{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "376c303c",
   "metadata": {},
   "source": [
    "# Interpreting game ratings\n",
    "\n",
    "## TL;DR\n",
    "\n",
    "At a game-level, we can use the following rubric to evaluate performances:\n",
    "\n",
    "| Impact       | Description   |\n",
    "| ------------ | ------------- |\n",
    "| 0.75+        | Elite         |\n",
    "| 0.4 to 0.75  | Good          |\n",
    "| 0.25 to 0.4  | Above average |\n",
    "| 0.15 to 0.25 | Average       |\n",
    "| 0.05 to 0.15 | Below average |\n",
    "| 0.01 to 0.05 | Bad           |\n",
    "| $\\leq$0.01   | ... Bad       |\n",
    "\n",
    "and at a season-level,\n",
    "\n",
    "| Impact       | Description   |\n",
    "| ------------ | ------------- |\n",
    "| 0.55+        | Elite         |\n",
    "| 0.3 to 0.55  | Good          |\n",
    "| 0.2 to 0.3   | Above average |\n",
    "| 0.15 to 0.2  | Average       |\n",
    "| 0.1 to 0.15  | Below Average |\n",
    "| 0.01 to 0.1  | Bad           |\n",
    "| $\\leq$0.01   | ... Bad       |\n",
    "\n",
    "## Looking at game ratings\n",
    "\n",
    "### Including exogenous variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8a6be6f6",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [],
   "source": [
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "from nbaspa.data.endpoints.parameters import SEASONS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d2d0716f",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [],
   "source": [
    "filelist = []\n",
    "for season in SEASONS:\n",
    "    if season.startswith(\"2021\") or season.startswith(\"2005\"):\n",
    "        continue\n",
    "    filelist += list(Path(os.environ[\"DATA_DIR\"], season).glob(\"game-impact/data_*.csv\"))\n",
    "\n",
    "ratings = pd.concat(\n",
    "    (\n",
    "        pd.read_csv(fpath, sep=\"|\", index_col=0, dtype={\"GAME_ID\": str}) for fpath in filelist\n",
    "    ),\n",
    "    ignore_index=True\n",
    ")\n",
    "ratings = ratings[ratings[\"IMPACT\"] != 0.0].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b2b70d5b",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.axes_style(style=\"darkgrid\"):\n",
    "    fig, ax = plt.subplots(figsize=(12, 8))\n",
    "    sns.histplot(\n",
    "        x=\"IMPACT\",\n",
    "        data=ratings,\n",
    "        ax=ax\n",
    "    ).set(\n",
    "        title=\"Distribution of game impact ratings\",\n",
    "        xlabel=\"Impact\",\n",
    "        ylabel=\"Count\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8bf0a2e5",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "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>Game impact</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.975</th>\n",
       "      <td>0.729719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.800</th>\n",
       "      <td>0.373039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.600</th>\n",
       "      <td>0.227745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.400</th>\n",
       "      <td>0.133761</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.200</th>\n",
       "      <td>0.062914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.025</th>\n",
       "      <td>0.007094</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Game impact\n",
       "0.975     0.729719\n",
       "0.800     0.373039\n",
       "0.600     0.227745\n",
       "0.400     0.133761\n",
       "0.200     0.062914\n",
       "0.025     0.007094"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tiers = ratings[\"IMPACT\"].quantile(q=[0.975, 0.8, 0.6, 0.4, 0.2, 0.025]).to_frame()\n",
    "tiers.rename(columns={\"IMPACT\": \"Game impact\"})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9f27fb4e",
   "metadata": {},
   "source": [
    "### Excluding exogenous variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "1bda8f4b",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [],
   "source": [
    "filelist = []\n",
    "for season in SEASONS:\n",
    "    if season.startswith(\"2021\") or season.startswith(\"2005\"):\n",
    "        continue\n",
    "    filelist += list(Path(os.environ[\"DATA_DIR\"], season).glob(\"game-impact-plus/data_*.csv\"))\n",
    "\n",
    "ratings = pd.concat(\n",
    "    (\n",
    "        pd.read_csv(fpath, sep=\"|\", index_col=0, dtype={\"GAME_ID\": str}) for fpath in filelist\n",
    "    ),\n",
    "    ignore_index=True\n",
    ")\n",
    "ratings = ratings[ratings[\"IMPACT\"] != 0.0].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "92bfcbbe",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.axes_style(style=\"darkgrid\"):\n",
    "    fig, ax = plt.subplots(figsize=(12, 8))\n",
    "    sns.histplot(\n",
    "        x=\"IMPACT\",\n",
    "        data=ratings,\n",
    "        ax=ax\n",
    "    ).set(\n",
    "        title=\"Distribution of game impact ratings\",\n",
    "        xlabel=\"Impact\",\n",
    "        ylabel=\"Count\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "226bd297",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "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>Game ratings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.975</th>\n",
       "      <td>0.758746</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.800</th>\n",
       "      <td>0.403610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.600</th>\n",
       "      <td>0.254601</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.400</th>\n",
       "      <td>0.155604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.200</th>\n",
       "      <td>0.073957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.025</th>\n",
       "      <td>0.009486</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Game ratings\n",
       "0.975      0.758746\n",
       "0.800      0.403610\n",
       "0.600      0.254601\n",
       "0.400      0.155604\n",
       "0.200      0.073957\n",
       "0.025      0.009486"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tiers = ratings[\"IMPACT\"].quantile(q=[0.975, 0.8, 0.6, 0.4, 0.2, 0.025]).to_frame()\n",
    "tiers.rename(columns={\"IMPACT\": \"Game ratings\"})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6041e9f1",
   "metadata": {},
   "source": [
    "## Looking at season long performance\n",
    "\n",
    "### Including exogenous variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ebc159f6",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [],
   "source": [
    "filelist = []\n",
    "for season in SEASONS:\n",
    "    if season.startswith(\"2021\") or season.startswith(\"2005\"):\n",
    "        continue\n",
    "    filelist.append(Path(os.environ[\"DATA_DIR\"], season, \"impact-summary.csv\"))\n",
    "\n",
    "ratings = pd.concat(\n",
    "    (pd.read_csv(fpath, sep=\"|\", index_col=0) for fpath in filelist), ignore_index=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d75364d8",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.axes_style(style=\"darkgrid\"):\n",
    "    fig, ax = plt.subplots(figsize=(12, 8))\n",
    "    sns.histplot(\n",
    "        x=\"IMPACT_mean\",\n",
    "        data=ratings,\n",
    "        ax=ax\n",
    "    ).set(\n",
    "        title=\"Distribution of average season-long impact\",\n",
    "        xlabel=\"Average impact\",\n",
    "        ylabel=\"Count\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "798aad27",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "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>Season-long average impact</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.975</th>\n",
       "      <td>0.507773</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.800</th>\n",
       "      <td>0.281275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.600</th>\n",
       "      <td>0.185309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.400</th>\n",
       "      <td>0.126032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.200</th>\n",
       "      <td>0.081936</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.025</th>\n",
       "      <td>0.015990</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Season-long average impact\n",
       "0.975                    0.507773\n",
       "0.800                    0.281275\n",
       "0.600                    0.185309\n",
       "0.400                    0.126032\n",
       "0.200                    0.081936\n",
       "0.025                    0.015990"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tiers = ratings[\"IMPACT_mean\"].quantile(q=[0.975, 0.8, 0.6, 0.4, 0.2, 0.025]).to_frame()\n",
    "tiers.rename(columns={\"IMPACT_mean\": \"Season-long average impact\"})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1415d6d4",
   "metadata": {},
   "source": [
    "### Excluding exogenous variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "d1c70d2b",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [],
   "source": [
    "filelist = []\n",
    "for season in SEASONS:\n",
    "    if season.startswith(\"2021\") or season.startswith(\"2005\"):\n",
    "        continue\n",
    "    filelist.append(Path(os.environ[\"DATA_DIR\"], season, \"impact-plus-summary.csv\"))\n",
    "\n",
    "ratings = pd.concat(\n",
    "    (pd.read_csv(fpath, sep=\"|\", index_col=0) for fpath in filelist), ignore_index=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "37ffb6db",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.axes_style(style=\"darkgrid\"):\n",
    "    fig, ax = plt.subplots(figsize=(12, 8))\n",
    "    sns.histplot(\n",
    "        x=\"IMPACT_mean\",\n",
    "        data=ratings,\n",
    "        ax=ax\n",
    "    ).set(\n",
    "        title=\"Distribution of average season-long impact\",\n",
    "        xlabel=\"Average impact\",\n",
    "        ylabel=\"Count\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f780ed0b",
   "metadata": {
    "tags": [
     "remove-input"
    ]
   },
   "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>Season-long average impact</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.975</th>\n",
       "      <td>0.554305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.800</th>\n",
       "      <td>0.308529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.600</th>\n",
       "      <td>0.203010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.400</th>\n",
       "      <td>0.138607</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.200</th>\n",
       "      <td>0.088596</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.025</th>\n",
       "      <td>0.017254</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Season-long average impact\n",
       "0.975                    0.554305\n",
       "0.800                    0.308529\n",
       "0.600                    0.203010\n",
       "0.400                    0.138607\n",
       "0.200                    0.088596\n",
       "0.025                    0.017254"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tiers = ratings[\"IMPACT_mean\"].quantile(q=[0.975, 0.8, 0.6, 0.4, 0.2, 0.025]).to_frame()\n",
    "tiers.rename(columns={\"IMPACT_mean\": \"Season-long average impact\"})"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Tags",
  "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": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}