n",
       "      <th>carries</th>\n",
       "      <th>progressive_carries</th>\n",
       "      <th>carries_into_final_third</th>\n",
       "      <th>carries_into_penalty_area</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Sportiello</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Toloi</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>5.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.008672</td>\n",
       "      <td>0.005124</td>\n",
       "      <td>0.014978</td>\n",
       "      <td>0.005912</td>\n",
       "      <td>0.018132</td>\n",
       "      <td>0.011037</td>\n",
       "      <td>0.430430</td>\n",
       "      <td>0.015767</td>\n",
       "      <td>0.014190</td>\n",
       "      <td>0.001182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>Djimsiti</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.009524</td>\n",
       "      <td>0.003429</td>\n",
       "      <td>0.011810</td>\n",
       "      <td>0.009905</td>\n",
       "      <td>0.028952</td>\n",
       "      <td>0.011810</td>\n",
       "      <td>0.460571</td>\n",
       "      <td>0.016000</td>\n",
       "      <td>0.019810</td>\n",
       "      <td>0.001905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Hateboer</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>7.5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.015891</td>\n",
       "      <td>0.006810</td>\n",
       "      <td>0.011351</td>\n",
       "      <td>0.004540</td>\n",
       "      <td>0.013053</td>\n",
       "      <td>0.011351</td>\n",
       "      <td>0.386493</td>\n",
       "      <td>0.023269</td>\n",
       "      <td>0.017026</td>\n",
       "      <td>0.003973</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Gosens</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>7.5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.016250</td>\n",
       "      <td>0.006342</td>\n",
       "      <td>0.019818</td>\n",
       "      <td>0.010305</td>\n",
       "      <td>0.014269</td>\n",
       "      <td>0.007927</td>\n",
       "      <td>0.417757</td>\n",
       "      <td>0.028141</td>\n",
       "      <td>0.022592</td>\n",
       "      <td>0.006342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10569</th>\n",
       "      <td>38</td>\n",
       "      <td>Bessa</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.029268</td>\n",
       "      <td>0.019512</td>\n",
       "      <td>0.026016</td>\n",
       "      <td>0.009756</td>\n",
       "      <td>0.003252</td>\n",
       "      <td>0.026016</td>\n",
       "      <td>0.401626</td>\n",
       "      <td>0.030894</td>\n",
       "      <td>0.030894</td>\n",
       "      <td>0.006504</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10570</th>\n",
       "      <td>38</td>\n",
       "      <td>Zaccagni</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.028363</td>\n",
       "      <td>0.024530</td>\n",
       "      <td>0.013798</td>\n",
       "      <td>0.041778</td>\n",
       "      <td>0.018015</td>\n",
       "      <td>0.021847</td>\n",
       "      <td>0.329628</td>\n",
       "      <td>0.045228</td>\n",
       "      <td>0.026447</td>\n",
       "      <td>0.017631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10571</th>\n",
       "      <td>38</td>\n",
       "      <td>Ilic</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.014169</td>\n",
       "      <td>0.007084</td>\n",
       "      <td>0.012534</td>\n",
       "      <td>0.015804</td>\n",
       "      <td>0.019074</td>\n",
       "      <td>0.013624</td>\n",
       "      <td>0.407084</td>\n",
       "      <td>0.020708</td>\n",
       "      <td>0.015804</td>\n",
       "      <td>0.000545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10572</th>\n",
       "      <td>38</td>\n",
       "      <td>Lasagna</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.040253</td>\n",
       "      <td>0.014207</td>\n",
       "      <td>0.025257</td>\n",
       "      <td>0.016575</td>\n",
       "      <td>0.020521</td>\n",
       "      <td>0.037885</td>\n",
       "      <td>0.183899</td>\n",
       "      <td>0.022099</td>\n",
       "      <td>0.014207</td>\n",
       "      <td>0.013418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10573</th>\n",
       "      <td>38</td>\n",
       "      <td>Kalinic</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.032926</td>\n",
       "      <td>0.012230</td>\n",
       "      <td>0.025400</td>\n",
       "      <td>0.031985</td>\n",
       "      <td>0.015992</td>\n",
       "      <td>0.047977</td>\n",
       "      <td>0.199436</td>\n",
       "      <td>0.006585</td>\n",
       "      <td>0.004704</td>\n",
       "      <td>0.002822</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10574 rows × 135 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       matchday      player      team oppteam  home  vote  goals  assists  \\\n",
       "0             1  Sportiello  Atalanta   Lazio     0   6.5     -1        0   \n",
       "1             1       Toloi  Atalanta   Lazio     0   6.0      0        0   \n",
       "2             1    Djimsiti  Atalanta   Lazio     0   6.5      0        0   \n",
       "3             1    Hateboer  Atalanta   Lazio     0   7.5      1        0   \n",
       "4             1      Gosens  Atalanta   Lazio     0   7.5      1        1   \n",
       "...         ...         ...       ...     ...   ...   ...    ...      ...   \n",
       "10569        38       Bessa    Verona  Napoli     0   6.0      0        0   \n",
       "10570        38    Zaccagni    Verona  Napoli     0   6.5      0        0   \n",
       "10571        38        Ilic    Verona  Napoli     0   6.5      0        0   \n",
       "10572        38     Lasagna    Verona  Napoli     0   5.5      0        0   \n",
       "10573        38     Kalinic    Verona  Napoli     0   5.5      0        0   \n",
       "\n",
       "       cards_malus  fantavote  ... miscontrols dispossessed     fouls  \\\n",
       "0              0.0        5.5  ...    0.000000     0.000000  0.000000   \n",
       "1              0.5        5.5  ...    0.008672     0.005124  0.014978   \n",
       "2              0.5        6.0  ...    0.009524     0.003429  0.011810   \n",
       "3              0.0       10.5  ...    0.015891     0.006810  0.011351   \n",
       "4              0.0       11.5  ...    0.016250     0.006342  0.019818   \n",
       "...            ...        ...  ...         ...          ...       ...   \n",
       "10569          0.0        6.0  ...    0.029268     0.019512  0.026016   \n",
       "10570          0.0        6.5  ...    0.028363     0.024530  0.013798   \n",
       "10571          0.5        6.0  ...    0.014169     0.007084  0.012534   \n",
       "10572          0.0        5.5  ...    0.040253     0.014207  0.025257   \n",
       "10573          0.0        5.5  ...    0.032926     0.012230  0.025400   \n",
       "\n",
       "         fouled  aerials_won  aerials_lost   carries  progressive_carries  \\\n",
       "0      0.000000     0.000000      0.000000  0.000000             0.000000   \n",
       "1      0.005912     0.018132      0.011037  0.430430             0.015767   \n",
       "2      0.009905     0.028952      0.011810  0.460571             0.016000   \n",
       "3      0.004540     0.013053      0.011351  0.386493             0.023269   \n",
       "4      0.010305     0.014269      0.007927  0.417757             0.028141   \n",
       "...         ...          ...           ...       ...                  ...   \n",
       "10569  0.009756     0.003252      0.026016  0.401626             0.030894   \n",
       "10570  0.041778     0.018015      0.021847  0.329628             0.045228   \n",
       "10571  0.015804     0.019074      0.013624  0.407084             0.020708   \n",
       "10572  0.016575     0.020521      0.037885  0.183899             0.022099   \n",
       "10573  0.031985     0.015992      0.047977  0.199436             0.006585   \n",
       "\n",
       "       carries_into_final_third  carries_into_penalty_area  \n",
       "0                      0.000000                   0.000000  \n",
       "1                      0.014190                   0.001182  \n",
       "2                      0.019810                   0.001905  \n",
       "3                      0.017026                   0.003973  \n",
       "4                      0.022592                   0.006342  \n",
       "...                         ...                        ...  \n",
       "10569                  0.030894                   0.006504  \n",
       "10570                  0.026447                   0.017631  \n",
       "10571                  0.015804                   0.000545  \n",
       "10572                  0.014207                   0.013418  \n",
       "10573                  0.004704                   0.002822  \n",
       "\n",
       "[10574 rows x 135 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "db"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "1db9f77f",
   "metadata": {},
   "outputs": [],
   "source": [
    "db = db.drop(db[db.r == 'P'].index) # remove goalkeepers\n",
    "db = db.drop(db[db.r != db.r].index) # remove non extisting players\n",
    "db = db.drop(db[db.r != db.r].index) # remove non extisting players\n",
    "db = db.drop(db[db.xg != db.xg].index) \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "f56df3ca",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>matchday</th>\n",
       "      <th>player</th>\n",
       "      <th>team</th>\n",
       "      <th>oppteam</th>\n",
       "      <th>home</th>\n",
       "      <th>vote</th>\n",
       "      <th>goals</th>\n",
       "      <th>assists</th>\n",
       "      <th>cards_malus</th>\n",
       "      <th>fantavote</th>\n",
       "      <th>...</th>\n",
       "      <th>miscontrols</th>\n",
       "      <th>dispossessed</th>\n",
       "      <th>fouls</th>\n",
       "      <th>fouled</th>\n",
       "      <th>aerials_won</th>\n",
       "      <th>aerials_lost</th>\n",
       "      <th>carries</th>\n",
       "      <th>progressive_carries</th>\n",
       "      <th>carries_into_final_third</th>\n",
       "      <th>carries_into_penalty_area</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Toloi</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>5.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.008672</td>\n",
       "      <td>0.005124</td>\n",
       "      <td>0.014978</td>\n",
       "      <td>0.005912</td>\n",
       "      <td>0.018132</td>\n",
       "      <td>0.011037</td>\n",
       "      <td>0.430430</td>\n",
       "      <td>0.015767</td>\n",
       "      <td>0.014190</td>\n",
       "      <td>0.001182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>Djimsiti</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.009524</td>\n",
       "      <td>0.003429</td>\n",
       "      <td>0.011810</td>\n",
       "      <td>0.009905</td>\n",
       "      <td>0.028952</td>\n",
       "      <td>0.011810</td>\n",
       "      <td>0.460571</td>\n",
       "      <td>0.016000</td>\n",
       "      <td>0.019810</td>\n",
       "      <td>0.001905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Hateboer</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>7.5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.015891</td>\n",
       "      <td>0.006810</td>\n",
       "      <td>0.011351</td>\n",
       "      <td>0.004540</td>\n",
       "      <td>0.013053</td>\n",
       "      <td>0.011351</td>\n",
       "      <td>0.386493</td>\n",
       "      <td>0.023269</td>\n",
       "      <td>0.017026</td>\n",
       "      <td>0.003973</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Gosens</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>7.5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.016250</td>\n",
       "      <td>0.006342</td>\n",
       "      <td>0.019818</td>\n",
       "      <td>0.010305</td>\n",
       "      <td>0.014269</td>\n",
       "      <td>0.007927</td>\n",
       "      <td>0.417757</td>\n",
       "      <td>0.028141</td>\n",
       "      <td>0.022592</td>\n",
       "      <td>0.006342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>Palomino</td>\n",
       "      <td>Atalanta</td>\n",
       "      <td>Lazio</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.005354</td>\n",
       "      <td>0.002059</td>\n",
       "      <td>0.019769</td>\n",
       "      <td>0.006590</td>\n",
       "      <td>0.024300</td>\n",
       "      <td>0.018534</td>\n",
       "      <td>0.404448</td>\n",
       "      <td>0.006590</td>\n",
       "      <td>0.004942</td>\n",
       "      <td>0.000824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10569</th>\n",
       "      <td>38</td>\n",
       "      <td>Bessa</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.029268</td>\n",
       "      <td>0.019512</td>\n",
       "      <td>0.026016</td>\n",
       "      <td>0.009756</td>\n",
       "      <td>0.003252</td>\n",
       "      <td>0.026016</td>\n",
       "      <td>0.401626</td>\n",
       "      <td>0.030894</td>\n",
       "      <td>0.030894</td>\n",
       "      <td>0.006504</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10570</th>\n",
       "      <td>38</td>\n",
       "      <td>Zaccagni</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.028363</td>\n",
       "      <td>0.024530</td>\n",
       "      <td>0.013798</td>\n",
       "      <td>0.041778</td>\n",
       "      <td>0.018015</td>\n",
       "      <td>0.021847</td>\n",
       "      <td>0.329628</td>\n",
       "      <td>0.045228</td>\n",
       "      <td>0.026447</td>\n",
       "      <td>0.017631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10571</th>\n",
       "      <td>38</td>\n",
       "      <td>Ilic</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>6.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.014169</td>\n",
       "      <td>0.007084</td>\n",
       "      <td>0.012534</td>\n",
       "      <td>0.015804</td>\n",
       "      <td>0.019074</td>\n",
       "      <td>0.013624</td>\n",
       "      <td>0.407084</td>\n",
       "      <td>0.020708</td>\n",
       "      <td>0.015804</td>\n",
       "      <td>0.000545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10572</th>\n",
       "      <td>38</td>\n",
       "      <td>Lasagna</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.040253</td>\n",
       "      <td>0.014207</td>\n",
       "      <td>0.025257</td>\n",
       "      <td>0.016575</td>\n",
       "      <td>0.020521</td>\n",
       "      <td>0.037885</td>\n",
       "      <td>0.183899</td>\n",
       "      <td>0.022099</td>\n",
       "      <td>0.014207</td>\n",
       "      <td>0.013418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10573</th>\n",
       "      <td>38</td>\n",
       "      <td>Kalinic</td>\n",
       "      <td>Verona</td>\n",
       "      <td>Napoli</td>\n",
       "      <td>0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.032926</td>\n",
       "      <td>0.012230</td>\n",
       "      <td>0.025400</td>\n",
       "      <td>0.031985</td>\n",
       "      <td>0.015992</td>\n",
       "      <td>0.047977</td>\n",
       "      <td>0.199436</td>\n",
       "      <td>0.006585</td>\n",
       "      <td>0.004704</td>\n",
       "      <td>0.002822</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>9663 rows × 135 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       matchday    player      team oppteam  home  vote  goals  assists  \\\n",
       "1             1     Toloi  Atalanta   Lazio     0   6.0      0        0   \n",
       "2             1  Djimsiti  Atalanta   Lazio     0   6.5      0        0   \n",
       "3             1  Hateboer  Atalanta   Lazio     0   7.5      1        0   \n",
       "4             1    Gosens  Atalanta   Lazio     0   7.5      1        1   \n",
       "5             1  Palomino  Atalanta   Lazio     0   6.5      0        0   \n",
       "...         ...       ...       ...     ...   ...   ...    ...      ...   \n",
       "10569        38     Bessa    Verona  Napoli     0   6.0      0        0   \n",
       "10570        38  Zaccagni    Verona  Napoli     0   6.5      0        0   \n",
       "10571        38      Ilic    Verona  Napoli     0   6.5      0        0   \n",
       "10572        38   Lasagna    Verona  Napoli     0   5.5      0        0   \n",
       "10573        38   Kalinic    Verona  Napoli     0   5.5      0        0   \n",
       "\n",
       "       cards_malus  fantavote  ... miscontrols dispossessed     fouls  \\\n",
       "1              0.5        5.5  ...    0.008672     0.005124  0.014978   \n",
       "2              0.5        6.0  ...    0.009524     0.003429  0.011810   \n",
       "3              0.0       10.5  ...    0.015891     0.006810  0.011351   \n",
       "4              0.0       11.5  ...    0.016250     0.006342  0.019818   \n",
       "5              0.0        6.5  ...    0.005354     0.002059  0.019769   \n",
       "...            ...        ...  ...         ...          ...       ...   \n",
       "10569          0.0        6.0  ...    0.029268     0.019512  0.026016   \n",
       "10570          0.0        6.5  ...    0.028363     0.024530  0.013798   \n",
       "10571          0.5        6.0  ...    0.014169     0.007084  0.012534   \n",
       "10572          0.0        5.5  ...    0.040253     0.014207  0.025257   \n",
       "10573          0.0        5.5  ...    0.032926     0.012230  0.025400   \n",
       "\n",
       "         fouled  aerials_won  aerials_lost   carries  progressive_carries  \\\n",
       "1      0.005912     0.018132      0.011037  0.430430             0.015767   \n",
       "2      0.009905     0.028952      0.011810  0.460571             0.016000   \n",
       "3      0.004540     0.013053      0.011351  0.386493             0.023269   \n",
       "4      0.010305     0.014269      0.007927  0.417757             0.028141   \n",
       "5      0.006590     0.024300      0.018534  0.404448             0.006590   \n",
       "...         ...          ...           ...       ...                  ...   \n",
       "10569  0.009756     0.003252      0.026016  0.401626             0.030894   \n",
       "10570  0.041778     0.018015      0.021847  0.329628             0.045228   \n",
       "10571  0.015804     0.019074      0.013624  0.407084             0.020708   \n",
       "10572  0.016575     0.020521      0.037885  0.183899             0.022099   \n",
       "10573  0.031985     0.015992      0.047977  0.199436             0.006585   \n",
       "\n",
       "       carries_into_final_third  carries_into_penalty_area  \n",
       "1                      0.014190                   0.001182  \n",
       "2                      0.019810                   0.001905  \n",
       "3                      0.017026                   0.003973  \n",
       "4                      0.022592                   0.006342  \n",
       "5                      0.004942                   0.000824  \n",
       "...                         ...                        ...  \n",
       "10569                  0.030894                   0.006504  \n",
       "10570                  0.026447                   0.017631  \n",
       "10571                  0.015804                   0.000545  \n",
       "10572                  0.014207                   0.013418  \n",
       "10573                  0.004704                   0.002822  \n",
       "\n",
       "[9663 rows x 135 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "db"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "df26c8e0",
   "metadata": {},
   "outputs": [],
   "source": [
    "db.to_excel('mid_outputs/season' + season + '/database_entries.xlsx')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "626ba83b",
   "metadata": {},
   "source": [
    "Goalkeepers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "ba7b838c",
   "metadata": {},
   "outputs": [],
   "source": [
    "features_abs_gk = [\n",
    " 'gk_games',\n",
    " 'gk_games_starts',\n",
    " 'gk_minutes',\n",
    " 'gk_goals_against_per90', \n",
    " 'gk_save_pct',\n",
    " 'gk_clean_sheets_pct',\n",
    " 'gk_psxg_net_per90',\n",
    " 'gk_passes_pct_launched',\n",
    " 'gk_pct_passes_launched',\n",
    " 'gk_passes_length_avg',\n",
    " 'gk_pct_goal_kicks_launched',\n",
    " 'gk_goal_kick_length_avg',\n",
    " 'gk_crosses_stopped_pct',\n",
    " 'gk_def_actions_outside_pen_area_per90',\n",
    " 'gk_avg_distance_def_actions',\n",
    "    \n",
    " 'team_possession',\n",
    " 'team_goals_assists_per90',\n",
    " 'team_goals_pens_per90',\n",
    " 'team_goals_assists_pens_per90',\n",
    " 'team_xg_per90',\n",
    " 'team_gk_goals_against_per90',\n",
    " 'team_gk_save_pct',\n",
    " 'team_gk_clean_sheets_pct',\n",
    " 'team_passes_pct',\n",
    " 'team_passes_pct_medium',\n",
    " 'team_passes_pct_long',\n",
    " 'team_sca_per90',\n",
    " 'team_gca_per90',\n",
    " 'team_dribble_tackles_pct',\n",
    " 'team_aerials_won_pct',\n",
    " 'vs_team_possession',\n",
    " 'vs_team_goals_per90',\n",
    " 'vs_team_assists_per90',\n",
    " 'vs_team_xg_per90',\n",
    " 'vs_team_gk_save_pct',\n",
    " 'vs_team_gk_clean_sheets_pct',\n",
    " 'vs_team_gk_pct_passes_launched',\n",
    " 'vs_team_gk_crosses_stopped_pct',\n",
    " 'vs_team_shots_on_target_per90',\n",
    " 'vs_team_passes_pct',\n",
    " 'vs_team_passes_pct_short',\n",
    " 'vs_team_passes_pct_medium',\n",
    " 'vs_team_passes_pct_long',\n",
    " 'vs_team_sca_per90',\n",
    " 'vs_team_gca_per90',\n",
    " 'vs_team_dribble_tackles_pct',\n",
    " 'vs_team_dribbles_completed_pct',\n",
    " 'vs_team_aerials_won_pct',\n",
    " 'opp_team_possession',\n",
    " 'opp_team_goals_assists_per90',\n",
    " 'opp_team_goals_pens_per90',\n",
    " 'opp_team_goals_assists_pens_per90',\n",
    " 'opp_team_xg_per90',\n",
    " 'opp_team_gk_goals_against_per90',\n",
    " 'opp_team_gk_save_pct',\n",
    " 'opp_team_gk_clean_sheets_pct',\n",
    " 'opp_team_passes_pct',\n",
    " 'opp_team_passes_pct_medium',\n",
    " 'opp_team_passes_pct_long',\n",
    " 'opp_team_sca_per90',\n",
    " 'opp_team_gca_per90',\n",
    " 'opp_team_dribble_tackles_pct',\n",
    " 'opp_team_aerials_won_pct',\n",
    " 'opp_vs_team_possession',\n",
    " 'opp_vs_team_goals_per90',\n",
    " 'opp_vs_team_assists_per90',\n",
    " 'opp_vs_team_xg_per90',\n",
    " 'opp_vs_team_gk_save_pct',\n",
    " 'opp_vs_team_gk_clean_sheets_pct',\n",
    " 'opp_vs_team_gk_pct_passes_launched',\n",
    " 'opp_vs_team_gk_crosses_stopped_pct',\n",
    " 'opp_vs_team_shots_on_target_per90',\n",
    " 'opp_vs_team_passes_pct',\n",
    " 'opp_vs_team_passes_pct_short',\n",
    " 'opp_vs_team_passes_pct_medium',\n",
    " 'opp_vs_team_passes_pct_long',\n",
    " 'opp_vs_team_sca_per90',\n",
    " 'opp_vs_team_gca_per90',\n",
    " 'opp_vs_team_dribble_tackles_pct',\n",
    " 'opp_vs_team_dribbles_completed_pct',\n",
    " 'opp_vs_team_aerials_won_pct',\n",
    "                \n",
    " 'vote_avg',\n",
    " 'vote_std']\n",
    "\n",
    "features_rel_gk = [\n",
    " 'gk_shots_on_target_against',\n",
    " 'gk_saves',\n",
    " 'gk_free_kick_goals_against',\n",
    " 'gk_corner_kick_goals_against',\n",
    " 'gk_own_goals_against',\n",
    " 'gk_psxg',\n",
    " 'gk_psnpxg_per_shot_on_target_against',\n",
    " 'gk_psxg_net',\n",
    " 'gk_passes_completed_launched',\n",
    " 'gk_passes_launched',\n",
    " 'gk_passes',\n",
    " 'gk_passes_throws',\n",
    " 'gk_goal_kicks',\n",
    " 'gk_crosses',\n",
    " 'gk_crosses_stopped',\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "1d2682b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "def player_match_data_ext_gk(player, pteam, oppteam):\n",
    "    pdata = player_match_data(player, pteam, oppteam)\n",
    "    \n",
    "    if(not isinstance(pdata, pd.DataFrame)):\n",
    "        return None\n",
    "    \n",
    "    if(pdata['gk_games'][0] <= 0):\n",
    "        return None\n",
    "    \n",
    "    out = pd.concat([pdata[features_abs_gk], pdata[features_rel_gk]], axis = 1)\n",
    "    \n",
    "    out[features_rel_gk] = out[features_rel_gk] / max(pdata['minutes'][0], 1)\n",
    "\n",
    "    return out\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "bc6e9d52",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gk_games</th>\n",
       "      <th>gk_games_starts</th>\n",
       "      <th>gk_minutes</th>\n",
       "      <th>gk_goals_against_per90</th>\n",
       "      <th>gk_save_pct</th>\n",
       "      <th>gk_clean_sheets_pct</th>\n",
       "      <th>gk_psxg_net_per90</th>\n",
       "      <th>gk_passes_pct_launched</th>\n",
       "      <th>gk_pct_passes_launched</th>\n",
       "      <th>gk_passes_length_avg</th>\n",
       "      <th>...</th>\n",
       "      <th>gk_psxg</th>\n",
       "      <th>gk_psnpxg_per_shot_on_target_against</th>\n",
       "      <th>gk_psxg_net</th>\n",
       "      <th>gk_passes_completed_launched</th>\n",
       "      <th>gk_passes_launched</th>\n",
       "      <th>gk_passes</th>\n",
       "      <th>gk_passes_throws</th>\n",
       "      <th>gk_goal_kicks</th>\n",
       "      <th>gk_crosses</th>\n",
       "      <th>gk_crosses_stopped</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Consigli</th>\n",
       "      <td>37.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>3322.0</td>\n",
       "      <td>1.71</td>\n",
       "      <td>67.8</td>\n",
       "      <td>8.1</td>\n",
       "      <td>-0.16</td>\n",
       "      <td>43.2</td>\n",
       "      <td>29.0</td>\n",
       "      <td>30.9</td>\n",
       "      <td>...</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0.28</td>\n",
       "      <td>-6.0</td>\n",
       "      <td>186.0</td>\n",
       "      <td>431.0</td>\n",
       "      <td>1219.0</td>\n",
       "      <td>174.0</td>\n",
       "      <td>246.0</td>\n",
       "      <td>491.0</td>\n",
       "      <td>21.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 98 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          gk_games  gk_games_starts  gk_minutes  gk_goals_against_per90  \\\n",
       "Consigli      37.0             37.0      3322.0                    1.71   \n",
       "\n",
       "          gk_save_pct  gk_clean_sheets_pct  gk_psxg_net_per90  \\\n",
       "Consigli         67.8                  8.1              -0.16   \n",
       "\n",
       "          gk_passes_pct_launched  gk_pct_passes_launched  \\\n",
       "Consigli                    43.2                    29.0   \n",
       "\n",
       "          gk_passes_length_avg  ...  gk_psxg  \\\n",
       "Consigli                  30.9  ...     56.0   \n",
       "\n",
       "          gk_psnpxg_per_shot_on_target_against  gk_psxg_net  \\\n",
       "Consigli                                  0.28         -6.0   \n",
       "\n",
       "          gk_passes_completed_launched  gk_passes_launched  gk_passes  \\\n",
       "Consigli                         186.0               431.0     1219.0   \n",
       "\n",
       "          gk_passes_throws  gk_goal_kicks  gk_crosses  gk_crosses_stopped  \n",
       "Consigli             174.0          246.0       491.0                21.0  \n",
       "\n",
       "[1 rows x 98 columns]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test = player_match_data_ext_gk('Consigli', 'Sassuolo', 'Atalanta')\n",
    "\n",
    "add_columns_gk = test.columns\n",
    "\n",
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6f08e718",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "100\n",
      "200\n",
      "300\n",
      "400\n",
      "500\n",
      "600\n",
      "700\n",
      "800\n",
      "900\n",
      "1000\n",
      "1100\n",
      "1200\n",
      "1300\n",
      "1400\n",
      "1500\n",
      "1600\n",
      "1700\n",
      "1800\n",
      "1900\n",
      "2000\n",
      "2100\n",
      "2200\n",
      "2300\n",
      "2400\n",
      "2500\n",
      "2600\n",
      "2700\n",
      "2800\n",
      "2900\n",
      "3000\n",
      "3100\n",
      "3200\n",
      "3300\n",
      "3400\n",
      "3500\n",
      "3600\n",
      "3700\n",
      "3800\n",
      "3900\n",
      "4000\n",
      "4100\n",
      "4200\n",
      "4300\n",
      "4400\n",
      "4500\n",
      "4600\n",
      "4700\n",
      "4800\n",
      "4900\n",
      "5000\n",
      "5100\n",
      "5200\n",
      "5300\n",
      "5400\n",
      "5500\n",
      "5600\n",
      "5700\n",
      "5800\n",
      "5900\n",
      "6000\n",
      "6100\n",
      "6200\n",
      "6300\n",
      "6400\n",
      "6500\n",
      "6600\n",
      "6700\n",
      "6800\n",
      "6900\n",
      "7000\n",
      "7100\n",
      "7200\n"
     ]
    }
   ],
   "source": [
    "db_gk = votes.copy()\n",
    "\n",
    "delta = pd.DataFrame(columns = add_columns_gk)\n",
    "\n",
    "for i in range(db_gk.shape[0]):\n",
    "    ext_data = player_match_data_ext_gk(db_gk['player'][i], db_gk['team'][i], db_gk['oppteam'][i])\n",
    "    \n",
    "    if(i % 100 == 0):\n",
    "        print(i)\n",
    "        \n",
    "    if(not isinstance(ext_data, pd.DataFrame)):\n",
    "        continue\n",
    "    \n",
    "    ext_data = ext_data.rename(index = {db_gk['player'][i] : i})\n",
    "    \n",
    "    delta = pd.concat([delta, ext_data], axis = 0)\n",
    "    \n",
    "    \n",
    "    \n",
    "db_gk = pd.concat([db_gk, delta], axis = 1)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2b540670",
   "metadata": {},
   "outputs": [],
   "source": [
    "db_gk = db_gk.drop(db_gk[db_gk.gk_games != db_gk.gk_games].index) # remove non extisting players"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "35495581",
   "metadata": {},
   "outputs": [],
   "source": [
    "db_gk.to_excel('mid_outputs/season' + season + '/database_entries_gk.xlsx')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "43b85925",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}