Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import fancyimpute\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.linear_model import LogisticRegressionCV\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "import xgboost as xgb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "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>Year</th>\n",
       "      <th>Name</th>\n",
       "      <th>Position</th>\n",
       "      <th>Round</th>\n",
       "      <th>Pick</th>\n",
       "      <th>College</th>\n",
       "      <th>Conference</th>\n",
       "      <th>Games</th>\n",
       "      <th>Seasons</th>\n",
       "      <th>Height</th>\n",
       "      <th>...</th>\n",
       "      <th>PR.TD</th>\n",
       "      <th>KRs</th>\n",
       "      <th>KR.Yds</th>\n",
       "      <th>KR.TD</th>\n",
       "      <th>XPA</th>\n",
       "      <th>XP.</th>\n",
       "      <th>FGA</th>\n",
       "      <th>FG.</th>\n",
       "      <th>Punts</th>\n",
       "      <th>Punt.Avg</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008</td>\n",
       "      <td>Husain Abdullah</td>\n",
       "      <td>FS</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Small</td>\n",
       "      <td>Pac-10</td>\n",
       "      <td>45</td>\n",
       "      <td>4</td>\n",
       "      <td>72</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2008</td>\n",
       "      <td>Jamar Adams</td>\n",
       "      <td>SS</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Michigan</td>\n",
       "      <td>Big Ten</td>\n",
       "      <td>38</td>\n",
       "      <td>3</td>\n",
       "      <td>74</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2008</td>\n",
       "      <td>Xavier Adibi</td>\n",
       "      <td>OLB</td>\n",
       "      <td>4</td>\n",
       "      <td>118</td>\n",
       "      <td>Virginia Tech</td>\n",
       "      <td>ACC</td>\n",
       "      <td>47</td>\n",
       "      <td>4</td>\n",
       "      <td>74</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2008</td>\n",
       "      <td>Erik Ainge</td>\n",
       "      <td>QB</td>\n",
       "      <td>5</td>\n",
       "      <td>162</td>\n",
       "      <td>Tennessee</td>\n",
       "      <td>SEC</td>\n",
       "      <td>43</td>\n",
       "      <td>4</td>\n",
       "      <td>77</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2008</td>\n",
       "      <td>Branden Albert</td>\n",
       "      <td>OG</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>Virginia</td>\n",
       "      <td>ACC</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 54 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Year             Name Position  Round  Pick        College Conference  \\\n",
       "0  2008  Husain Abdullah       FS      0     0          Small     Pac-10   \n",
       "1  2008      Jamar Adams       SS      0     0       Michigan    Big Ten   \n",
       "2  2008     Xavier Adibi      OLB      4   118  Virginia Tech        ACC   \n",
       "3  2008       Erik Ainge       QB      5   162      Tennessee        SEC   \n",
       "4  2008   Branden Albert       OG      1    15       Virginia        ACC   \n",
       "\n",
       "   Games  Seasons  Height    ...     PR.TD  KRs  KR.Yds  KR.TD  XPA  XP.  FGA  \\\n",
       "0     45        4      72    ...         0    0       0      0    0  0.0    0   \n",
       "1     38        3      74    ...         0    0       0      0    0  0.0    0   \n",
       "2     47        4      74    ...         0    0       0      0    0  0.0    0   \n",
       "3     43        4      77    ...         0    0       0      0    0  0.0    0   \n",
       "4      0        0      78    ...         0    0       0      0    0  0.0    0   \n",
       "\n",
       "   FG.  Punts  Punt.Avg  \n",
       "0  0.0      0       0.0  \n",
       "1  0.0      0       0.0  \n",
       "2  0.0      0       0.0  \n",
       "3  0.0      0       0.0  \n",
       "4  0.0      0       0.0  \n",
       "\n",
       "[5 rows x 54 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"data/full2.csv\")\n",
    "# Examine head of data\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Imputing row 1/3151 with 1 missing, elapsed time: 1.903\n",
      "Imputing row 101/3151 with 0 missing, elapsed time: 1.911\n",
      "Imputing row 201/3151 with 0 missing, elapsed time: 1.917\n",
      "Imputing row 301/3151 with 0 missing, elapsed time: 1.923\n",
      "Imputing row 401/3151 with 1 missing, elapsed time: 1.929\n",
      "Imputing row 501/3151 with 0 missing, elapsed time: 1.935\n",
      "Imputing row 601/3151 with 0 missing, elapsed time: 1.942\n",
      "Imputing row 701/3151 with 0 missing, elapsed time: 1.947\n",
      "Imputing row 801/3151 with 5 missing, elapsed time: 1.954\n",
      "Imputing row 901/3151 with 0 missing, elapsed time: 1.960\n",
      "Imputing row 1001/3151 with 0 missing, elapsed time: 1.965\n",
      "Imputing row 1101/3151 with 0 missing, elapsed time: 1.969\n",
      "Imputing row 1201/3151 with 1 missing, elapsed time: 1.974\n",
      "Imputing row 1301/3151 with 0 missing, elapsed time: 1.980\n",
      "Imputing row 1401/3151 with 5 missing, elapsed time: 1.986\n",
      "Imputing row 1501/3151 with 1 missing, elapsed time: 1.992\n",
      "Imputing row 1601/3151 with 3 missing, elapsed time: 2.002\n",
      "Imputing row 1701/3151 with 0 missing, elapsed time: 2.011\n",
      "Imputing row 1801/3151 with 2 missing, elapsed time: 2.016\n",
      "Imputing row 1901/3151 with 2 missing, elapsed time: 2.022\n",
      "Imputing row 2001/3151 with 0 missing, elapsed time: 2.029\n",
      "Imputing row 2101/3151 with 4 missing, elapsed time: 2.035\n",
      "Imputing row 2201/3151 with 1 missing, elapsed time: 2.041\n",
      "Imputing row 2301/3151 with 0 missing, elapsed time: 2.046\n",
      "Imputing row 2401/3151 with 0 missing, elapsed time: 2.054\n",
      "Imputing row 2501/3151 with 1 missing, elapsed time: 2.060\n",
      "Imputing row 2601/3151 with 6 missing, elapsed time: 2.065\n",
      "Imputing row 2701/3151 with 5 missing, elapsed time: 2.073\n",
      "Imputing row 2801/3151 with 6 missing, elapsed time: 2.079\n",
      "Imputing row 2901/3151 with 0 missing, elapsed time: 2.085\n",
      "Imputing row 3001/3151 with 2 missing, elapsed time: 2.091\n",
      "Imputing row 3101/3151 with 0 missing, elapsed time: 2.098\n"
     ]
    }
   ],
   "source": [
    "# We know that combine data is missing, so let's impute it.\n",
    "imp = df[[\"Position\", \"Height\", \"Weight\", \"X40.Yard\", \"Bench\", \"Broad.Jump\", \"Shuttle\", \"X3.Cone\", \"Vertical\"]]\n",
    "imp=imp.replace({0:np.nan})\n",
    "number = LabelEncoder()\n",
    "imp[\"Position\"] = number.fit_transform(imp[\"Position\"])\n",
    "imp=fancyimpute.KNN(k=5).complete(imp)\n",
    "imp = pd.DataFrame(imp)\n",
    "imp.columns = [\"Position\", \"Height\", \"Weight\", \"X40.Yard\", \"Bench\", \"Broad.Jump\", \"Shuttle\", \"X3.Cone\", \"Vertical\"]\n",
    "imp = imp.round({\"X40.Yard\": 2, \"Bench\": 0, \"Broad.Jump\": 0, \"X3.Cone\": 2, \"Shuttle\": 2})\n",
    "imp[\"Vertical\"] = round(imp[\"Vertical\"] * 2) / 2\n",
    "df[[\"Height\", \"Weight\", \"X40.Yard\", \"Bench\", \"Broad.Jump\", \"Shuttle\", \"X3.Cone\", \"Vertical\"]] = imp.drop(\"Position\", axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Imputing row 1/3151 with 0 missing, elapsed time: 1.367\n",
      "Imputing row 101/3151 with 0 missing, elapsed time: 1.368\n",
      "Imputing row 201/3151 with 2 missing, elapsed time: 1.369\n",
      "Imputing row 301/3151 with 0 missing, elapsed time: 1.370\n",
      "Imputing row 401/3151 with 0 missing, elapsed time: 1.372\n",
      "Imputing row 501/3151 with 0 missing, elapsed time: 1.374\n",
      "Imputing row 601/3151 with 2 missing, elapsed time: 1.375\n",
      "Imputing row 701/3151 with 0 missing, elapsed time: 1.376\n",
      "Imputing row 801/3151 with 0 missing, elapsed time: 1.378\n",
      "Imputing row 901/3151 with 0 missing, elapsed time: 1.380\n",
      "Imputing row 1001/3151 with 0 missing, elapsed time: 1.381\n",
      "Imputing row 1101/3151 with 0 missing, elapsed time: 1.383\n",
      "Imputing row 1201/3151 with 0 missing, elapsed time: 1.384\n",
      "Imputing row 1301/3151 with 0 missing, elapsed time: 1.385\n",
      "Imputing row 1401/3151 with 0 missing, elapsed time: 1.387\n",
      "Imputing row 1501/3151 with 0 missing, elapsed time: 1.388\n",
      "Imputing row 1601/3151 with 0 missing, elapsed time: 1.389\n",
      "Imputing row 1701/3151 with 0 missing, elapsed time: 1.389\n",
      "Imputing row 1801/3151 with 0 missing, elapsed time: 1.390\n",
      "Imputing row 1901/3151 with 0 missing, elapsed time: 1.393\n",
      "Imputing row 2001/3151 with 0 missing, elapsed time: 1.394\n",
      "Imputing row 2101/3151 with 0 missing, elapsed time: 1.398\n",
      "Imputing row 2201/3151 with 0 missing, elapsed time: 1.400\n",
      "Imputing row 2301/3151 with 0 missing, elapsed time: 1.402\n",
      "Imputing row 2401/3151 with 0 missing, elapsed time: 1.404\n",
      "Imputing row 2501/3151 with 0 missing, elapsed time: 1.405\n",
      "Imputing row 2601/3151 with 0 missing, elapsed time: 1.407\n",
      "Imputing row 2701/3151 with 0 missing, elapsed time: 1.408\n",
      "Imputing row 2801/3151 with 0 missing, elapsed time: 1.410\n",
      "Imputing row 2901/3151 with 0 missing, elapsed time: 1.411\n",
      "Imputing row 3001/3151 with 0 missing, elapsed time: 1.412\n",
      "Imputing row 3101/3151 with 0 missing, elapsed time: 1.416\n"
     ]
    }
   ],
   "source": [
    "# Impute missing games and season values\n",
    "imp = df[[\"Position\", \"Games\", \"Seasons\"]]\n",
    "imp=imp.replace({0:np.nan})\n",
    "imp[\"Position\"] = number.fit_transform(imp[\"Position\"])\n",
    "imp=fancyimpute.KNN(k=10).complete(imp)\n",
    "imp = pd.DataFrame(imp)\n",
    "imp = imp.round(0)\n",
    "df[[\"Games\", \"Seasons\"]] = imp.drop(0, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fix Position Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Position\n",
       "C        96\n",
       "CB      342\n",
       "DB        2\n",
       "DE      246\n",
       "DT      247\n",
       "EDGE     19\n",
       "FB       45\n",
       "FS      102\n",
       "ILB     153\n",
       "K        46\n",
       "LB        2\n",
       "LS       14\n",
       "NT        3\n",
       "OG      125\n",
       "OL        2\n",
       "OLB     241\n",
       "OT      175\n",
       "P        60\n",
       "QB      177\n",
       "RB      295\n",
       "S        24\n",
       "SS      110\n",
       "TE      166\n",
       "WR      459\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's get some information about the positions.\n",
    "df.groupby(\"Position\").size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "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>Year</th>\n",
       "      <th>Name</th>\n",
       "      <th>Position</th>\n",
       "      <th>Round</th>\n",
       "      <th>Pick</th>\n",
       "      <th>College</th>\n",
       "      <th>Conference</th>\n",
       "      <th>Games</th>\n",
       "      <th>Seasons</th>\n",
       "      <th>Height</th>\n",
       "      <th>...</th>\n",
       "      <th>PR.TD</th>\n",
       "      <th>KRs</th>\n",
       "      <th>KR.Yds</th>\n",
       "      <th>KR.TD</th>\n",
       "      <th>XPA</th>\n",
       "      <th>XP.</th>\n",
       "      <th>FGA</th>\n",
       "      <th>FG.</th>\n",
       "      <th>Punts</th>\n",
       "      <th>Punt.Avg</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2954</th>\n",
       "      <td>2018</td>\n",
       "      <td>Minkah Fitzpatrick</td>\n",
       "      <td>DB</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>SEC</td>\n",
       "      <td>42.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3039</th>\n",
       "      <td>2018</td>\n",
       "      <td>Quenton Meeks</td>\n",
       "      <td>DB</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Stanford</td>\n",
       "      <td>Pac-12</td>\n",
       "      <td>36.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>74.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 54 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Year                Name Position  Round  Pick   College Conference  \\\n",
       "2954  2018  Minkah Fitzpatrick       DB      1    11   Alabama        SEC   \n",
       "3039  2018       Quenton Meeks       DB      0     0  Stanford     Pac-12   \n",
       "\n",
       "      Games  Seasons  Height    ...     PR.TD  KRs  KR.Yds  KR.TD  XPA  XP.  \\\n",
       "2954   42.0      3.0    73.0    ...         0    1      16      1    0  0.0   \n",
       "3039   36.0      3.0    74.0    ...         0    0       0      0    0  0.0   \n",
       "\n",
       "      FGA  FG.  Punts  Punt.Avg  \n",
       "2954    0  0.0      0       0.0  \n",
       "3039    0  0.0      0       0.0  \n",
       "\n",
       "[2 rows x 54 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.Position==\"DB\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fitzpatrick will play mostly as a free safety and Meeks as a CB.\n",
    "df.at[2954, \"Position\"] = \"FS\"\n",
    "df.at[3039, \"Position\"] = \"CB\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Year</th>\n",
       "      <th>Name</th>\n",
       "      <th>Position</th>\n",
       "      <th>Round</th>\n",
       "      <th>Pick</th>\n",
       "      <th>College</th>\n",
       "      <th>Conference</th>\n",
       "      <th>Games</th>\n",
       "      <th>Seasons</th>\n",
       "      <th>Height</th>\n",
       "      <th>...</th>\n",
       "      <th>PR.TD</th>\n",
       "      <th>KRs</th>\n",
       "      <th>KR.Yds</th>\n",
       "      <th>KR.TD</th>\n",
       "      <th>XPA</th>\n",
       "      <th>XP.</th>\n",
       "      <th>FGA</th>\n",
       "      <th>FG.</th>\n",
       "      <th>Punts</th>\n",
       "      <th>Punt.Avg</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3030</th>\n",
       "      <td>2018</td>\n",
       "      <td>Cole Madison</td>\n",
       "      <td>OL</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Small</td>\n",
       "      <td>Pac-12</td>\n",
       "      <td>40.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>77.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3073</th>\n",
       "      <td>2018</td>\n",
       "      <td>Martinas Rankin</td>\n",
       "      <td>OL</td>\n",
       "      <td>3</td>\n",
       "      <td>80</td>\n",
       "      <td>Small</td>\n",
       "      <td>SEC</td>\n",
       "      <td>41.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>76.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 54 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Year             Name Position  Round  Pick College Conference  Games  \\\n",
       "3030  2018     Cole Madison       OL      0     0   Small     Pac-12   40.0   \n",
       "3073  2018  Martinas Rankin       OL      3    80   Small        SEC   41.0   \n",
       "\n",
       "      Seasons  Height    ...     PR.TD  KRs  KR.Yds  KR.TD  XPA  XP.  FGA  \\\n",
       "3030      4.0    77.0    ...         0    0       0      0    0  0.0    0   \n",
       "3073      4.0    76.0    ...         0    0       0      0    0  0.0    0   \n",
       "\n",
       "      FG.  Punts  Punt.Avg  \n",
       "3030  0.0      0       0.0  \n",
       "3073  0.0      0       0.0  \n",
       "\n",
       "[2 rows x 54 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.Position==\"OL\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Madison is a guard and Rankin is a tackle.\n",
    "df.at[3030, \"Position\"] = \"OG\"\n",
    "df.at[3073, \"Position\"] = \"OT\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[df.Position==\"LB\", \"Position\"] = \"OLB\"\n",
    "df.loc[df.Position==\"NT\", \"Position\"] = \"DT\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegressionCV(Cs=10, class_weight=None, cv=None, dual=False,\n",
       "           fit_intercept=True, intercept_scaling=1.0, max_iter=100,\n",
       "           multi_class='ovr', n_jobs=1, penalty='l2', random_state=None,\n",
       "           refit=True, scoring=None, solver='lbfgs', tol=0.0001, verbose=0)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We next need to divide free safeties and strong safeties in our 2018 data. Let's\n",
    "# examine what our safeties look like in our data. Let's use a simple logistic\n",
    "# regression model based on Height, Weight, Total Tackles, and Interceptions.\n",
    "s = df[(df.Position==\"FS\") | (df.Position==\"SS\")]\n",
    "s = s[[\"Position\", \"Height\", \"Weight\", \"Total.Tackles\", \"Int\", \"Sacks\"]]\n",
    "s.head()\n",
    "s_predict = LogisticRegressionCV()\n",
    "s_predict.fit(s.drop(\"Position\", 1), s.Position)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "safeties = df[df.Position==\"S\"]\n",
    "df.loc[safeties.index, \"Position\"] = s_predict.predict(safeties[[\"Height\", \"Weight\", \"Total.Tackles\", \"Int\", \"Sacks\"]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's do the same for edge rushers, forecasting their position by height and weight differences,\n",
    "# as well as total tackles\n",
    "outside = df[(df.Position==\"OLB\") | (df.Position==\"DE\")]\n",
    "outside = outside[[\"Position\", \"Height\", \"Weight\", \"Total.Tackles\"]]\n",
    "outside.head()\n",
    "edge_predict = LogisticRegressionCV()\n",
    "edge_predict.fit(outside.drop(\"Position\", 1), outside.Position)\n",
    "edge = df[df.Position==\"EDGE\"]\n",
    "df.loc[edge.index, \"Position\"] = edge_predict.predict(edge[[\"Height\", \"Weight\", \"Total.Tackles\"]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploratory Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Round</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Position</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>C</th>\n",
       "      <td>0.034156</td>\n",
       "      <td>0.023669</td>\n",
       "      <td>0.028481</td>\n",
       "      <td>0.026549</td>\n",
       "      <td>0.039275</td>\n",
       "      <td>0.013699</td>\n",
       "      <td>0.035714</td>\n",
       "      <td>0.034934</td>\n",
       "      <td>0.030467</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CB</th>\n",
       "      <td>0.083491</td>\n",
       "      <td>0.130178</td>\n",
       "      <td>0.136076</td>\n",
       "      <td>0.147493</td>\n",
       "      <td>0.108761</td>\n",
       "      <td>0.136986</td>\n",
       "      <td>0.095238</td>\n",
       "      <td>0.078603</td>\n",
       "      <td>0.108854</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DE</th>\n",
       "      <td>0.062619</td>\n",
       "      <td>0.121302</td>\n",
       "      <td>0.094937</td>\n",
       "      <td>0.097345</td>\n",
       "      <td>0.075529</td>\n",
       "      <td>0.051370</td>\n",
       "      <td>0.083333</td>\n",
       "      <td>0.096070</td>\n",
       "      <td>0.080292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DT</th>\n",
       "      <td>0.063567</td>\n",
       "      <td>0.100592</td>\n",
       "      <td>0.104430</td>\n",
       "      <td>0.109145</td>\n",
       "      <td>0.063444</td>\n",
       "      <td>0.065068</td>\n",
       "      <td>0.075397</td>\n",
       "      <td>0.087336</td>\n",
       "      <td>0.079340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FB</th>\n",
       "      <td>0.018027</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002950</td>\n",
       "      <td>0.012085</td>\n",
       "      <td>0.023973</td>\n",
       "      <td>0.019841</td>\n",
       "      <td>0.039301</td>\n",
       "      <td>0.014281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FS</th>\n",
       "      <td>0.044592</td>\n",
       "      <td>0.026627</td>\n",
       "      <td>0.037975</td>\n",
       "      <td>0.038348</td>\n",
       "      <td>0.030211</td>\n",
       "      <td>0.037671</td>\n",
       "      <td>0.043651</td>\n",
       "      <td>0.026201</td>\n",
       "      <td>0.037766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ILB</th>\n",
       "      <td>0.052182</td>\n",
       "      <td>0.026627</td>\n",
       "      <td>0.047468</td>\n",
       "      <td>0.038348</td>\n",
       "      <td>0.057402</td>\n",
       "      <td>0.061644</td>\n",
       "      <td>0.051587</td>\n",
       "      <td>0.048035</td>\n",
       "      <td>0.048556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>K</th>\n",
       "      <td>0.032258</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.003165</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.003021</td>\n",
       "      <td>0.013699</td>\n",
       "      <td>0.011905</td>\n",
       "      <td>0.013100</td>\n",
       "      <td>0.014599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LS</th>\n",
       "      <td>0.008539</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.003425</td>\n",
       "      <td>0.015873</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.004443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OG</th>\n",
       "      <td>0.034156</td>\n",
       "      <td>0.038462</td>\n",
       "      <td>0.031646</td>\n",
       "      <td>0.047198</td>\n",
       "      <td>0.045317</td>\n",
       "      <td>0.065068</td>\n",
       "      <td>0.039683</td>\n",
       "      <td>0.030568</td>\n",
       "      <td>0.039987</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OLB</th>\n",
       "      <td>0.063567</td>\n",
       "      <td>0.085799</td>\n",
       "      <td>0.098101</td>\n",
       "      <td>0.088496</td>\n",
       "      <td>0.108761</td>\n",
       "      <td>0.095890</td>\n",
       "      <td>0.067460</td>\n",
       "      <td>0.074236</td>\n",
       "      <td>0.080927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OT</th>\n",
       "      <td>0.033207</td>\n",
       "      <td>0.118343</td>\n",
       "      <td>0.069620</td>\n",
       "      <td>0.064897</td>\n",
       "      <td>0.051360</td>\n",
       "      <td>0.065068</td>\n",
       "      <td>0.027778</td>\n",
       "      <td>0.061135</td>\n",
       "      <td>0.055855</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>P</th>\n",
       "      <td>0.043643</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002950</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.017123</td>\n",
       "      <td>0.023810</td>\n",
       "      <td>0.008734</td>\n",
       "      <td>0.019042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>QB</th>\n",
       "      <td>0.064516</td>\n",
       "      <td>0.091716</td>\n",
       "      <td>0.034810</td>\n",
       "      <td>0.038348</td>\n",
       "      <td>0.042296</td>\n",
       "      <td>0.041096</td>\n",
       "      <td>0.067460</td>\n",
       "      <td>0.048035</td>\n",
       "      <td>0.056173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RB</th>\n",
       "      <td>0.104364</td>\n",
       "      <td>0.068047</td>\n",
       "      <td>0.085443</td>\n",
       "      <td>0.067847</td>\n",
       "      <td>0.102719</td>\n",
       "      <td>0.095890</td>\n",
       "      <td>0.119048</td>\n",
       "      <td>0.087336</td>\n",
       "      <td>0.093621</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SS</th>\n",
       "      <td>0.040797</td>\n",
       "      <td>0.029586</td>\n",
       "      <td>0.031646</td>\n",
       "      <td>0.017699</td>\n",
       "      <td>0.048338</td>\n",
       "      <td>0.041096</td>\n",
       "      <td>0.027778</td>\n",
       "      <td>0.061135</td>\n",
       "      <td>0.037448</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TE</th>\n",
       "      <td>0.045541</td>\n",
       "      <td>0.026627</td>\n",
       "      <td>0.056962</td>\n",
       "      <td>0.073746</td>\n",
       "      <td>0.060423</td>\n",
       "      <td>0.054795</td>\n",
       "      <td>0.051587</td>\n",
       "      <td>0.074236</td>\n",
       "      <td>0.052682</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WR</th>\n",
       "      <td>0.170778</td>\n",
       "      <td>0.112426</td>\n",
       "      <td>0.139241</td>\n",
       "      <td>0.138643</td>\n",
       "      <td>0.151057</td>\n",
       "      <td>0.116438</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.131004</td>\n",
       "      <td>0.145668</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Round            0         1         2         3         4         5  \\\n",
       "Position                                                               \n",
       "C         0.034156  0.023669  0.028481  0.026549  0.039275  0.013699   \n",
       "CB        0.083491  0.130178  0.136076  0.147493  0.108761  0.136986   \n",
       "DE        0.062619  0.121302  0.094937  0.097345  0.075529  0.051370   \n",
       "DT        0.063567  0.100592  0.104430  0.109145  0.063444  0.065068   \n",
       "FB        0.018027  0.000000  0.000000  0.002950  0.012085  0.023973   \n",
       "FS        0.044592  0.026627  0.037975  0.038348  0.030211  0.037671   \n",
       "ILB       0.052182  0.026627  0.047468  0.038348  0.057402  0.061644   \n",
       "K         0.032258  0.000000  0.003165  0.000000  0.003021  0.013699   \n",
       "LS        0.008539  0.000000  0.000000  0.000000  0.000000  0.003425   \n",
       "OG        0.034156  0.038462  0.031646  0.047198  0.045317  0.065068   \n",
       "OLB       0.063567  0.085799  0.098101  0.088496  0.108761  0.095890   \n",
       "OT        0.033207  0.118343  0.069620  0.064897  0.051360  0.065068   \n",
       "P         0.043643  0.000000  0.000000  0.002950  0.000000  0.017123   \n",
       "QB        0.064516  0.091716  0.034810  0.038348  0.042296  0.041096   \n",
       "RB        0.104364  0.068047  0.085443  0.067847  0.102719  0.095890   \n",
       "SS        0.040797  0.029586  0.031646  0.017699  0.048338  0.041096   \n",
       "TE        0.045541  0.026627  0.056962  0.073746  0.060423  0.054795   \n",
       "WR        0.170778  0.112426  0.139241  0.138643  0.151057  0.116438   \n",
       "All       1.000000  1.000000  1.000000  1.000000  1.000000  1.000000   \n",
       "\n",
       "Round            6         7       All  \n",
       "Position                                \n",
       "C         0.035714  0.034934  0.030467  \n",
       "CB        0.095238  0.078603  0.108854  \n",
       "DE        0.083333  0.096070  0.080292  \n",
       "DT        0.075397  0.087336  0.079340  \n",
       "FB        0.019841  0.039301  0.014281  \n",
       "FS        0.043651  0.026201  0.037766  \n",
       "ILB       0.051587  0.048035  0.048556  \n",
       "K         0.011905  0.013100  0.014599  \n",
       "LS        0.015873  0.000000  0.004443  \n",
       "OG        0.039683  0.030568  0.039987  \n",
       "OLB       0.067460  0.074236  0.080927  \n",
       "OT        0.027778  0.061135  0.055855  \n",
       "P         0.023810  0.008734  0.019042  \n",
       "QB        0.067460  0.048035  0.056173  \n",
       "RB        0.119048  0.087336  0.093621  \n",
       "SS        0.027778  0.061135  0.037448  \n",
       "TE        0.051587  0.074236  0.052682  \n",
       "WR        0.142857  0.131004  0.145668  \n",
       "All       1.000000  1.000000  1.000000  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# All of our \"hybrid\" positions have been dealt with. Let's now examine preliminary trends.\n",
    "round_info = pd.crosstab(index=df[\"Position\"], columns=df[\"Round\"], margins=True) \n",
    "round_info / round_info.loc[\"All\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As can be seen, there are a relatively high proportion of WRs, OTs, CBs, DEs, and DTs in the first round. In later rounds, More RBs and skill positions are taken. Let's now examine the proportions of rounds each position is typically drafted in."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Round</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Position</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>C</th>\n",
       "      <td>0.375000</td>\n",
       "      <td>0.083333</td>\n",
       "      <td>0.093750</td>\n",
       "      <td>0.093750</td>\n",
       "      <td>0.135417</td>\n",
       "      <td>0.041667</td>\n",
       "      <td>0.093750</td>\n",
       "      <td>0.083333</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CB</th>\n",
       "      <td>0.256560</td>\n",
       "      <td>0.128280</td>\n",
       "      <td>0.125364</td>\n",
       "      <td>0.145773</td>\n",
       "      <td>0.104956</td>\n",
       "      <td>0.116618</td>\n",
       "      <td>0.069971</td>\n",
       "      <td>0.052478</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DE</th>\n",
       "      <td>0.260870</td>\n",
       "      <td>0.162055</td>\n",
       "      <td>0.118577</td>\n",
       "      <td>0.130435</td>\n",
       "      <td>0.098814</td>\n",
       "      <td>0.059289</td>\n",
       "      <td>0.083004</td>\n",
       "      <td>0.086957</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DT</th>\n",
       "      <td>0.268000</td>\n",
       "      <td>0.136000</td>\n",
       "      <td>0.132000</td>\n",
       "      <td>0.148000</td>\n",
       "      <td>0.084000</td>\n",
       "      <td>0.076000</td>\n",
       "      <td>0.076000</td>\n",
       "      <td>0.080000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FB</th>\n",
       "      <td>0.422222</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.022222</td>\n",
       "      <td>0.088889</td>\n",
       "      <td>0.155556</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FS</th>\n",
       "      <td>0.394958</td>\n",
       "      <td>0.075630</td>\n",
       "      <td>0.100840</td>\n",
       "      <td>0.109244</td>\n",
       "      <td>0.084034</td>\n",
       "      <td>0.092437</td>\n",
       "      <td>0.092437</td>\n",
       "      <td>0.050420</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ILB</th>\n",
       "      <td>0.359477</td>\n",
       "      <td>0.058824</td>\n",
       "      <td>0.098039</td>\n",
       "      <td>0.084967</td>\n",
       "      <td>0.124183</td>\n",
       "      <td>0.117647</td>\n",
       "      <td>0.084967</td>\n",
       "      <td>0.071895</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>K</th>\n",
       "      <td>0.739130</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.021739</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.021739</td>\n",
       "      <td>0.086957</td>\n",
       "      <td>0.065217</td>\n",
       "      <td>0.065217</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LS</th>\n",
       "      <td>0.642857</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.071429</td>\n",
       "      <td>0.285714</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OG</th>\n",
       "      <td>0.285714</td>\n",
       "      <td>0.103175</td>\n",
       "      <td>0.079365</td>\n",
       "      <td>0.126984</td>\n",
       "      <td>0.119048</td>\n",
       "      <td>0.150794</td>\n",
       "      <td>0.079365</td>\n",
       "      <td>0.055556</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OLB</th>\n",
       "      <td>0.262745</td>\n",
       "      <td>0.113725</td>\n",
       "      <td>0.121569</td>\n",
       "      <td>0.117647</td>\n",
       "      <td>0.141176</td>\n",
       "      <td>0.109804</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OT</th>\n",
       "      <td>0.198864</td>\n",
       "      <td>0.227273</td>\n",
       "      <td>0.125000</td>\n",
       "      <td>0.125000</td>\n",
       "      <td>0.096591</td>\n",
       "      <td>0.107955</td>\n",
       "      <td>0.039773</td>\n",
       "      <td>0.079545</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>P</th>\n",
       "      <td>0.766667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.016667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.083333</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>QB</th>\n",
       "      <td>0.384181</td>\n",
       "      <td>0.175141</td>\n",
       "      <td>0.062147</td>\n",
       "      <td>0.073446</td>\n",
       "      <td>0.079096</td>\n",
       "      <td>0.067797</td>\n",
       "      <td>0.096045</td>\n",
       "      <td>0.062147</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RB</th>\n",
       "      <td>0.372881</td>\n",
       "      <td>0.077966</td>\n",
       "      <td>0.091525</td>\n",
       "      <td>0.077966</td>\n",
       "      <td>0.115254</td>\n",
       "      <td>0.094915</td>\n",
       "      <td>0.101695</td>\n",
       "      <td>0.067797</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SS</th>\n",
       "      <td>0.364407</td>\n",
       "      <td>0.084746</td>\n",
       "      <td>0.084746</td>\n",
       "      <td>0.050847</td>\n",
       "      <td>0.135593</td>\n",
       "      <td>0.101695</td>\n",
       "      <td>0.059322</td>\n",
       "      <td>0.118644</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TE</th>\n",
       "      <td>0.289157</td>\n",
       "      <td>0.054217</td>\n",
       "      <td>0.108434</td>\n",
       "      <td>0.150602</td>\n",
       "      <td>0.120482</td>\n",
       "      <td>0.096386</td>\n",
       "      <td>0.078313</td>\n",
       "      <td>0.102410</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WR</th>\n",
       "      <td>0.392157</td>\n",
       "      <td>0.082789</td>\n",
       "      <td>0.095861</td>\n",
       "      <td>0.102397</td>\n",
       "      <td>0.108932</td>\n",
       "      <td>0.074074</td>\n",
       "      <td>0.078431</td>\n",
       "      <td>0.065359</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>0.334497</td>\n",
       "      <td>0.107268</td>\n",
       "      <td>0.100286</td>\n",
       "      <td>0.107585</td>\n",
       "      <td>0.105046</td>\n",
       "      <td>0.092669</td>\n",
       "      <td>0.079975</td>\n",
       "      <td>0.072675</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Round            0         1         2         3         4         5  \\\n",
       "Position                                                               \n",
       "C         0.375000  0.083333  0.093750  0.093750  0.135417  0.041667   \n",
       "CB        0.256560  0.128280  0.125364  0.145773  0.104956  0.116618   \n",
       "DE        0.260870  0.162055  0.118577  0.130435  0.098814  0.059289   \n",
       "DT        0.268000  0.136000  0.132000  0.148000  0.084000  0.076000   \n",
       "FB        0.422222  0.000000  0.000000  0.022222  0.088889  0.155556   \n",
       "FS        0.394958  0.075630  0.100840  0.109244  0.084034  0.092437   \n",
       "ILB       0.359477  0.058824  0.098039  0.084967  0.124183  0.117647   \n",
       "K         0.739130  0.000000  0.021739  0.000000  0.021739  0.086957   \n",
       "LS        0.642857  0.000000  0.000000  0.000000  0.000000  0.071429   \n",
       "OG        0.285714  0.103175  0.079365  0.126984  0.119048  0.150794   \n",
       "OLB       0.262745  0.113725  0.121569  0.117647  0.141176  0.109804   \n",
       "OT        0.198864  0.227273  0.125000  0.125000  0.096591  0.107955   \n",
       "P         0.766667  0.000000  0.000000  0.016667  0.000000  0.083333   \n",
       "QB        0.384181  0.175141  0.062147  0.073446  0.079096  0.067797   \n",
       "RB        0.372881  0.077966  0.091525  0.077966  0.115254  0.094915   \n",
       "SS        0.364407  0.084746  0.084746  0.050847  0.135593  0.101695   \n",
       "TE        0.289157  0.054217  0.108434  0.150602  0.120482  0.096386   \n",
       "WR        0.392157  0.082789  0.095861  0.102397  0.108932  0.074074   \n",
       "All       0.334497  0.107268  0.100286  0.107585  0.105046  0.092669   \n",
       "\n",
       "Round            6         7  All  \n",
       "Position                           \n",
       "C         0.093750  0.083333  1.0  \n",
       "CB        0.069971  0.052478  1.0  \n",
       "DE        0.083004  0.086957  1.0  \n",
       "DT        0.076000  0.080000  1.0  \n",
       "FB        0.111111  0.200000  1.0  \n",
       "FS        0.092437  0.050420  1.0  \n",
       "ILB       0.084967  0.071895  1.0  \n",
       "K         0.065217  0.065217  1.0  \n",
       "LS        0.285714  0.000000  1.0  \n",
       "OG        0.079365  0.055556  1.0  \n",
       "OLB       0.066667  0.066667  1.0  \n",
       "OT        0.039773  0.079545  1.0  \n",
       "P         0.100000  0.033333  1.0  \n",
       "QB        0.096045  0.062147  1.0  \n",
       "RB        0.101695  0.067797  1.0  \n",
       "SS        0.059322  0.118644  1.0  \n",
       "TE        0.078313  0.102410  1.0  \n",
       "WR        0.078431  0.065359  1.0  \n",
       "All       0.079975  0.072675  1.0  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round_info = pd.crosstab(index=df[\"Position\"], columns=df[\"Round\"], margins=True) \n",
    "round_info.div(round_info[\"All\"], axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "At least 1/3 of players in most positions are undrafted; OTs have a significantly low amount undrafted, however. Most OTs are picked in the first round, with mostly decreasing numbers after that. We can see that positions such as kickers, punters, and fullbacks are primarily drafted in the late rounds."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Simple Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For this, we will run a \"one versus all\" regression to predict what players will go in the first round."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[df.Round != 1, \"Round\"] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_X = df.drop([\"Name\", \"Round\", \"Pick\", \"College\", \"Conference\"], 1)\n",
    "all_X = pd.get_dummies(all_X)\n",
    "\n",
    "train_X = all_X[(all_X.Year != 2018)].drop(\"Year\", 1)\n",
    "test_X = all_X[all_X.Year == 2018].drop(\"Year\", 1)\n",
    "train_y = df[(df.Year != 2018)].Round\n",
    "test_y = df[df.Year == 2018].Round"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegressionCV(Cs=10, class_weight=None, cv=None, dual=False,\n",
       "           fit_intercept=True, intercept_scaling=1.0, max_iter=100,\n",
       "           multi_class='ovr', n_jobs=1, penalty='l2', random_state=None,\n",
       "           refit=True, scoring=None, solver='lbfgs', tol=0.0001, verbose=0)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log = LogisticRegressionCV()\n",
    "log.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also use XGBoost."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "XGB = xgb.XGBClassifier(colsample_bytree=0.7,\n",
    " eta= 0.001,\n",
    " eval_metric= 'mae',\n",
    " max_depth= 6,\n",
    " min_child_weight= 15,\n",
    " objective= 'binary:logistic',\n",
    " subsample= 0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "       colsample_bytree=0.7, eta=0.001, eval_metric='mae', gamma=0,\n",
       "       learning_rate=0.1, max_delta_step=0, max_depth=6,\n",
       "       min_child_weight=15, missing=None, n_estimators=100, n_jobs=1,\n",
       "       nthread=None, objective='binary:logistic', random_state=0,\n",
       "       reg_alpha=0, reg_lambda=1, scale_pos_weight=1, seed=None,\n",
       "       silent=True, subsample=0.7)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "XGB.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "            min_impurity_split=1e-07, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            n_estimators=1000, n_jobs=1, oob_score=False,\n",
       "            random_state=None, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf = RandomForestClassifier(n_estimators=1000)\n",
    "clf.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "preds = 0.4 * clf.predict_proba(test_X) + 0.4 * XGB.predict_proba(test_X) + 0.2 * log.predict_proba(test_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 Kolton Miller\n",
      "2 Mike McGlinchey\n",
      "3 Saquon Barkley\n",
      "4 Harold Landry\n",
      "5 Terrell Edmunds\n",
      "6 David Bright\n",
      "7 Lamar Jackson\n",
      "8 Justin Reid\n",
      "9 Bradley Chubb\n",
      "10 Derwin James\n",
      "11 Martinas Rankin\n",
      "12 Frank Ragnow\n",
      "13 Josh Allen\n",
      "14 Joseph Noteboom\n",
      "15 D.J. Moore\n",
      "16 Dane Cruikshank\n",
      "17 Avonte Maddox\n",
      "18 Sam Darnold\n",
      "19 Harrison Phillips\n",
      "20 Riley Ferguson\n",
      "21 Jaire Alexander\n",
      "22 Troy Apke\n",
      "23 Minkah Fitzpatrick\n",
      "24 Taven Bryan\n",
      "25 Malik Jefferson\n",
      "26 D.J. Chark\n",
      "27 Kylie Fitts\n",
      "28 Darius Phillips\n",
      "29 Marquis Haynes\n",
      "30 B.J. Hill\n",
      "31 Josh Rosen\n",
      "32 Godwin Igwebuike\n",
      "33 Josh Sweat\n",
      "34 Rod Taylor\n",
      "35 Leighton Vander esch\n",
      "36 Rick Leonard\n",
      "37 Rashaad Penny\n",
      "38 Quenton Nelson\n",
      "39 Tremaine Edmunds\n",
      "40 Courtland Sutton\n",
      "41 Brian O'neill\n",
      "42 Genard Avery\n",
      "43 Sam Hubbard\n",
      "44 Donte Jackson\n",
      "45 Vita Vea\n",
      "46 Scott Quessenberry\n",
      "47 Quinton Flowers\n",
      "48 Wyatt Teller\n",
      "49 Jessie Bates\n",
      "50 D.J. Reed\n",
      "51 James Hearns\n",
      "52 Anthony Averett\n",
      "53 Baker Mayfield\n",
      "54 Marcus Davenport\n",
      "55 Matthew Thomas\n",
      "56 Denzel Ward\n",
      "57 Nyheim Hines\n",
      "58 Deadrin Senat\n",
      "59 Logan Woodside\n",
      "60 Joshua Kalu\n",
      "61 Mike Gesicki\n",
      "62 Christian Kirk\n",
      "63 Isaiah Wynn\n",
      "64 Parry Nickerson\n",
      "65 Quenton Meeks\n",
      "66 Oren Burks\n",
      "67 Braden Smith\n",
      "68 Shaquem Griffin\n",
      "69 Mason Rudolph\n",
      "70 Hercules Mata'afa\n",
      "71 Tyquan Lewis\n",
      "72 Will Richardson\n",
      "73 Jojo Wicker\n",
      "74 Keke Coutee\n",
      "75 Deshon Elliott\n",
      "76 Tyrell Crosby\n",
      "77 Rasheem Green\n",
      "78 Michael Gallup\n",
      "79 Will Hernandez\n",
      "80 Justin Jones\n",
      "81 Will Clapp\n",
      "82 Cole Madison\n",
      "83 Duke Ejiofor\n",
      "84 Anthony Winbush\n",
      "85 Ogbonnia Okoronkwo\n",
      "86 James Washington\n",
      "87 Josh Jackson\n",
      "88 Andre Chachere\n",
      "89 Arden Key\n",
      "90 Deon Cain\n",
      "91 Cedrick Wilson\n",
      "92 Kemoko Turay \n",
      "93 Calvin Ridley\n",
      "94 Toby Weathersby\n",
      "95 Connor Williams\n",
      "96 Roquan Smith\n",
      "97 Derrick Nnadi\n",
      "98 Peter Kalambayi\n",
      "99 Colby Gossett\n",
      "100 Ryan Nall\n",
      "101 Tarvarus McFadden\n",
      "102 Trevon Young\n",
      "103 Nick Chubb\n",
      "104 Mike Hughes\n",
      "105 Anthony Miller\n",
      "106 Derrius Guice\n",
      "107 Bo Scarbrough\n",
      "108 Tegray Scales\n",
      "109 Andrew Brown\n",
      "110 Quin Blanding\n",
      "111 Daron Payne\n",
      "112 Armani Watts\n",
      "113 Jerome Baker\n",
      "114 Quadree Henderson\n",
      "115 M.J. Stewart\n",
      "116 Lowell Lotulelei\n",
      "117 Tim Settle\n",
      "118 Tre Flowers\n",
      "119 Dylan Cantrell\n",
      "120 Antonio Callaway\n",
      "121 Chase Litton\n",
      "122 Jordan Whitehead\n",
      "123 Rashard Fant\n",
      "124 Chandon Sullivan\n",
      "125 Natrell Jamerson\n",
      "126 Nick Gates\n",
      "127 Jordan Akins\n",
      "128 Lorenzo Carter\n",
      "129 Fred Warner\n",
      "130 Allen Lazard\n",
      "131 Brandon Facyson\n",
      "132 J.T. Barrett\n",
      "133 Ade Aruna\n",
      "134 Equanimeous St. brown\n",
      "135 J.C. Jackson\n",
      "136 Jordan Thomas\n",
      "137 Dee Delaney\n",
      "138 Leon Jacobs\n",
      "139 Byron Pringle\n",
      "140 Mark Andrews\n",
      "141 Marcus Allen\n",
      "142 Chukwuma Okorafor\n",
      "143 Jack Cichy\n",
      "144 Duke Dawson\n",
      "145 Skai Moore\n",
      "146 Kyzir White\n",
      "147 Tanner Lee\n",
      "148 Dante Pettis \n",
      "149 Kurt Benkert\n",
      "150 Ronnie Harrison\n",
      "151 Kendrick Norton\n",
      "152 Holton Hill\n",
      "153 Jeff Holland\n",
      "154 Jester Weah\n",
      "155 Nic Shimonek\n",
      "156 Isaiah Oliver\n",
      "157 Jaryd Jones-Smith\n",
      "158 Eddy Wilson\n",
      "159 Greg Stroman\n",
      "160 Nick Nelson\n",
      "161 Salesi Uhatafe\n",
      "162 Kc McDermott\n",
      "163 Billy Price\n",
      "164 Grant Haley\n",
      "165 Brett Toth\n",
      "166 Simmie Cobbs\n",
      "167 Carlton Davis\n",
      "168 Ian Thomas\n",
      "169 Austin Corbett\n",
      "170 Deontay Burnett\n",
      "171 Marcell Ateman\n",
      "172 Jamarco Jones\n",
      "173 Jamarcus King\n",
      "174 Hayden Hurst\n",
      "175 Matt Dickerson\n",
      "176 Garret Dooley\n",
      "177 Dorian O'daniel\n",
      "178 Kamryn Pettway\n",
      "179 Mike McCray\n",
      "180 Royce Freeman\n",
      "181 Jalyn Holmes\n",
      "182 Dominick Sanders\n",
      "183 Micah Kiser\n",
      "184 Josh Adams\n",
      "185 Ronald Jones\n",
      "186 Orlando Brown\n",
      "187 Sean Chandler\n",
      "188 Joshua Frazier\n",
      "189 K.J. Malone\n",
      "190 Keishawn Bierria\n",
      "191 Jacob Pugh\n",
      "192 Van Smith\n",
      "193 Tony Brown\n",
      "194 Uchenna Nwosu\n",
      "195 Mason Cole\n",
      "196 Jordan Lasley\n",
      "197 Sam Jones\n",
      "198 Justin Jackson\n",
      "199 Nick Bawden\n",
      "200 Kerryon Johnson\n",
      "201 Christian Sam\n",
      "202 Chad Thomas\n",
      "203 Levi Wallace\n",
      "204 Kenny Young\n",
      "205 Josey Jewell\n",
      "206 Jaleel Scott\n",
      "207 Andre Smith\n",
      "208 Korey Robertson\n",
      "209 Luke Falk\n",
      "210 Auden Tate\n",
      "211 Azeem Victor\n",
      "212 Dimitri Flowers\n",
      "213 John Atkins\n",
      "214 Chris Jones\n",
      "215 Marquez Valdes-Scantling\n",
      "216 Danny Etling\n",
      "217 Taylor Hearn\n",
      "218 Marcell Harris\n",
      "219 Javon Wims\n",
      "220 James Daniels\n",
      "221 Rashaan Gaulden\n",
      "222 James Looney\n",
      "223 Jason Cabinda\n",
      "224 Trenton Thompson\n",
      "225 Kalen Ballage\n",
      "226 Rashaan Evans\n",
      "227 Tyler Conklin\n",
      "228 Geron Christian\n",
      "229 Taylor Stallworth\n",
      "230 Troy Fumagalli\n",
      "231 Chris Worley\n",
      "232 Jordan Wilkins\n",
      "233 Davon Grayson\n",
      "234 Zaycoven Henderson\n",
      "235 Jaylen Samuels\n",
      "236 Breeland Speaks\n",
      "237 Kamrin Moore\n",
      "238 Shaun dion Hamilton\n",
      "239 Austin Allen\n",
      "240 Jarvion Franklin\n",
      "241 Mark Walton\n",
      "242 Daesean Hamilton\n",
      "243 Kameron Kelly\n",
      "244 Kentavius Street\n",
      "245 Chris Covington\n",
      "246 John Kelly\n",
      "247 Damon Webb\n",
      "248 Isaac Yiadom\n",
      "249 Robert Foster\n",
      "250 Sony Michel\n",
      "251 Dalton Schultz\n",
      "252 Justin Crawford\n",
      "253 Coleman Shelton\n",
      "254 Joseph Davidson\n",
      "255 Durham Smythe\n",
      "256 David Wells\n",
      "257 Davin Bellamy\n",
      "258 Sean Welsh\n",
      "259 Cam Phillips\n",
      "260 Stephen Roberts\n",
      "261 Brian Allen\n",
      "262 Trey Quinn\n",
      "263 Will Dissly\n",
      "264 Ryan Izzo\n",
      "265 Chris Lacy\n",
      "266 Trevor Daniel\n",
      "267 Akrum Wadley\n",
      "268 Ray-ray McCloud\n",
      "269 Braxton Berrios\n",
      "270 Marcus Baugh\n",
      "271 Michael Dickson\n",
      "272 Tanner Carew\n",
      "273 Daniel Carlson\n",
      "274 Steven Mitchell\n",
      "275 Lavon Coleman\n",
      "276 Darrel Williams\n",
      "277 Eddy Pineiro\n",
      "278 Shane Tripucka\n",
      "279 Jk Scott\n",
      "280 Kyle Hicks\n",
      "281 Ryan Winslow\n",
      "282 Drew Brown\n",
      "283 Johnny Townsend\n"
     ]
    }
   ],
   "source": [
    "count = 1\n",
    "for i in pd.DataFrame(preds).sort_values(by=1, ascending=False).index:\n",
    "    print(str(count) + \" \" + str(df[df.Year==2018].reset_index().at[i, \"Name\"]))\n",
    "    count += 1"
   ]
  }
 ],
 "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}