NCAATourneyPredictor.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Runtime: 24.53 s\n",
      "\n",
      "Team Averages: \n",
      "      Season  TeamID  Wins  Losses        PPG       PAPG        RPG        APG  \\\n",
      "0    2006.0  1104.0  17.0    12.0  69.965517  67.137931  37.000000  12.206897   \n",
      "1    2006.0  1266.0  19.0    10.0  74.206897  69.655172  35.551724  15.275862   \n",
      "2    2006.0  1130.0  25.0     7.0  74.531250  66.875000  35.187500  17.812500   \n",
      "3    2006.0  1334.0  22.0     7.0  73.724138  63.586207  35.172414  15.655172   \n",
      "4    2006.0  1181.0  30.0     3.0  82.515152  68.969697  32.909091  15.363636   \n",
      "..      ...     ...   ...     ...        ...        ...        ...        ...   \n",
      "891  2019.0  1205.0  20.0    11.0  75.774194  73.419355  32.870968  13.612903   \n",
      "892  2019.0  1439.0  24.0     8.0  74.000000  62.093750  32.593750  15.343750   \n",
      "893  2019.0  1387.0  23.0    12.0  67.057143  63.742857  39.771429  12.971429   \n",
      "894  2019.0  1449.0  26.0     8.0  69.823529  64.382353  31.588235  11.676471   \n",
      "895  2019.0  1429.0  27.0     6.0  79.030303  67.090909  40.060606  17.090909   \n",
      "\n",
      "          SPG       BPG       PFPG       TOPG       SOS       FGP  \n",
      "0    6.344828  5.034483  13.827586  13.517241  0.578590  0.453444  \n",
      "1    7.517241  3.068966  19.482759  14.793103  0.567223  0.450089  \n",
      "2    5.937500  4.281250  16.500000  13.312500  0.532000  0.489275  \n",
      "3    5.931034  2.103448  15.379310  12.551724  0.460328  0.485192  \n",
      "4    9.545455  5.424242  17.484848  13.393939  0.586100  0.494334  \n",
      "..        ...       ...        ...        ...       ...       ...  \n",
      "891  6.225806  2.483871  16.193548  11.774194  0.446706  0.477726  \n",
      "892  6.656250  2.312500  15.281250  11.375000  0.546237  0.475476  \n",
      "893  7.085714  4.057143  17.485714  12.714286  0.508255  0.418000  \n",
      "894  9.000000  5.735294  18.411765  13.352941  0.541900  0.452575  \n",
      "895  6.121212  4.272727  18.636364  12.545455  0.510435  0.474974  \n",
      "\n",
      "[896 rows x 14 columns]\n",
      "All regular season matchups: \n",
      "        Season DayNum WTeamID WScore LTeamID LScore WLoc NumOT WFGM WFGA  ...  \\\n",
      "1        2003     10    1104     68    1328     62    N     0   27   58  ...   \n",
      "2        2003     10    1272     70    1393     63    N     0   26   62  ...   \n",
      "3        2003     11    1266     73    1437     61    N     0   24   58  ...   \n",
      "4        2003     11    1296     56    1457     50    N     0   18   38  ...   \n",
      "5        2003     11    1400     77    1208     71    N     0   30   61  ...   \n",
      "...       ...    ...     ...    ...     ...    ...  ...   ...  ...  ...  ...   \n",
      "87500    2019    132    1153     69    1222     57    N     0   22   50  ...   \n",
      "87501    2019    132    1209     73    1426     64    N     0   20   50  ...   \n",
      "87502    2019    132    1277     65    1276     60    N     0   22   55  ...   \n",
      "87503    2019    132    1387     55    1382     53    N     0   22   59  ...   \n",
      "87504    2019    132    1463     97    1217     85    H     0   32   53  ...   \n",
      "\n",
      "      LFGA3 LFTM LFTA LOR LDR LAst LTO LStl LBlk LPF  \n",
      "1        10   16   22  10  22    8  18    9    2  20  \n",
      "2        24    9   20  20  25    7  12    8    6  16  \n",
      "3        26   14   23  31  22    9  12    2    5  23  \n",
      "4        22    8   15  17  20    9  19    4    3  23  \n",
      "5        16   17   27  21  15   12  10    7    1  14  \n",
      "...     ...  ...  ...  ..  ..  ...  ..  ...  ...  ..  \n",
      "87500    33   11   18  17  16    8   7    2    4  19  \n",
      "87501    33   11   17  13  28   12  14    5    2  24  \n",
      "87502    25   10   12   3  26   17   6    5    5  11  \n",
      "87503    19    8   10  13  30    9  11    2    7  16  \n",
      "87504    32   19   24  12  15    7   9    1    2  22  \n",
      "\n",
      "[87504 rows x 34 columns]\n"
     ]
    }
   ],
   "source": [
    "import matplotlib\n",
    "from matplotlib import pylab\n",
    "import pandas as pd\n",
    "from import_data import import_legacy\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.model_selection import StratifiedShuffleSplit\n",
    "from matplotlib import pyplot\n",
    "import numpy as np\n",
    "from training import randomize_winner\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "import time\n",
    "\n",
    "#1\n",
    "# Import data from older seasons\n",
    "\n",
    "#Specify columns for team_avgs calculation from following list:\n",
    "    # ~~ Must always include Season, TeamID, Wins, Losses ~~\n",
    "    # PPG, PAPG, RPG, APG, SPG, BPG, PFPG, TOPG, SOS, FGP\n",
    "cols = ['Season', 'TeamID', 'Wins', 'Losses', 'PPG', 'PAPG', 'RPG', 'APG', 'SPG', 'BPG', 'PFPG', 'TOPG', 'SOS', 'FGP']\n",
    "\n",
    "# Specify seasons to import data from, from 2006 to 2019 (inclusive)\n",
    "years = [2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019]\n",
    "\n",
    "# import 2006-2019 regular season and tournament data, calculate team averages for tournament teams in each year\n",
    "runtime, team_avgs, reg_season_df, tourney_df, seeds_df = import_legacy(cols, years)\n",
    "\n",
    "# Estimated runtime: ~20 s\n",
    "print(f\"Runtime: {runtime:.2f} s\\n\")\n",
    "print(\"Team Averages: \\n\", team_avgs)\n",
    "print(\"All regular season matchups: \\n\", reg_season_df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Runtime: 0.64 s\n",
      "\n",
      "Team Averages: \n",
      "     Season  TeamID  Wins  Losses        PPG       PAPG        RPG        APG  \\\n",
      "0   2022.0  1211.0  26.0     3.0  87.827586  65.344828  41.482759  18.206897   \n",
      "1   2022.0  1112.0  31.0     3.0  84.558824  67.529412  41.352941  19.911765   \n",
      "2   2022.0  1242.0  28.0     6.0  78.588235  68.147059  37.352941  15.441176   \n",
      "3   2022.0  1124.0  26.0     6.0  76.500000  63.625000  37.125000  15.843750   \n",
      "4   2022.0  1120.0  27.0     5.0  78.718750  67.031250  39.937500  14.593750   \n",
      "..     ...     ...   ...     ...        ...        ...        ...        ...   \n",
      "63  2022.0  1313.0  24.0     6.0  75.100000  63.733333  38.766667  12.633333   \n",
      "64  2022.0  1460.0  21.0    13.0  75.470588  71.058824  34.882353  13.941176   \n",
      "65  2022.0  1136.0  22.0     9.0  77.903226  72.741935  41.548387  14.096774   \n",
      "66  2022.0  1411.0  18.0    12.0  69.233333  65.500000  39.533333  10.833333   \n",
      "67  2022.0  1394.0  23.0    11.0  76.911765  69.941176  38.352941  15.029412   \n",
      "\n",
      "         SPG       BPG       PFPG       TOPG       SOS       FGP  \n",
      "0   6.655172  5.862069  15.413793  11.758621  0.462941  0.527333  \n",
      "1   6.705882  5.705882  16.470588  13.176471  0.435251  0.495927  \n",
      "2   6.411765  4.147059  15.911765  12.500000  0.463304  0.480988  \n",
      "3   8.812500  3.406250  15.843750  12.500000  0.462177  0.463552  \n",
      "4   8.781250  7.937500  18.562500  12.093750  0.441351  0.439528  \n",
      "..       ...       ...        ...        ...       ...       ...  \n",
      "63  6.700000  3.666667  17.666667  14.000000  0.367469  0.465769  \n",
      "64  5.764706  2.941176  14.382353  12.294118  0.382200  0.464677  \n",
      "65  6.322581  4.451613  16.806452  13.903226  0.376877  0.436663  \n",
      "66  5.566667  5.000000  17.800000  14.933333  0.376836  0.436854  \n",
      "67  8.500000  1.735294  20.264706  14.205882  0.373556  0.444983  \n",
      "\n",
      "[68 rows x 14 columns]\n",
      "All regular season matchups: \n",
      "       Season DayNum WTeamID WScore LTeamID LScore WLoc NumOT WFGM WFGA  ...  \\\n",
      "1       2022      6       0     60    1240     57    H     0   22   57  ...   \n",
      "2       2022      7    1240     70       0     47    H     0   20   52  ...   \n",
      "3       2022      8       0     69    1240     65    H     3   23   68  ...   \n",
      "4       2022      9       0     78    1240     70    N     0   26   58  ...   \n",
      "5       2022     10    1240     72       0     64    N     0   25   50  ...   \n",
      "...      ...    ...     ...    ...     ...    ...  ...   ...  ...  ...  ...   \n",
      "1906    2022     37       0     61    1308     59    H     0   24   56  ...   \n",
      "1907    2022     38       0     73    1308     71    A     0   28   57  ...   \n",
      "1908    2022     39    1308     62       0     46    H     0   26   60  ...   \n",
      "1909    2022     40    1308     75       0     70    H     0   24   50  ...   \n",
      "1910    2022     41    1308     66       0     52    H     0   20   53  ...   \n",
      "\n",
      "     LFGA3 LFTM LFTA LOR LDR LAst LTO LStl LBlk LPF  \n",
      "1       20    5   11  10  21   12  17    6    6  12  \n",
      "2       22    6    9  10  22    6  12    7    0  18  \n",
      "3       23   16   26  10  40    9  20   14    2  28  \n",
      "4       27   14   20   9  29    7  14    4    4  20  \n",
      "5       24    5    6   6  19   11  12    5    1  15  \n",
      "...    ...  ...  ...  ..  ..  ...  ..  ...  ...  ..  \n",
      "1906    18   11   16  14  24   14  11    8    3  10  \n",
      "1907    20   14   25  12  22   15  18    2    7  17  \n",
      "1908    15    8   10   5  26    7  14    5    1  10  \n",
      "1909    30   12   17  19  21   15  13    6    0  19  \n",
      "1910    22   12   16   9  19    6  11    5    1  18  \n",
      "\n",
      "[1910 rows x 34 columns]\n"
     ]
    }
   ],
   "source": [
    "from import_data import import_year\n",
    "\n",
    "#2\n",
    "# Import data from 2022 season\n",
    "\n",
    "tourney_team_ids = [1211, 1112, 1242, 1124, 1120, 1246, 1437, 1181, 1458, 1397, 1345, 1403, 1417, 1228, 1344, 1116,\n",
    "                        1163, 1222, 1388, 1234, 1104, 1261, 1400, 1161, 1425, 1293, 1277, 1326, 1129, 1314, 1361, 1371,\n",
    "                        1166, 1395, 1266, 1272, 1362, 1274, 1260, 1172, 1235, 1276, 1461, 1353, 1231, 1439, 1323, 1412,\n",
    "                        1350, 1308, 1151, 1355, 1436, 1103, 1255, 1463, 1159, 1286, 1174, 1389, 1240, 1168, 1209, 1313,\n",
    "                        1460, 1136, 1411, 1394]\n",
    "\n",
    "#Specify columns for team_avgs calculation from following list:\n",
    "    # ~~ Must always include Season, TeamID, Wins, Losses ~~\n",
    "    # PPG, PAPG, RPG, APG, SPG, BPG, PFPG, TOPG, SOS, FGP\n",
    "cols = ['Season', 'TeamID', 'Wins', 'Losses', 'PPG', 'PAPG', 'RPG', 'APG', 'SPG', 'BPG', 'PFPG', 'TOPG', 'SOS', 'FGP']\n",
    "\n",
    "runtime, team_avgs_2022, reg_2022_df = import_year(cols)\n",
    "\n",
    "# Estimated runtime: 1 s\n",
    "print(f\"Runtime: {runtime:.2f} s\\n\")\n",
    "print(\"Team Averages: \\n\", team_avgs_2022)\n",
    "print(\"All regular season matchups: \\n\", reg_2022_df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Runtime: 2.95 s\n",
      "\n",
      "     Season  Team1ID  Team2ID  Team1W  Team1L   Team1PPG  Team1PAPG  \\\n",
      "252  2010.0   1124.0   1358.0    24.0     7.0  76.709677  66.548387   \n",
      "253  2010.0   1139.0   1431.0    28.0     4.0  70.093750  60.031250   \n",
      "254  2010.0   1140.0   1196.0    28.0     5.0  82.606061  65.666667   \n",
      "255  2010.0   1242.0   1250.0    32.0     2.0  81.764706  63.794118   \n",
      "256  2010.0   1243.0   1317.0    25.0     7.0  79.468750  69.218750   \n",
      "..      ...      ...      ...     ...     ...        ...        ...   \n",
      "877  2019.0   1246.0   1120.0    27.0     6.0  76.727273  65.424242   \n",
      "878  2019.0   1181.0   1277.0    29.0     5.0  83.500000  67.558824   \n",
      "879  2019.0   1277.0   1403.0    28.0     6.0  78.823529  65.529412   \n",
      "880  2019.0   1438.0   1120.0    29.0     3.0  71.843750  55.062500   \n",
      "881  2019.0   1438.0   1403.0    29.0     3.0  71.843750  55.062500   \n",
      "\n",
      "      Team1RPG   Team1APG  Team1SPG  ...   Team2PPG  Team2PAPG   Team2RPG  \\\n",
      "252  38.645161  13.645161  6.387097  ...  79.892857  72.321429  36.428571   \n",
      "253  32.843750  12.812500  6.843750  ...  75.781250  64.187500  36.343750   \n",
      "254  36.969697  15.787879  8.545455  ...  71.606061  65.424242  36.969697   \n",
      "255  40.529412  17.147059  8.411765  ...  75.125000  70.031250  36.500000   \n",
      "256  39.250000  14.625000  7.468750  ...  73.466667  69.900000  36.433333   \n",
      "..         ...        ...       ...  ...        ...        ...        ...   \n",
      "877  38.424242  13.848485  6.030303  ...  78.882353  68.588235  33.676471   \n",
      "878  41.764706  15.911765  9.470588  ...  78.823529  65.529412  40.911765   \n",
      "879  40.911765  18.941176  5.235294  ...  73.093750  59.281250  34.281250   \n",
      "880  34.625000  14.500000  5.625000  ...  78.882353  68.588235  33.676471   \n",
      "881  34.625000  14.500000  5.625000  ...  73.093750  59.281250  34.281250   \n",
      "\n",
      "      Team2APG  Team2SPG  Team2BPG  Team2PFPG  Team2TOPG  Team2SOS  Team2FGP  \n",
      "252  19.642857  7.535714  2.250000  20.571429  13.428571  0.472801  0.466707  \n",
      "253  14.781250  8.750000  3.718750  19.343750  13.156250  0.513885  0.475851  \n",
      "254  14.060606  7.151515  2.727273  14.454545  12.181818  0.558072  0.443983  \n",
      "255  15.125000  6.156250  2.531250  17.906250  13.000000  0.439309  0.456170  \n",
      "256  11.966667  4.433333  3.000000  18.100000  14.766667  0.452013  0.459100  \n",
      "..         ...       ...       ...        ...        ...       ...       ...  \n",
      "877  14.411765  9.294118  4.764706  18.382353  12.147059  0.566801  0.447202  \n",
      "878  18.941176  5.235294  5.470588  16.911765  12.852941  0.564069  0.485597  \n",
      "879  14.031250  7.375000  4.906250  17.781250  12.375000  0.542954  0.471549  \n",
      "880  14.411765  9.294118  4.764706  18.382353  12.147059  0.566801  0.447202  \n",
      "881  14.031250  7.375000  4.906250  17.781250  12.375000  0.542954  0.471549  \n",
      "\n",
      "[630 rows x 27 columns]\n",
      "252    1\n",
      "253    1\n",
      "254    1\n",
      "255    1\n",
      "256    1\n",
      "      ..\n",
      "877    2\n",
      "878    2\n",
      "879    2\n",
      "880    1\n",
      "881    1\n",
      "Name: Winner, Length: 630, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "from training import randomize_winner\n",
    "#3\n",
    "# Randomize order of winner in historical seasons\n",
    "# Can't apply this to 2022 season due to differences in data\n",
    "\n",
    "start_time = time.time()\n",
    "\n",
    "train_cols = [\"Season\", \"Team1ID\", \"Team2ID\", \"Winner\", \"Team1W\", \"Team1L\", \"Team1PPG\", \"Team1PAPG\", \"Team1RPG\", \"Team1APG\",\n",
    "              \"Team1SPG\", \"Team1BPG\", \"Team1PFPG\", \"Team1TOPG\", \"Team1SOS\", \"Team1FGP\", \"Team2W\", \"Team2L\", \"Team2PPG\",\n",
    "              \"Team2PAPG\", \"Team2RPG\", \"Team2APG\", \"Team2SPG\", \"Team2BPG\", \"Team2PFPG\", \"Team2TOPG\", \"Team2SOS\", \"Team2FGP\"]\n",
    "\n",
    "trim_tourney_df = tourney_df[[\"Season\", \"WTeamID\", \"LTeamID\"]]\n",
    "\n",
    "# For historical data, randomize the order of winning and losing teams\n",
    "train_DF = randomize_winner(trim_tourney_df, team_avgs, train_cols, seeds_df, cols, years)\n",
    "\n",
    "# Limit data to more recent games (since 2010)\n",
    "train_data = train_DF[train_DF['Season'] >= 2010]\n",
    "train_labels = train_data['Winner'].astype(\"int\")\n",
    "train_data = train_data.drop(columns=['Winner'])\n",
    "\n",
    "print(f\"Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "print(train_data)\n",
    "print(train_labels)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Runtime: 0.11 s\n",
      "\n",
      "    Season  Team1ID  Team2ID  Team1W  Team1L   Team1PPG  Team1PAPG   Team1RPG  \\\n",
      "0   2022.0   1211.0   1209.0    26.0     3.0  87.827586  65.344828  41.482759   \n",
      "1   2022.0   1129.0   1272.0    27.0     7.0  68.705882  60.764706  34.970588   \n",
      "2   2022.0   1163.0   1308.0    23.0     9.0  75.187500  65.156250  40.875000   \n",
      "3   2022.0   1116.0   1436.0    25.0     8.0  76.939394  68.545455  38.151515   \n",
      "4   2022.0   1104.0   1323.0    19.0    13.0  79.968750  76.406250  39.968750   \n",
      "..     ...      ...      ...     ...     ...        ...        ...        ...   \n",
      "58  2022.0   1437.0   1222.0    26.0     7.0  72.636364  63.090909  34.939394   \n",
      "59  2022.0   1242.0   1274.0    28.0     6.0  78.588235  68.147059  37.352941   \n",
      "60  2022.0   1181.0   1314.0    28.0     6.0  80.176471  67.147059  38.294118   \n",
      "61  2022.0   1242.0   1437.0    28.0     6.0  78.588235  68.147059  37.352941   \n",
      "62  2022.0   1242.0   1314.0    28.0     6.0  78.588235  68.147059  37.352941   \n",
      "\n",
      "     Team1APG  Team1SPG  ...   Team2PPG  Team2PAPG   Team2RPG   Team2APG  \\\n",
      "0   18.206897  6.655172  ...  70.607143  64.750000  37.250000  13.535714   \n",
      "1   11.529412  5.911765  ...  75.419355  68.387097  38.741935  16.064516   \n",
      "2   14.000000  5.968750  ...  73.312500  65.031250  38.468750  13.906250   \n",
      "3   14.090909  7.727273  ...  74.939394  60.303030  35.090909  15.212121   \n",
      "4   14.625000  7.000000  ...  72.500000  66.843750  34.062500  14.156250   \n",
      "..        ...       ...  ...        ...        ...        ...        ...   \n",
      "58  12.060606  6.242424  ...  75.823529  58.882353  39.235294  16.735294   \n",
      "59  15.441176  6.411765  ...  74.818182  71.030303  30.454545  14.030303   \n",
      "60  16.911765  6.470588  ...  77.515152  71.727273  39.757576  14.818182   \n",
      "61  15.441176  6.411765  ...  72.636364  63.090909  34.939394  12.060606   \n",
      "62  15.441176  6.411765  ...  77.515152  71.727273  39.757576  14.818182   \n",
      "\n",
      "    Team2SPG  Team2BPG  Team2PFPG  Team2TOPG  Team2SOS  Team2FGP  \n",
      "0   8.928571  4.464286  15.750000  11.857143  0.372757  0.402650  \n",
      "1   8.709677  5.677419  18.709677  16.354839  0.424656  0.470688  \n",
      "2   5.062500  4.156250  15.875000  14.000000  0.375876  0.457459  \n",
      "3   5.212121  2.515152  13.787879   9.696970  0.386042  0.493492  \n",
      "4   5.093750  1.875000  12.875000  10.343750  0.437692  0.462396  \n",
      "..       ...       ...        ...        ...       ...       ...  \n",
      "58  8.235294  5.205882  17.382353  11.323529  0.403810  0.468888  \n",
      "59  8.727273  2.939394  14.939394   9.606061  0.415655  0.478192  \n",
      "60  5.424242  3.848485  14.363636  11.666667  0.446561  0.452830  \n",
      "61  6.242424  2.272727  15.242424  10.000000  0.489467  0.437029  \n",
      "62  5.424242  3.848485  14.363636  11.666667  0.446561  0.452830  \n",
      "\n",
      "[63 rows x 27 columns]\n"
     ]
    }
   ],
   "source": [
    "from training import tourney_2022\n",
    "import time\n",
    "\n",
    "#4\n",
    "# Calculate stat averages for 2022 tourney teams\n",
    "\n",
    "start_time = time.time()\n",
    "\n",
    "d = {'Team1ID': [1211,1129,1163,1116,1104,1403,1277,1181,1124,1314,1388,1417,1400,1345,1293,1246,1112,1371,1222,1228,1161,1397,1326,1437,1242,1361,1234,1344,1261,1458,1425,1120,1211,1116,1403,1181,1124,1417,1345,1293,1112,1228,1397,1437,1242,1344,1458,1120,1211,1181,1417,1345,1112,1437,1242,1274,1181,1314,1437,1242,1181,1242,1242], 'Team2ID': [1209,1272,1308,1436,1323,1286,1172,1168,1313,1266,1231,1103,1439,1463,1362,1389,1460,1395,1412,1151,1276,1255,1260,1174,1411,1166,1350,1355,1235,1159,1274,1240,1272,1308,1323,1277,1314,1388,1400,1389,1395,1222,1276,1326,1166,1350,1235,1274,1116,1403,1314,1389,1222,1276,1344,1235,1116,1389,1222,1274,1314,1437,1314]}\n",
    "test_cols = [\"Season\", \"Team1ID\", \"Team2ID\", \"Team1W\", \"Team1L\", \"Team1PPG\", \"Team1PAPG\", \"Team1RPG\", \"Team1APG\", \"Team1SPG\", \\\n",
    "              \"Team1BPG\", \"Team1PFPG\", \"Team1TOPG\", \"Team1SOS\", \"Team1FGP\", \"Team2W\", \"Team2L\", \"Team2PPG\", \"Team2PAPG\", \"Team2RPG\", \"Team2APG\", \"Team2SPG\",\\\n",
    "              \"Team2BPG\", \"Team2PFPG\", \"Team2TOPG\", \"Team2SOS\", \"Team2FGP\"]\n",
    "\n",
    "\n",
    "tourney_2022_df = tourney_2022(d, team_avgs_2022, test_cols)\n",
    "print(f\"Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "print(tourney_2022_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Years: 2010 to 2019\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#5\n",
    "# Shows histogram distribution of Point Differential and Strength of Schedule\n",
    "# For winning teams and losing teams\n",
    "\n",
    "# What years would you like to see data for? (max range is 2010 to 2019)\n",
    "\n",
    "train_DF = randomize_winner(trim_tourney_df, team_avgs, train_cols, seeds_df, cols, years)\n",
    "\n",
    "year_range = [2010, 2019]\n",
    "\n",
    "def team_lists(team_avgs, game_data, year):\n",
    "    temp_df = game_data[game_data[\"Season\"] == year]\n",
    "    winner_list = temp_df[\"WTeamID\"].tolist()\n",
    "    loser_list = temp_df[\"LTeamID\"].tolist()\n",
    "    return winner_list, loser_list\n",
    "\n",
    "\n",
    "def print_histogram(team_avgs, game_data, years):\n",
    "\n",
    "    sos_winner = []\n",
    "    sos_loser = []\n",
    "    pd_winner = []\n",
    "    pd_loser = []\n",
    "\n",
    "    for year in years:\n",
    "        winner_list, loser_list = team_lists(team_avgs, game_data, year)\n",
    "        temp_df = team_avgs[team_avgs[\"Season\"] == year]\n",
    "        for i in range(32):\n",
    "            current = temp_df[temp_df[\"TeamID\"] == int(winner_list[i])]\n",
    "            sos_winner.append(current[\"SOS\"].values[0])\n",
    "            pd_winner.append(current[\"PPG\"].values[0] - current[\"PAPG\"].values[0])\n",
    "\n",
    "            current = temp_df.loc[temp_df[\"TeamID\"] == int(loser_list[i])]\n",
    "            sos_loser.append(current[\"SOS\"].values[0])\n",
    "            pd_loser.append(current[\"PPG\"].values[0] - current[\"PAPG\"].values[0])\n",
    "\n",
    "\n",
    "    #pyplot.hist(ppg_all, alpha=0.5, label=\"All teams\")\n",
    "    print(f\"Years: {year_range[0]} to {year_range[1]}\")\n",
    "    plot1 = pyplot.subplot(1, 2, 1)\n",
    "    pyplot.hist(x=[pd_winner, pd_loser], alpha=0.5, label=[\"Won at least 1 game\", \"Lost in 1st round\"])\n",
    "    pyplot.title(\"Point Differential\")\n",
    "    pyplot.legend(loc='best')\n",
    "    plot2 = pyplot.subplot(1, 2, 2)\n",
    "    pyplot.hist(x=[sos_winner, sos_loser], alpha=0.5, label=[\"Won at least 1 game\", \"Lost in 1st round\"])\n",
    "    pyplot.title(\"Strength of Schedule\")\n",
    "    pyplot.legend(loc='best')\n",
    "\n",
    "    #pyplot.hist(ppg_loser, alpha=0.5, label=\"Lost in 1st round\")\n",
    "\n",
    "    pyplot.show()\n",
    "\n",
    "print_histogram(team_avgs, tourney_df, list(range(year_range[0], year_range[1]+1)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Runtime: 0.0003 s\n",
      "\n",
      "Vegas # of Games correct:  42\n",
      "Vegas # of Games missed:  21\n",
      "Vegas Accuracy:  0.6666666666666666\n"
     ]
    }
   ],
   "source": [
    "#6\n",
    "# Calculate vegas game-by-game accuracy for 2022 tournament\n",
    "start_time = time.time()\n",
    "winners = [1,2,2,1,2,1,1,1,1,1,1,1,1,1,1,2,1,2,1,1,2,1,1,1,1,2,2,1,2,1,2,1,1,1,1,1,2,1,1,2,1,2,2,1,1,1,2,2,2,1,2,2,2,1,1,1,1,1,1,1,2,1,1]\n",
    "vegas_winners = [1,2,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,2,1,1,1,1,1,1,1,2,1,1,1,1,1,1,1,1,2,1,1,1,1]\n",
    "\n",
    "vright = 0\n",
    "vwrong = 0\n",
    "for i in range(0,len(vegas_winners)):\n",
    "    if vegas_winners[i] == winners[i]:\n",
    "        vright += 1\n",
    "    else:\n",
    "        vwrong += 1\n",
    "\n",
    "#print(f\"# of Games correct: \",right)\n",
    "#print(f\"# of Games missed: \",wrong)\n",
    "#print(f\"Accuracy: \", (right) / (right + wrong))\n",
    "\n",
    "print(f\"Runtime: {time.time() - start_time:.4f} s\\n\")\n",
    "\n",
    "print(f\"Vegas # of Games correct: \",vright)\n",
    "print(f\"Vegas # of Games missed: \",vwrong)\n",
    "print(f\"Vegas Accuracy: \", (vright) / (vright + vwrong))\n",
    "\n",
    "def accuracy(preds):\n",
    "    right = 0\n",
    "    wrong = 0\n",
    "    for i in range(0,len(preds)):\n",
    "        if preds[i] == winners[i]:\n",
    "            right += 1\n",
    "        else:\n",
    "            wrong += 1\n",
    "    return right / (right + wrong)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "import xgboost as xgb\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn import svm\n",
    "\n",
    "xgb.set_config(verbosity=0)\n",
    "\n",
    "#7\n",
    "# Defines a function to predict games using a specified classification model\n",
    "# NO OUTPUT for this cell, but it should be run to define the functions\n",
    "\n",
    "def trainAndPredict(classifier, test_df, sampleSize, num_games):\n",
    "\n",
    "    predicted_games = [0] * num_games\n",
    "\n",
    "    for i in range(sampleSize):\n",
    "\n",
    "        train_df_year = train_DF[train_DF['Season'] >= 2010]\n",
    "\n",
    "        X = train_df_year.drop(\"Season\", axis=1)\n",
    "        X = X.drop([\"Winner\", \"Team1ID\", \"Team2ID\"], axis=1)\n",
    "        y = train_df_year[\"Winner\"]\n",
    "        y = y.astype(int)\n",
    "\n",
    "        test_X = test_df.drop([\"Season\", \"Team1ID\", \"Team2ID\"], axis=1)\n",
    "\n",
    "        scaler = StandardScaler()\n",
    "        scaled_X = scaler.fit_transform(X.values)\n",
    "        scaled_test = scaler.fit_transform(test_X.values)\n",
    "        X_proc = pd.DataFrame(scaled_X, index = X.index, columns = X.columns)\n",
    "        test_proc = pd.DataFrame(scaled_test, index = test_X.index, columns = test_X.columns)\n",
    "\n",
    "        # This code takes a VERY long time to run (>30 min) so it is commented out\n",
    "\n",
    "        #C_range = np.logspace(-2, 10, 13)\n",
    "        #gamma_range = np.logspace(-9, 3, 13)\n",
    "        #param_grid = dict(gamma=gamma_range, C=C_range)\n",
    "        #cv = StratifiedShuffleSplit(n_splits=2, test_size=0.2, random_state=42)\n",
    "        #grid = GridSearchCV(svm.SVC(), param_grid=param_grid, cv=cv)\n",
    "        #grid.fit(X, y)\n",
    "\n",
    "        #print(\n",
    "        #    \"The best parameters are %s with a score of %0.2f\"\n",
    "        #    % (grid.best_params_, grid.best_score_)\n",
    "        #)\n",
    "\n",
    "        classifier.fit(X_proc, y)\n",
    "\n",
    "        preds = classifier.predict(test_proc)\n",
    "        for i in range(len(preds)):\n",
    "            if preds[i] == 1:\n",
    "                predicted_games[i] += 1\n",
    "        \n",
    "    overall_pred = [1] * num_games\n",
    "    for i in range(len(predicted_games)):\n",
    "        if predicted_games[i] < (sampleSize / 2):\n",
    "            overall_pred[i] = 2\n",
    "    return overall_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Runtime: 0.02 s\n",
      "\n",
      "SVM: 0.6508\n"
     ]
    }
   ],
   "source": [
    "#8\n",
    "#Game-by-game prediction accuracy:\n",
    "\n",
    "start_time = time.time()\n",
    "\n",
    "svm_preds = trainAndPredict(svm.SVC(C=4.2, gamma=0.01), tourney_2022_df, 1, 63)\n",
    "\n",
    "print(f\"Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "print(f\"SVM: {accuracy(svm_preds):.4f}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM Runtime: 0.22 s\n",
      "\n",
      "Avg game-by-game accuracy across all games from 2010-2019: 0.7032\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(array([2., 3., 2., 1., 0., 0., 1., 0., 1., 1.]),\n",
       " array([0.65079365, 0.66507937, 0.67936508, 0.69365079, 0.70793651,\n",
       "        0.72222222, 0.73650794, 0.75079365, 0.76507937, 0.77936508,\n",
       "        0.79365079]),\n",
       " <a list of 10 Patch objects>)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#9\n",
    "# Predicts game-by-game outcomes from specified years\n",
    "\n",
    "def predictYear(classifier, year, num_games):\n",
    "    \n",
    "    predicted_games = [0] * num_games\n",
    "    \n",
    "    test_data = train_DF[train_DF['Season'] == year]\n",
    "    test_labels = test_data['Winner'].astype(\"int\")\n",
    "    test_data = test_data.drop(columns=['Winner'])\n",
    "\n",
    "    training_data = train_DF[train_DF['Season'] != year]\n",
    "    y = training_data['Winner']\n",
    "\n",
    "    test_X = test_data.drop([\"Season\", \"Team1ID\", \"Team2ID\"], axis=1)\n",
    "    X = training_data.drop(\"Season\", axis=1)\n",
    "    X = X.drop([\"Winner\", \"Team1ID\", \"Team2ID\"], axis=1)\n",
    "    y = y.astype(int)\n",
    "\n",
    "    scaler = StandardScaler()\n",
    "    scaled_X = scaler.fit_transform(X.values)\n",
    "    scaled_test = scaler.fit_transform(test_X.values)\n",
    "    test_proc = pd.DataFrame(scaled_test, index = test_X.index, columns = test_X.columns)\n",
    "\n",
    "    X_proc = pd.DataFrame(scaled_X, index = X.index, columns = X.columns)\n",
    "\n",
    "    accuracy_scores = []\n",
    "    from sklearn.metrics import accuracy_score\n",
    "\n",
    "    # Model Fit\n",
    "    classifier.fit(X_proc, y)\n",
    "\n",
    "    # Model Prediction\n",
    "    preds = classifier.predict(test_proc)\n",
    "\n",
    "    # Accuracy Score\n",
    "    return accuracy_score(test_labels, preds)\n",
    "\n",
    "\n",
    "def predictHistorical(classifier):\n",
    "    acc_scores = []\n",
    "    for year in years:\n",
    "        if year >= 2010:\n",
    "            score = predictYear(classifier, year, 63)\n",
    "            acc_scores.append(score)\n",
    "    return acc_scores\n",
    "\n",
    "start_time = time.time()\n",
    "svm_accuracies = predictHistorical(svm.SVC(C=4.6, gamma=0.01))\n",
    "print(f\"SVM Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "print(f\"Avg game-by-game accuracy across all games from 2010-2019: {sum(svm_accuracies) / len(svm_accuracies):.4f}\")\n",
    "\n",
    "svm_accuracies.append(accuracy(svm_preds))\n",
    "pyplot.hist(svm_accuracies)        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM Bracket Score (2022): 880\n",
      "Runtime: 0.20 s\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#10\n",
    "# Predicts and calculates bracket score for 2022 bracket\n",
    "\n",
    "#All of the matchups had to be entered by hand due to the nature of the data collection for 2022\n",
    "#This is something we would change the next time we do the project\n",
    "firstRound = {'Team1ID': [1211,1129,1163,1116,1104,1403,1277,1181,1124,1314,1388,1417,1400,1345,1293,1246,1112,1371,1222,1228,1161,1397,1326,1437,1242,1361,1234,1344,1261,1458,1425,1120], 'Team2ID': [1209,1272,1308,1436,1323,1286,1172,1168,1313,1266,1231,1103,1439,1463,1362,1389,1460,1395,1412,1151,1276,1255,1260,1174,1411,1166,1350,1355,1235,1159,1274,1240]}\n",
    "tournamentRounds = {'Round1': [1211,1272,1308,1116,1323,1403,1277,1181,1124,1314,1388,1417,1400,1345,1293,1389,1112,1395,1222,1228,1276,1397,1326,1437,1242,1166,1350,1344,1235,1458,1274,1120], \\\n",
    "                    'Round2': [1211,1116,1403,1181,1314,1417,1345,1389,1112,1222,1276,1437,1242,1344,1235,1274], \\\n",
    "                    'Round3': [1116,1181,1314,1389,1222,1437,1242,1274], \\\n",
    "                    'Round4': [1181,1314,1437,1242], 'Round5': [1314, 1242], 'Round6': [1242]}\n",
    "\n",
    "#Each of the teams are assigned their correct seed, to account for sorting. This was done using an algorithm\n",
    "#but was included as hard-coded since the seeds never change to reduce runtime\n",
    "seed_dict = {1211: 'W01', 1209: 'W16', 1129: 'W08', 1272: 'W09', 1163: 'W05', 1308: 'W12', 1116: 'W04', 1436: 'W13', 1104: 'W06', 1323: 'W11', 1403: 'W03', 1286: 'W14', 1277: 'W07', 1172: 'W10', 1181: 'W02', 1168: 'W15', 1124: 'X01', 1313: 'X16', 1314: 'X08', 1266: 'X09', 1388: 'X05', 1231: 'X12', 1417: 'X04', 1103: 'X13', 1400: 'X06', 1439: 'X11', 1345: 'X03', 1463: 'X14', 1293: 'X07', 1362: 'X10', 1246: 'X02', 1389: 'X15', 1112: 'Y01', 1460: 'Y16', 1371: 'Y08', 1395: 'Y09', 1222: 'Y05', 1412: 'Y12', 1228: 'Y04', 1151: 'Y13', 1161: 'Y06', 1276: 'Y11', 1397: 'Y03', 1255: 'Y14', 1326: 'Y07', 1260: 'Y10', 1437: 'Y02', 1174: 'Y15', 1242: 'Z01', 1411: 'Z16', 1361: 'Z08', 1166: 'Z09', 1234: 'Z05', 1350: 'Z12', 1344: 'Z04', 1355: 'Z13', 1261: 'Z06', 1235: 'Z11', 1458: 'Z03', 1159: 'Z14', 1425: 'Z07', 1274: 'Z10', 1120: 'Z02', 1240: 'Z15'}\n",
    "bracket_df = pd.DataFrame(data=firstRound)\n",
    "\n",
    "#Predicts tournament outcome and calculates score\n",
    "def tournament(classifier, round_bracket, round_num, bracket_score):\n",
    "    round_df = pd.DataFrame(columns=test_cols)\n",
    "    for row in round_bracket.iterrows():\n",
    "        temp_list = [2022, row[1]['Team1ID'], row[1]['Team2ID']]\n",
    "        \n",
    "        #ordering new teams by seed\n",
    "        if round_num > 1:\n",
    "            team1_seed = seed_dict[temp_list[1]]\n",
    "            team2_seed = seed_dict[temp_list[2]]\n",
    "\n",
    "            \n",
    "            team1_prefix = team1_seed[0:1]\n",
    "            team2_prefix = team2_seed[0:1]\n",
    "\n",
    "            team1_seed = int(team1_seed[1:3])\n",
    "            team2_seed = int(team2_seed[1:3])\n",
    "            \n",
    "            if team1_seed > team2_seed:\n",
    "                temp = temp_list[1]\n",
    "                temp_list[1] = temp_list[2]\n",
    "                temp_list[2] = temp\n",
    "                \n",
    "                \n",
    "        #print(temp_list)\n",
    "        team1_row = team_avgs_2022[team_avgs_2022[\"TeamID\"] == temp_list[1]].values.tolist()\n",
    "        team2_row = team_avgs_2022[team_avgs_2022[\"TeamID\"] == temp_list[2]].values.tolist()\n",
    "        team1_row[0].pop(0)\n",
    "        team1_row[0].pop(0)\n",
    "        team2_row[0].pop(0)\n",
    "        team2_row[0].pop(0)\n",
    "\n",
    "        temp_list.extend(team1_row[0])\n",
    "        temp_list.extend(team2_row[0])\n",
    "        # add that list to our DF\n",
    "        temp_series = pd.Series(temp_list, index=test_cols)\n",
    "        round_df = round_df.append(temp_series, ignore_index=True)\n",
    "        \n",
    "    #print(round_df)\n",
    "    preds = trainAndPredict(classifier, round_df, 1, len(round_df.index))\n",
    "\n",
    "    #if(round_num == 1):\n",
    "        #return;\n",
    "    \n",
    "    round_dict = {'Team1ID': [], 'Team2ID': []}\n",
    "    winners_arr = []\n",
    "    for i in range(0, len(preds)):\n",
    "        if i % 2 == 0:\n",
    "            length = len(round_dict['Team1ID'])\n",
    "            val = 0\n",
    "            if preds[i] == 1:\n",
    "                val = round_df.iloc[i]['Team1ID']\n",
    "            else:\n",
    "                val = round_df.iloc[i]['Team2ID']\n",
    "                \n",
    "            round_dict['Team1ID'].append(val)\n",
    "            winners_arr.append(val)\n",
    "        else:\n",
    "            length = len(round_dict['Team2ID'])\n",
    "            val = 0\n",
    "            if preds[i] == 1:\n",
    "                val = round_df.iloc[i]['Team1ID']\n",
    "            else:\n",
    "                val = round_df.iloc[i]['Team2ID']\n",
    "                \n",
    "            round_dict['Team2ID'].append(val)\n",
    "            winners_arr.append(val)\n",
    "    for i in range(0, len(winners_arr)):\n",
    "        #print(f\"Predicted: {winners_arr[i]}  Real: {tournamentRounds['Round' + str(round_num)][i]}\")\n",
    "        if winners_arr[i] == tournamentRounds['Round' + str(round_num)][i]:\n",
    "            bracket_score += 10 * (2**(round_num - 1))\n",
    "            #print(bracket_score)\n",
    "    if round_num < 6:\n",
    "        new_df = pd.DataFrame(data=round_dict)\n",
    "        return tournament(classifier, new_df, round_num+1, bracket_score)\n",
    "    else:\n",
    "        return bracket_score\n",
    "\n",
    "start_time = time.time()\n",
    "\n",
    "svm_bracket_score = tournament(svm.SVC(C=4.6, gamma=0.01), bracket_df, 1, 0)\n",
    "print(f\"SVM Bracket Score (2022): {svm_bracket_score}\")\n",
    "print(f\"Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "\n",
    "\n",
    "#xgboost    \n",
    "#eta-[0.05, 0.06, 0.11, 0.13, .15]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM Avg Bracket Score 2010-2019: 443.3333333333333\n",
      "Runtime: 3.281 s\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#11\n",
    "# Predicts and calculates bracket score for previous years' tournaments\n",
    "\n",
    "def predictByYear(classifier, round_df, year, num_games):\n",
    "    \n",
    "    predicted_games = [0] * num_games\n",
    "    \n",
    "    test_data = round_df\n",
    "\n",
    "    training_data = train_DF[train_DF['Season'] != year]\n",
    "    y = training_data['Winner']\n",
    "\n",
    "    test_X = test_data.drop([\"Season\", \"Team1ID\", \"Team2ID\"], axis=1)\n",
    "    X = training_data.drop(\"Season\", axis=1)\n",
    "    X = X.drop([\"Winner\", \"Team1ID\", \"Team2ID\"], axis=1)\n",
    "    y = y.astype(int)\n",
    "\n",
    "    scaler = StandardScaler()\n",
    "    scaled_X = scaler.fit_transform(X.values)\n",
    "    scaled_test = scaler.fit_transform(test_X.values)\n",
    "    test_proc = pd.DataFrame(scaled_test, index = test_X.index, columns = test_X.columns)\n",
    "\n",
    "    X_proc = pd.DataFrame(scaled_X, index = X.index, columns = X.columns)\n",
    "\n",
    "    accuracy_scores = []\n",
    "    from sklearn.metrics import accuracy_score\n",
    "\n",
    "    # Model Fit\n",
    "    classifier.fit(X_proc, y)\n",
    "\n",
    "    # Model Prediction\n",
    "    preds = classifier.predict(test_proc)\n",
    "    return preds\n",
    "    \n",
    "def tourneyByYear(classifier, round_bracket, round_num, start_index, bracket_score, year):\n",
    "    round_df = pd.DataFrame(columns=test_cols)\n",
    "    for row in round_bracket.iterrows():\n",
    "        temp_list = [2022, row[1]['Team1ID'], row[1]['Team2ID']]\n",
    "        seeds_year_df = seeds_df[seeds_df['Season'] == year]\n",
    "        \n",
    "        #ordering new teams by seed\n",
    "        if round_num > 1:\n",
    "            #print(row[1])\n",
    "            team1_seed = seeds_year_df[seeds_year_df['TeamID'] == row[1][\"Team1ID\"]].values.tolist()[0][1]\n",
    "            team2_seed = seeds_year_df[seeds_year_df['TeamID'] == row[1][\"Team2ID\"]].values.tolist()[0][1]\n",
    "            team1_prefix = team1_seed[0:1]\n",
    "            team2_prefix = team2_seed[0:1]\n",
    "            \n",
    "            team1_seed = int(team1_seed[1:3])\n",
    "            team2_seed = int(team2_seed[1:3])\n",
    "            \n",
    "            if team1_seed > team2_seed:\n",
    "                temp = temp_list[1]\n",
    "                temp_list[1] = temp_list[2]\n",
    "                temp_list[2] = temp\n",
    "                \n",
    "                \n",
    "        #print(temp_list)\n",
    "        team1_row = team_avgs[team_avgs['Season'] == year]\n",
    "        team1_row = team_avgs[team_avgs[\"TeamID\"] == temp_list[1]].values.tolist()\n",
    "        \n",
    "        team2_row = team_avgs[team_avgs['Season'] == year]\n",
    "        team2_row = team_avgs[team_avgs[\"TeamID\"] == temp_list[2]].values.tolist()\n",
    "        team1_row[0].pop(0)\n",
    "        team1_row[0].pop(0)\n",
    "        team2_row[0].pop(0)\n",
    "        team2_row[0].pop(0)\n",
    "\n",
    "        temp_list.extend(team1_row[0])\n",
    "        temp_list.extend(team2_row[0])\n",
    "        # add that list to our DF\n",
    "        temp_series = pd.Series(temp_list, index=test_cols)\n",
    "        round_df = round_df.append(temp_series, ignore_index=True)\n",
    "        \n",
    "    preds = predictByYear(classifier, round_df, 1, len(round_df.index))\n",
    "    \n",
    "    round_dict = {'Team1ID': [], 'Team2ID': []}\n",
    "    winners_arr = []\n",
    "    for i in range(0, len(preds)):\n",
    "        if i % 2 == 0:\n",
    "            length = len(round_dict['Team1ID'])\n",
    "            val = 0\n",
    "            if preds[i] == 1:\n",
    "                val = round_df.iloc[i]['Team1ID']\n",
    "            else:\n",
    "                val = round_df.iloc[i]['Team2ID']\n",
    "                \n",
    "            round_dict['Team1ID'].append(val)\n",
    "            winners_arr.append(val)\n",
    "        else:\n",
    "            length = len(round_dict['Team2ID'])\n",
    "            val = 0\n",
    "            if preds[i] == 1:\n",
    "                val = round_df.iloc[i]['Team1ID']\n",
    "            else:\n",
    "                val = round_df.iloc[i]['Team2ID']\n",
    "                \n",
    "            round_dict['Team2ID'].append(val)\n",
    "            winners_arr.append(val)\n",
    "    tourney_year = trim_tourney_df[trim_tourney_df['Season'] == year].apply(pd.to_numeric)\n",
    "    for i in range(0, len(winners_arr)):\n",
    "        new_row = tourney_year.iloc[start_index + i]\n",
    "        if winners_arr[i] == new_row['WTeamID']:\n",
    "            bracket_score += 10 * (2**(round_num - 1))\n",
    "    if round_num < 6:\n",
    "        new_df = pd.DataFrame(data=round_dict)\n",
    "        return tourneyByYear(classifier, new_df, round_num+1, start_index + int(32 / (2**round_num-1)), bracket_score, year)\n",
    "    else:\n",
    "        return bracket_score\n",
    "        \n",
    "def tourneyHistorical(classifier):\n",
    "    historical_scores = []\n",
    "    start_time = time.time()\n",
    "    for year in years:\n",
    "        if(year > 2010):\n",
    "            old_round_df = train_DF[train_DF['Season'] == year]\n",
    "            old_round_df = old_round_df.head(32)\n",
    "            historical_scores.append(tourneyByYear(classifier, old_round_df,1,0,0,year))\n",
    "    return historical_scores\n",
    "\n",
    "start_time = time.time()\n",
    "svm_historical_scores = tourneyHistorical(svm.SVC(C=4.6, gamma=0.01))\n",
    "print(f\"SVM Avg Bracket Score 2010-2019: {sum(svm_historical_scores) / len(svm_historical_scores)}\")\n",
    "print(f\"Runtime: {time.time() - start_time:.3f} s\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Runtime: 0.076 s\n",
      "\n",
      "Average of 1000 coin flip brackets: 309.54\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'whiskers': [<matplotlib.lines.Line2D at 0x7f58796c97c0>,\n",
       "  <matplotlib.lines.Line2D at 0x7f58796c91f0>],\n",
       " 'caps': [<matplotlib.lines.Line2D at 0x7f58796c9760>,\n",
       "  <matplotlib.lines.Line2D at 0x7f58832571f0>],\n",
       " 'boxes': [<matplotlib.lines.Line2D at 0x7f58796c9f70>],\n",
       " 'medians': [<matplotlib.lines.Line2D at 0x7f5883257520>],\n",
       " 'fliers': [<matplotlib.lines.Line2D at 0x7f5879806250>],\n",
       " 'means': []}"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from random import randint\n",
    "import time\n",
    "\n",
    "#12\n",
    "# Demonstration of distribution of coin-flip brackets\n",
    "\n",
    "actual_winners = ( #Teams are given an index starting at the top left and ending at the bottom right\n",
    "      [], #placeholder so each round can have the right index\n",
    "      [0,3,5,6,9,10,12,14,16,18,20,22,24,26,28,31,32,35,36,38,41,42,44,46,48,51,53,54,57,58,61,62], #rd 1\n",
    "      [0,6,10,14,18,22,26,31,32,36,41,46,48,54,57,61], #rd 2\n",
    "      [6,14,18,31,36,46,48,61], #rd 3 (E8)\n",
    "      [14,18,46,48], #rd 4 (F4)\n",
    "      [18,48], #rd 5 (final)\n",
    "      [48]) #rd 6 (winner)\n",
    "\n",
    "# coin-flip bracket\n",
    "def pick_round(teams):\n",
    "    loser_list = []\n",
    "    \n",
    "    for i in range(0, len(teams) - 1, 2):\n",
    "        loser_list.append(i + randint(0,1)) # get a list of random losers\n",
    "        \n",
    "    loser_list.reverse() #reverse the list so it's in reverse numerical order (so we can remove by index)\n",
    "    for loser in loser_list:\n",
    "        teams.pop(loser)\n",
    "        \n",
    "    return teams\n",
    "\n",
    "def coin_flip(actual_winners):\n",
    "    sim_tourney = list(range(0, 64))\n",
    "    predicted_winners = [[]]\n",
    "    score = 0\n",
    "    \n",
    "    for rd in range(1,7):\n",
    "        sim_tourney = pick_round(sim_tourney)\n",
    "        predicted_winners.append(sim_tourney[:])\n",
    "    \n",
    "    for rd in range(1,7):\n",
    "        for game in range(0, len(predicted_winners[rd])):\n",
    "            score = (score + 10 * (2 ** (rd - 1))) if predicted_winners[rd][game] == actual_winners[rd][game] else score\n",
    "        \n",
    "    return score\n",
    "\n",
    "coin_flips = []\n",
    "\n",
    "start_time = time.time()\n",
    "\n",
    "# How many coin flip brackets would you like? Feel free to change it (it runs slowly [>10s] above 200,000)\n",
    "num_coin_flips = 1000\n",
    "for i in range(0, num_coin_flips):\n",
    "    coin_flips.append(coin_flip(actual_winners))\n",
    "\n",
    "print(f\"Runtime: {time.time() - start_time:.3f} s\\n\")\n",
    "\n",
    "print(f\"Average of {num_coin_flips} coin flip brackets: {sum(coin_flips) / len(coin_flips)}\")\n",
    "\n",
    "from matplotlib import pyplot\n",
    "\n",
    "pyplot.boxplot(coin_flips)\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Decision Tree Accuracy: 0.6032\n",
      "Decision Tree Runtime: 1.110 s\n",
      "\n",
      "Random Forest Accuracy: 0.6667\n",
      "Random Forest Runtime: 17.861 s\n",
      "\n",
      "Neural Networks Accuracy: 0.5079\n",
      "Neural Networks Runtime: 47.839 s\n",
      "\n",
      "XGB Accuracy: 0.6190\n",
      "XGB Runtime: 0.365 s\n",
      "\n",
      "Decision Tree Avg game-by-game accuracy across all games from 2010-2019: 0.6317\n",
      "Decision Tree Runtime: 0.14 s\n",
      "\n",
      "Random Forest Avg game-by-game accuracy across all games from 2010-2019: 0.6905\n",
      "Random Forest Runtime: 2.03 s\n",
      "\n",
      "Neural Networks Avg game-by-game accuracy across all games from 2010-2019: 0.6937\n",
      "Neural Networks Runtime: 6.59 s\n",
      "\n",
      "XGB Avg game-by-game accuracy across all games from 2010-2019: 0.6921\n",
      "XGB Runtime: 6.66 s\n"
     ]
    }
   ],
   "source": [
    "#13 \n",
    "# Other Model's game-by-game prediction accuracies for comparison\n",
    "start_time = time.time()\n",
    "dt_preds = trainAndPredict(DecisionTreeClassifier(), tourney_2022_df, 101, 63)\n",
    "print(f\"Decision Tree Accuracy: {accuracy(dt_preds):.4f}\")\n",
    "print(f\"Decision Tree Runtime: {time.time() - start_time:.3f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "rf_preds = trainAndPredict(RandomForestClassifier(), tourney_2022_df, 101, 63)\n",
    "print(f\"Random Forest Accuracy: {accuracy(rf_preds):.4f}\")\n",
    "print(f\"Random Forest Runtime: {time.time() - start_time:.3f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "nn_preds = trainAndPredict(MLPClassifier(), tourney_2022_df, 101, 63)\n",
    "print(f\"Neural Networks Accuracy: {accuracy(nn_preds):.4f}\")\n",
    "print(f\"Neural Networks Runtime: {time.time() - start_time:.3f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "xgb_preds = trainAndPredict(xgb.XGBClassifier(eta=0.11), tourney_2022_df, 1, 63)\n",
    "print(f\"XGB Accuracy: {accuracy(xgb_preds):.4f}\")\n",
    "print(f\"XGB Runtime: {time.time() - start_time:.3f} s\\n\")\n",
    "\n",
    "# historical 2010-2019\n",
    "start_time = time.time()\n",
    "accuracies = predictHistorical(DecisionTreeClassifier())\n",
    "print(f\"Decision Tree Avg game-by-game accuracy across all games from 2010-2019: {sum(accuracies) / len(accuracies):.4f}\")\n",
    "print(f\"Decision Tree Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "accuracies = predictHistorical(RandomForestClassifier())\n",
    "print(f\"Random Forest Avg game-by-game accuracy across all games from 2010-2019: {sum(accuracies) / len(accuracies):.4f}\")\n",
    "print(f\"Random Forest Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "accuracies = predictHistorical(MLPClassifier())\n",
    "print(f\"Neural Networks Avg game-by-game accuracy across all games from 2010-2019: {sum(accuracies) / len(accuracies):.4f}\")\n",
    "print(f\"Neural Networks Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "accuracies = predictHistorical(xgb.XGBClassifier(eta=0.11))\n",
    "print(f\"XGB Avg game-by-game accuracy across all games from 2010-2019: {sum(accuracies) / len(accuracies):.4f}\")\n",
    "print(f\"XGB Runtime: {time.time() - start_time:.2f} s\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Decision Tree Bracket Score (2022): 380\n",
      "Decision Tree Runtime: 0.19 s\n",
      "\n",
      "Random Forest Bracket Score (2022): 580\n",
      "Random Forest Runtime: 1.16 s\n",
      "\n",
      "Neural Networks Bracket Score (2022): 290\n",
      "Neural Networks Runtime: 3.14 s\n",
      "\n",
      "XGB Avg Bracket Bracket Score (2022): 1090\n",
      "XGB Avg Bracket Runtime: 2.63 s\n",
      "\n",
      "Decision Tree Avg Bracket Score 2010-2019: 335.55555555555554\n",
      "Decision Tree Runtime: 2.983 s\n",
      "\n",
      "Random Forest Avg Bracket Score 2010-2019: 430.0\n",
      "Random Forest Runtime: 13.979 s\n",
      "\n",
      "Neural Networks Avg Bracket Score 2010-2019: 326.6666666666667\n",
      "Neural Networks Runtime: 38.148 s\n",
      "\n",
      "XGB Avg Bracket Score 2010-2019: 411.1111111111111\n",
      "XGB Runtime: 24.521 s\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#14\n",
    "# Other Model's tournament prediction scores for comparison\n",
    "start_time = time.time()\n",
    "bracket_score_2022 = tournament(DecisionTreeClassifier(), bracket_df, 1, 0)\n",
    "print(f\"Decision Tree Bracket Score (2022): {bracket_score_2022}\")\n",
    "print(f\"Decision Tree Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "bracket_score_2022 = tournament(RandomForestClassifier(), bracket_df, 1, 0)\n",
    "print(f\"Random Forest Bracket Score (2022): {bracket_score_2022}\")\n",
    "print(f\"Random Forest Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "bracket_score_2022 = tournament(MLPClassifier(), bracket_df, 1, 0)\n",
    "print(f\"Neural Networks Bracket Score (2022): {bracket_score_2022}\")\n",
    "print(f\"Neural Networks Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "bracket_score_2022 = tournament(xgb.XGBClassifier(eta=0.11), bracket_df, 1, 0)\n",
    "print(f\"XGB Avg Bracket Bracket Score (2022): {bracket_score_2022}\")\n",
    "print(f\"XGB Avg Bracket Runtime: {time.time() - start_time:.2f} s\\n\")\n",
    "\n",
    "# historical tourney scores 2010-2019\n",
    "start_time = time.time()\n",
    "historical_scores = tourneyHistorical(DecisionTreeClassifier())\n",
    "print(f\"Decision Tree Avg Bracket Score 2010-2019: {sum(historical_scores) / len(historical_scores)}\")\n",
    "print(f\"Decision Tree Runtime: {time.time() - start_time:.3f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "historical_scores = tourneyHistorical(RandomForestClassifier())\n",
    "print(f\"Random Forest Avg Bracket Score 2010-2019: {sum(historical_scores) / len(historical_scores)}\")\n",
    "print(f\"Random Forest Runtime: {time.time() - start_time:.3f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "historical_scores = tourneyHistorical(MLPClassifier())\n",
    "print(f\"Neural Networks Avg Bracket Score 2010-2019: {sum(historical_scores) / len(historical_scores)}\")\n",
    "print(f\"Neural Networks Runtime: {time.time() - start_time:.3f} s\\n\")\n",
    "\n",
    "start_time = time.time()\n",
    "historical_scores = tourneyHistorical(xgb.XGBClassifier(eta=0.11))\n",
    "print(f\"XGB Avg Bracket Score 2010-2019: {sum(historical_scores) / len(historical_scores)}\")\n",
    "print(f\"XGB Runtime: {time.time() - start_time:.3f} s\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}