{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "81215b75-f5f8-4c17-9cd4-08b1b0ed4234",
   "metadata": {},
   "source": [
    "# Goal Expectancy\n",
    "\n",
    "Estimates the goal expectancy (the number of goals the bookmaker expects each team to score) based on their home, draw, away probabilities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1f931497-c1f9-4cb4-969a-058676e42a24",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import penaltyblog as pb\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a1b5c76-8f47-4f59-8351-d5add2f69309",
   "metadata": {},
   "source": [
    "## Get data from football-data.co.uk"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "949b129d-e4e5-4975-8318-dd601d918e90",
   "metadata": {
    "tags": []
   },
   "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>team_home</th>\n",
       "      <th>team_away</th>\n",
       "      <th>psh</th>\n",
       "      <th>psd</th>\n",
       "      <th>psa</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</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>1565308800---liverpool---norwich</th>\n",
       "      <td>Liverpool</td>\n",
       "      <td>Norwich</td>\n",
       "      <td>1.15</td>\n",
       "      <td>9.59</td>\n",
       "      <td>18.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1565395200---bournemouth---sheffield_united</th>\n",
       "      <td>Bournemouth</td>\n",
       "      <td>Sheffield United</td>\n",
       "      <td>2.04</td>\n",
       "      <td>3.57</td>\n",
       "      <td>3.90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1565395200---burnley---southampton</th>\n",
       "      <td>Burnley</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>2.71</td>\n",
       "      <td>3.31</td>\n",
       "      <td>2.81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1565395200---crystal_palace---everton</th>\n",
       "      <td>Crystal Palace</td>\n",
       "      <td>Everton</td>\n",
       "      <td>3.21</td>\n",
       "      <td>3.37</td>\n",
       "      <td>2.39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1565395200---tottenham---aston_villa</th>\n",
       "      <td>Tottenham</td>\n",
       "      <td>Aston Villa</td>\n",
       "      <td>1.30</td>\n",
       "      <td>5.84</td>\n",
       "      <td>10.96</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                  team_home         team_away  \\\n",
       "id                                                                              \n",
       "1565308800---liverpool---norwich                  Liverpool           Norwich   \n",
       "1565395200---bournemouth---sheffield_united     Bournemouth  Sheffield United   \n",
       "1565395200---burnley---southampton                  Burnley       Southampton   \n",
       "1565395200---crystal_palace---everton        Crystal Palace           Everton   \n",
       "1565395200---tottenham---aston_villa              Tottenham       Aston Villa   \n",
       "\n",
       "                                              psh   psd    psa  \n",
       "id                                                              \n",
       "1565308800---liverpool---norwich             1.15  9.59  18.05  \n",
       "1565395200---bournemouth---sheffield_united  2.04  3.57   3.90  \n",
       "1565395200---burnley---southampton           2.71  3.31   2.81  \n",
       "1565395200---crystal_palace---everton        3.21  3.37   2.39  \n",
       "1565395200---tottenham---aston_villa         1.30  5.84  10.96  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fb = pb.scrapers.FootballData(\"ENG Premier League\", \"2019-2020\")\n",
    "df = fb.get_fixtures()\n",
    "\n",
    "cols = [\"team_home\", \"team_away\", \"psh\", \"psd\", \"psa\"]\n",
    "df = df[cols]\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "faf4dd3f-34ca-4547-bcac-46d45ae2e1f2",
   "metadata": {},
   "source": [
    "## Remove the overround from the odds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "57093b8f-9725-4740-9a93-35da39171380",
   "metadata": {
    "tags": []
   },
   "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>team_home</th>\n",
       "      <th>team_away</th>\n",
       "      <th>psh</th>\n",
       "      <th>psd</th>\n",
       "      <th>psa</th>\n",
       "      <th>home</th>\n",
       "      <th>draw</th>\n",
       "      <th>away</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</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>1565308800---liverpool---norwich</th>\n",
       "      <td>Liverpool</td>\n",
       "      <td>Norwich</td>\n",
       "      <td>1.15</td>\n",
       "      <td>9.59</td>\n",
       "      <td>18.05</td>\n",
       "      <td>0.859818</td>\n",
       "      <td>0.094528</td>\n",
       "      <td>0.045654</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1565395200---bournemouth---sheffield_united</th>\n",
       "      <td>Bournemouth</td>\n",
       "      <td>Sheffield United</td>\n",
       "      <td>2.04</td>\n",
       "      <td>3.57</td>\n",
       "      <td>3.90</td>\n",
       "      <td>0.481290</td>\n",
       "      <td>0.271206</td>\n",
       "      <td>0.247504</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1565395200---burnley---southampton</th>\n",
       "      <td>Burnley</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>2.71</td>\n",
       "      <td>3.31</td>\n",
       "      <td>2.81</td>\n",
       "      <td>0.360007</td>\n",
       "      <td>0.293118</td>\n",
       "      <td>0.346875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1565395200---crystal_palace---everton</th>\n",
       "      <td>Crystal Palace</td>\n",
       "      <td>Everton</td>\n",
       "      <td>3.21</td>\n",
       "      <td>3.37</td>\n",
       "      <td>2.39</td>\n",
       "      <td>0.302636</td>\n",
       "      <td>0.287845</td>\n",
       "      <td>0.409519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1565395200---tottenham---aston_villa</th>\n",
       "      <td>Tottenham</td>\n",
       "      <td>Aston Villa</td>\n",
       "      <td>1.30</td>\n",
       "      <td>5.84</td>\n",
       "      <td>10.96</td>\n",
       "      <td>0.758663</td>\n",
       "      <td>0.160665</td>\n",
       "      <td>0.080673</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                  team_home         team_away  \\\n",
       "id                                                                              \n",
       "1565308800---liverpool---norwich                  Liverpool           Norwich   \n",
       "1565395200---bournemouth---sheffield_united     Bournemouth  Sheffield United   \n",
       "1565395200---burnley---southampton                  Burnley       Southampton   \n",
       "1565395200---crystal_palace---everton        Crystal Palace           Everton   \n",
       "1565395200---tottenham---aston_villa              Tottenham       Aston Villa   \n",
       "\n",
       "                                              psh   psd    psa      home  \\\n",
       "id                                                                         \n",
       "1565308800---liverpool---norwich             1.15  9.59  18.05  0.859818   \n",
       "1565395200---bournemouth---sheffield_united  2.04  3.57   3.90  0.481290   \n",
       "1565395200---burnley---southampton           2.71  3.31   2.81  0.360007   \n",
       "1565395200---crystal_palace---everton        3.21  3.37   2.39  0.302636   \n",
       "1565395200---tottenham---aston_villa         1.30  5.84  10.96  0.758663   \n",
       "\n",
       "                                                 draw      away  \n",
       "id                                                               \n",
       "1565308800---liverpool---norwich             0.094528  0.045654  \n",
       "1565395200---bournemouth---sheffield_united  0.271206  0.247504  \n",
       "1565395200---burnley---southampton           0.293118  0.346875  \n",
       "1565395200---crystal_palace---everton        0.287845  0.409519  \n",
       "1565395200---tottenham---aston_villa         0.160665  0.080673  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def remove_overround(row):\n",
    "    odds = [\n",
    "        row[\"psh\"],\n",
    "        row[\"psd\"],\n",
    "        row[\"psa\"],\n",
    "    ]\n",
    "\n",
    "    odds = pb.implied.differential_margin_weighting(odds)\n",
    "\n",
    "    return pd.Series(odds[\"implied_probabilities\"])\n",
    "\n",
    "\n",
    "df[[\"home\", \"draw\", \"away\"]] = df.apply(remove_overround, axis=1)\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9257f0fc-5f2b-402f-9209-d005d14880be",
   "metadata": {},
   "source": [
    "## Get the goal expectancy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6b8b7d83-273e-4ab1-bac0-b7387556a03f",
   "metadata": {
    "tags": []
   },
   "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>team_home</th>\n",
       "      <th>team_away</th>\n",
       "      <th>home_expectancy</th>\n",
       "      <th>away_expectancy</th>\n",
       "      <th>success</th>\n",
       "      <th>error</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Liverpool</td>\n",
       "      <td>Norwich</td>\n",
       "      <td>3.219933</td>\n",
       "      <td>0.667064</td>\n",
       "      <td>True</td>\n",
       "      <td>5.381587e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Bournemouth</td>\n",
       "      <td>Sheffield United</td>\n",
       "      <td>1.403588</td>\n",
       "      <td>0.923020</td>\n",
       "      <td>True</td>\n",
       "      <td>5.932409e-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Burnley</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>1.099876</td>\n",
       "      <td>1.073906</td>\n",
       "      <td>True</td>\n",
       "      <td>1.595460e-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Crystal Palace</td>\n",
       "      <td>Everton</td>\n",
       "      <td>0.999968</td>\n",
       "      <td>1.213245</td>\n",
       "      <td>True</td>\n",
       "      <td>6.410602e-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Tottenham</td>\n",
       "      <td>Aston Villa</td>\n",
       "      <td>2.312856</td>\n",
       "      <td>0.607438</td>\n",
       "      <td>True</td>\n",
       "      <td>1.123882e-10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        team_home         team_away  home_expectancy  away_expectancy  \\\n",
       "0       Liverpool           Norwich         3.219933         0.667064   \n",
       "1     Bournemouth  Sheffield United         1.403588         0.923020   \n",
       "2         Burnley       Southampton         1.099876         1.073906   \n",
       "3  Crystal Palace           Everton         0.999968         1.213245   \n",
       "4       Tottenham       Aston Villa         2.312856         0.607438   \n",
       "\n",
       "   success         error  \n",
       "0     True  5.381587e-09  \n",
       "1     True  5.932409e-11  \n",
       "2     True  1.595460e-11  \n",
       "3     True  6.410602e-11  \n",
       "4     True  1.123882e-10  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output = list()\n",
    "for idx, row in df.head().iterrows():\n",
    "    res = pb.models.goal_expectancy(row[\"home\"], row[\"draw\"], row[\"away\"])\n",
    "\n",
    "    tmp = {\n",
    "        \"team_home\": row[\"team_home\"],\n",
    "        \"team_away\": row[\"team_away\"],\n",
    "        \"home_expectancy\": res[\"home_exp\"],\n",
    "        \"away_expectancy\": res[\"away_exp\"],\n",
    "        \"success\": res[\"success\"],\n",
    "        \"error\": res[\"error\"],\n",
    "    }\n",
    "\n",
    "    output.append(tmp)\n",
    "\n",
    "output = pd.DataFrame(output)\n",
    "output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e5997054-9134-43e3-9d08-797c183fe642",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.13.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}