seasonal_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "seasons = [1516, 1617, 1718, 1819, 1920, 2021]\n",
    "seasons_2 = [20152016, 20162017, 20172018, 20182019, 20192020, 20202021]\n",
    "results_list = []\n",
    "\n",
    "for i in range(len(seasons)):\n",
    "  matches_df = pd.read_csv(f\"./datasets/matches/epl_matches_{seasons_2[i]}.csv\")\n",
    "  softmax_df = pd.read_csv(f\"./betting_simulation/new_probabilities/{seasons[i]}.csv\")\n",
    "  softmax_old_df = pd.read_csv(f\"./betting_simulation/original_probabilities/{seasons[i]}.csv\")\n",
    "  softmax_old_df = softmax_old_df.rename(columns={'home_prob_softmax': 'home_prob_softmax_old'})\n",
    "  combined_df = pd.merge(matches_df, softmax_df, on=\"match_url\")\n",
    "  combined_df = pd.merge(combined_df, softmax_old_df, on=\"match_url\")\n",
    "  combined_df['home_prob_softmax'] = combined_df['home_prob_softmax'] > 0.5\n",
    "  combined_df['home_prob_softmax'] = combined_df['home_prob_softmax'].astype(int)\n",
    "  combined_df['home_prob_softmax_old'] = combined_df['home_prob_softmax_old'] > 0.5\n",
    "  combined_df['home_prob_softmax_old'] = combined_df['home_prob_softmax_old'].astype(int)\n",
    "  results_df = pd.DataFrame()\n",
    "  results_df['match_url'] = combined_df['match_url']\n",
    "  results_df[f'{seasons[i]}_home_softmax'] = combined_df['home_prob_softmax']\n",
    "  results_df[f'{seasons[i]}_home_wins'] = combined_df['home_goals'] / combined_df['away_goals'] > 1.0\n",
    "  results_df[f'{seasons[i]}_home_wins'] = results_df[f'{seasons[i]}_home_wins'].astype(int)\n",
    "  results_df[f'{seasons[i]}_away_wins'] = combined_df['home_goals'] / combined_df['away_goals'] < 1.0\n",
    "  results_df[f'{seasons[i]}_away_wins'] = results_df[f'{seasons[i]}_away_wins'].astype(int)\n",
    "  results_df[f'{seasons[i]}_draws'] = combined_df['home_goals'] / combined_df['away_goals'] == 1.0\n",
    "  results_df[f'{seasons[i]}_draws'] = results_df[f'{seasons[i]}_draws'].astype(int)\n",
    "  results_df['home_prob_softmax_old'] = combined_df['home_prob_softmax_old']\n",
    "  results_list.append({\n",
    "      'season': seasons[i],\n",
    "      'actual_home_wins': results_df[f'{seasons[i]}_home_wins'].sum(),\n",
    "      'new_home_wins': results_df[f'{seasons[i]}_home_softmax'].sum(),\n",
    "      'old_home_wins': results_df['home_prob_softmax_old'].sum(),\n",
    "      'draws': results_df[f'{seasons[i]}_draws'].sum(),\n",
    "      'results_df': results_df,\n",
    "  })\n",
    "\n",
    "# print(results_list)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "seasons = [str(r['season']) for r in results_list]\n",
    "actual_home_wins = [r['actual_home_wins'] for r in results_list]\n",
    "new_home_wins = [r['new_home_wins'] for r in results_list]\n",
    "old_home_wins = [r['old_home_wins'] for r in results_list]\n",
    "draws = [r['draws'] for r in results_list]\n",
    "\n",
    "# Set the width of the bars\n",
    "bar_width = 0.15\n",
    "\n",
    "# Set the positions of the bars on the x-axis\n",
    "r1 = range(len(seasons))\n",
    "r2 = [x + bar_width for x in r1]\n",
    "r3 = [x + bar_width for x in r2]\n",
    "r4 = [x + bar_width for x in r3]\n",
    "\n",
    "# Create the actual and predicted bars\n",
    "plt.bar(r1, actual_home_wins, color='lightblue', width=bar_width, label='Actual Home Wins')\n",
    "plt.bar(r2, new_home_wins, color='green', width=bar_width, label='New Model Home Wins')\n",
    "plt.bar(r3, old_home_wins, color='red', width=bar_width, label='Old Model Home Wins')\n",
    "plt.bar(r4, draws, color='grey', width=bar_width, label='Draws')\n",
    "\n",
    "# Add x-axis ticks and labels\n",
    "plt.xticks([r + bar_width/2 for r in r2], seasons)\n",
    "\n",
    "# Add axis labels and title\n",
    "plt.xlabel('Season')\n",
    "plt.ylabel('Number of Home Wins')\n",
    "plt.title('Actual vs New Model Home Wins vs Old Model Home Wins vs Draws')\n",
    "\n",
    "# Add legend\n",
    "plt.legend()\n",
    "\n",
    "# Show the plot\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1516 57.10526315789474 56.315789473684205 80.26315789473685\n",
      "1516 45.26315789473684 49.73684210526316 80.26315789473685\n",
      "1617 59.473684210526315 60.0 85.0\n",
      "1617 52.10526315789473 51.05263157894737 85.0\n",
      "1718 57.89473684210527 55.52631578947368 82.36842105263158\n",
      "1718 45.78947368421053 47.10526315789474 82.36842105263158\n",
      "1819 66.05263157894737 68.42105263157895 87.10526315789474\n",
      "1819 42.368421052631575 41.05263157894737 87.10526315789474\n",
      "1920 61.05263157894737 61.05263157894737 81.3157894736842\n",
      "1920 50.0 48.68421052631579 81.57894736842105\n",
      "2021 66.84782608695652 64.13043478260869 83.69565217391305\n",
      "2021 45.108695652173914 47.82608695652174 83.69565217391305\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "seasons = [1516, 1617, 1718, 1819, 1920, 2021]\n",
    "seasons_2 = [20152016, 20162017, 20172018, 20182019, 20192020, 20202021]\n",
    "old_data, new_data = [], []\n",
    "for i in range(len(seasons)):\n",
    "  matches_df = pd.read_csv(f\"./datasets/matches/epl_matches_{seasons_2[i]}.csv\")\n",
    "  softmax_df = pd.read_csv(f\"./betting_simulation/new_probabilities/{seasons[i]}.csv\")\n",
    "  softmax_old_df = pd.read_csv(f\"./betting_simulation/original_probabilities/{seasons[i]}.csv\")\n",
    "  softmax_old_df = softmax_old_df.rename(columns={'home_prob_softmax': 'home_prob_softmax_old'})\n",
    "  combined_df = pd.merge(matches_df, softmax_df, on=\"match_url\")\n",
    "  combined_df = pd.merge(combined_df, softmax_old_df, on=\"match_url\")\n",
    "  combined_df['actual_home_wins'] = combined_df['home_goals'] / combined_df['away_goals'] > 1.0\n",
    "  combined_df['actual_away_wins'] = combined_df['home_goals'] / combined_df['away_goals'] < 1.0\n",
    "  combined_df['actual_draws'] = combined_df['home_goals'] / combined_df['away_goals'] == 1.0\n",
    "  combined_df['old_vs_actual_home_wins'] = (combined_df['home_prob_softmax_old'] > 0.5) == combined_df['actual_home_wins']\n",
    "  combined_df['old_vs_actual_away_wins'] = (combined_df['home_prob_softmax_old'] < 0.5) == combined_df['actual_away_wins']\n",
    "  combined_df['old_vs_actual_draws'] = (combined_df['home_prob_softmax_old'] == 0.5) == combined_df['actual_draws']\n",
    "  combined_df['new_vs_actual_home_wins'] = (combined_df['home_prob_softmax'] > 0.5) == combined_df['actual_home_wins']\n",
    "  combined_df['new_vs_actual_away_wins'] = (combined_df['home_prob_softmax'] < 0.5) == combined_df['actual_away_wins']\n",
    "  combined_df['new_vs_actual_draws'] = (combined_df['home_prob_softmax'] == 0.5) == combined_df['actual_draws']\n",
    "\n",
    "  # Get the fraction of true values in the old_vs_actual_home_wins column\n",
    "  old_vs_actual_home_wins = combined_df['old_vs_actual_home_wins'].mean() * 100\n",
    "  old_vs_actual_away_wins = combined_df['old_vs_actual_away_wins'].mean() * 100\n",
    "  old_vs_actual_draws = combined_df['old_vs_actual_draws'].mean() * 100\n",
    "  new_vs_actual_home_wins = combined_df['new_vs_actual_home_wins'].mean() * 100\n",
    "  new_vs_actual_away_wins = combined_df['new_vs_actual_away_wins'].mean() * 100\n",
    "  new_vs_actual_draws = combined_df['new_vs_actual_draws'].mean() * 100\n",
    "  print(f\"{seasons[i]}\", old_vs_actual_home_wins, old_vs_actual_away_wins, old_vs_actual_draws)\n",
    "  old_data.append((seasons[i], old_vs_actual_home_wins, old_vs_actual_away_wins, old_vs_actual_draws))\n",
    "  print(f\"{seasons[i]}\", new_vs_actual_home_wins, new_vs_actual_away_wins, new_vs_actual_draws)\n",
    "  new_data.append((seasons[i], new_vs_actual_home_wins, new_vs_actual_away_wins, new_vs_actual_draws))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Set the width of the bars\n",
    "bar_width = 0.1\n",
    "\n",
    "# Set the positions of the bars on the x-axis\n",
    "r1 = range(len(old_data))\n",
    "r2 = [x + bar_width for x in r1]\n",
    "r3 = [x + bar_width for x in r2]\n",
    "r4 = [x + bar_width for x in r3]\n",
    "r5 = [x + bar_width for x in r4]\n",
    "r6 = [x + bar_width for x in r5]\n",
    "\n",
    "# Create the old and new bars\n",
    "plt.bar(r1, [d[1] for d in old_data], color='green', width=bar_width, label='Old Model (Home Wins)')\n",
    "plt.bar(r2, [d[1] for d in new_data], color='yellow', width=bar_width, label='New Model (Home Wins)')\n",
    "plt.bar(r3, [d[2] for d in old_data], color='blue', width=bar_width, label='Old Model(Away Wins)')\n",
    "plt.bar(r4, [d[2] for d in new_data], color='orange', width=bar_width, label='New Model(Away Wins)')\n",
    "plt.bar(r5, [d[3] for d in old_data], color='brown', width=bar_width, label='Old Model(Draws)')\n",
    "plt.bar(r6, [d[3] for d in new_data], color='grey', width=bar_width, label='New Model(Draws)')\n",
    "\n",
    "# Add x-axis ticks and labels\n",
    "plt.xticks([r + bar_width/2 for r in r3], [d[0] for d in old_data])\n",
    "\n",
    "# Add axis labels and title\n",
    "plt.xlabel('Season')\n",
    "plt.ylabel('Accuracy (%)')\n",
    "plt.title('Comparison of Old and New Models Accuracy to Ground Truth')\n",
    "\n",
    "# # Add legend\n",
    "# plt.legend()\n",
    "\n",
    "# Show the plot\n",
    "plt.show()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "cs2109s",
   "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": "3.9.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}