LR_as_Elo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic Regression as Elo\n",
    "[Benjamin Morris](https://twitter.com/skepticalsports/status/1147225488273788929) set me on the right path. The previous experiments used a general purpose classifier that just looked at the rating difference between teams to predict a winner. You can use the logistic regression framework to solve for optimal Elo ratings. So let's do that here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import defaultdict, deque\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn import linear_model\n",
    "plt.style.use('fivethirtyeight')\n",
    "plt.style.use('seaborn-white')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the data and drop the away games\n",
    "df = pd.read_csv('nbaallelo.csv',index_col='game_id')\n",
    "df = df[df['game_location']!='A'] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# teams moved around a lot\n",
    "from_new_to_old_team = {\n",
    "    'CHO':'CHA','NOP':'NOH','BRK':'NJN','OKC':'SEA','NOK':'NOH','NOH':'CHH','MEM':'VAN','WAS':'WSB','SAC':'KCK','LAC':'SDC','UTA':'NOJ','SDC':'BUF','NJN':'NYN','GSW':'SFW','DLC':'TEX','HOU':'SDR','CHA':'CHH','SAA':'TEX','SAS':'SAA','DEN':'DNA','DNA':'DNR','WSB':'CAP','CAP':'BAL','BAL':'CHZ','CHZ':'CHP','SDS':'SDA','FLO':'MMF','MMF':'MNM','SFW':'PHW','LAL':'MNL','LAS':'ANA','UTS':'LAS','CAR':'HSM','SSL':'CAR','DET':'FTW','MLH':'TRI','STL':'MLH','ATL':'STL','PHI':'SYR','CIN':'ROC','PTC':'PTP','MNP':'PTP','PTP':'MNP','MMP':'NOB','MMT':'MMP','MMS':'MMT','VIR':'WSA','WSA':'OAK'\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# the basic 538 Elo\n",
    "def get_elo_ratings(df_local,start_year = 1947):\n",
    "    K=20\n",
    "    HFA=100\n",
    "    SW=0.75\n",
    "    MOVM=3\n",
    "    ELOW=400\n",
    "    INTERCEPT=7.5\n",
    "    SLOPE=0.006\n",
    "    POW=0.8\n",
    "    NEW_TEAM=1300\n",
    "    year_id = start_year\n",
    "    ELOW=400\n",
    "    \n",
    "    df_local = df_local[df_local.year_id >= start_year]\n",
    "    team_year_elo = defaultdict(lambda : defaultdict(list))\n",
    "    \n",
    "    elo = {}\n",
    "\n",
    "    for i,row in enumerate(df_local.itertuples()):\n",
    "        # skip duplicates\n",
    "        if row[3] != 0:\n",
    "            continue\n",
    "        # update the year\n",
    "        if row[4] != year_id:\n",
    "            for k in elo:\n",
    "                elo[k] = SW*elo[k] + (1-SW)*1505\n",
    "            year_id +=1\n",
    "            \n",
    "            # update teams that renamed themselves\n",
    "            teams1 = set(list(df[df.year_id == year_id-1].team_id.unique()) + list(df[df.year_id == year_id-1].opp_id.unique()))\n",
    "            teams2 = set(list(df[df.year_id == year_id].team_id.unique()) + list(df[df.year_id == year_id].opp_id.unique()))\n",
    "            for t in [_ for _ in teams2 if not _ in teams1]:\n",
    "                if t in from_new_to_old_team and not t in elo and from_new_to_old_team[t] in elo:\n",
    "                    elo[t] = elo[from_new_to_old_team[t]]\n",
    "                    team_year_elo[t].update(team_year_elo[from_new_to_old_team[t]])\n",
    "        \n",
    "        # get the stats\n",
    "        t1,t2=row[8],row[14]\n",
    "        elo_i,elo_n = row[11],row[12]\n",
    "        opp_elo_i,opp_elo_n = row[17],row[18]\n",
    "        p1,p2 = row[10],row[16]\n",
    "        \n",
    "        # get the Elo\n",
    "        elo1 = elo.get(t1,NEW_TEAM) + HFA\n",
    "        if row[-4] == 'N':\n",
    "            elo1 -= HFA\n",
    "        elo2 = elo.get(t2,NEW_TEAM)\n",
    "\n",
    "        # get the win %\n",
    "        winp = 1.0 / (10 ** (-(elo1-elo2)/ELOW) + 1)\n",
    "    \n",
    "        # compute margin of victory correction\n",
    "        mov = abs(p1-p2)\n",
    "        elo_diff_w  = (elo2 - elo1)\n",
    "        mofv_m1 = ((mov+MOVM) ** POW)/(INTERCEPT + SLOPE*(-elo_diff_w))\n",
    "        mofv_m2 = ((mov+MOVM) ** POW)/(INTERCEPT + SLOPE*(elo_diff_w))\n",
    "\n",
    "        # find the elo update amount\n",
    "        if p1 > p2:\n",
    "            hm,am = 1,-1\n",
    "            add_t = K*(1-winp)*mofv_m1\n",
    "        else:\n",
    "            hm,am = -1,1\n",
    "            add_t = K*(winp)*mofv_m2\n",
    "\n",
    "        # apply it\n",
    "        if row[-4] == 'N':\n",
    "            elo[t1] = elo1 + hm*add_t\n",
    "        else:\n",
    "            elo[t1] = elo1 + hm*add_t - HFA\n",
    "\n",
    "        elo[t2] = elo2 + am*add_t\n",
    "        \n",
    "        # save it\n",
    "        team_year_elo[t1][year_id].append(elo[t1] - hm*add_t)\n",
    "        team_year_elo[t2][year_id].append(elo[t2] - am*add_t)\n",
    "    return team_year_elo\n",
    "team_year_elo = get_elo_ratings(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6475659229208925\n"
     ]
    }
   ],
   "source": [
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
    "\n",
    "if False:\n",
    "    import autograd.numpy as np\n",
    "    import autograd\n",
    "    def generate_ratings2(dset,teams,weight_recent = 0,HFA=75/173):\n",
    "        N = len(teams)\n",
    "        X = []\n",
    "        Y = []\n",
    "        for t1,games in dset.items():\n",
    "            for t2,tW in games:\n",
    "                v = np.zeros(N)\n",
    "                v[teams[t1]] = 1\n",
    "                v[teams[t2]] = -1\n",
    "                #v[-1] = 70/173\n",
    "                Y.append(tW)\n",
    "                X.append(v)\n",
    "        try:\n",
    "            w = np.ones_like(Y)\n",
    "            if weight_recent != 0:\n",
    "                w = [np.exp(-(i*i/weight_recent)) for i in range(len(Y))]\n",
    "                w = np.array(w)[::-1]\n",
    "                w /= w.mean()\n",
    "            X = np.array(X).astype(np.float)\n",
    "            Y = np.array(Y).astype(np.float)\n",
    "            W = 1e-2*np.random.normal(size=(X.shape[1]))\n",
    "            def loss_f(xt):\n",
    "                p = 1/(1+np.exp(- (X @ xt + HFA)))\n",
    "                loss = np.mean(w*(-Y * np.log(p) - (1-Y) *np.log(1-p+1e-12)))\n",
    "                return loss\n",
    "            GRAD = autograd.grad(loss_f)\n",
    "            alpha = 3e-1\n",
    "            v = np.zeros_like(W)\n",
    "            for i in range(1000):\n",
    "                loss = loss_f(W)\n",
    "                v = 0.9*v - GRAD(W)\n",
    "                W += (alpha) * v\n",
    "                #if (i%1000== 0):\n",
    "                #    break\n",
    "                if (np.linalg.norm(v) < 1e-2):\n",
    "                    print('early')\n",
    "                    break\n",
    "                if (i%50) == 0:\n",
    "                    print(loss,)\n",
    "                if i > 0 and (i%250) ==0:\n",
    "                    alpha /= 4\n",
    "            return (W * 173 + 1500,HFA)\n",
    "        except:\n",
    "            return ([1500 for i in range(N)],0)\n",
    "import scipy.optimize as opt\n",
    "prev_w = None\n",
    "def log_logit(xs):\n",
    "    xs = np.array(xs)\n",
    "    return np.choose(xs > 0,[-np.log(1. + np.exp(-xs)),xs - np.log(1. + np.exp(xs))])\n",
    "def generate_ratings2(dset,teams,weight_recent = 0,HFA=75/173):\n",
    "    global prev_w\n",
    "    N = len(teams)\n",
    "    X = []\n",
    "    Y = []\n",
    "    for t1,games in dset.items():\n",
    "        for t2,tW in games:\n",
    "            v = np.zeros(N)\n",
    "            v[teams[t1]] = 1\n",
    "            v[teams[t2]] = -1\n",
    "            #v[-1] = 70/173\n",
    "            Y.append(tW)\n",
    "            X.append(v)\n",
    "    try:\n",
    "        w = np.ones_like(Y)\n",
    "        if weight_recent != 0:\n",
    "            w = [np.exp(-(i*i/weight_recent)) for i in range(len(Y))]\n",
    "            w = np.array(w)[::-1]\n",
    "            w /= w.mean()\n",
    "        X = np.array(X).astype(np.float)\n",
    "        Y = np.array(Y).astype(np.float)\n",
    "        if prev_w is None or prev_w.shape[0] != X.shape[1]:\n",
    "            prev_w = 1e-2*np.random.normal(size=(X.shape[1]))\n",
    "        W = prev_w\n",
    "\n",
    "\n",
    "        def loss_f(xt):\n",
    "            V = X @ xt + HFA\n",
    "            loss = np.mean(w*(-Y * log_logit(V) - (1-Y) *log_logit(-V)))\n",
    "            return loss\n",
    "        res = opt.minimize(loss_f,W)\n",
    "        W = res.x\n",
    "        prev_w = W\n",
    "        return (W * 173 + 1500,HFA)\n",
    "    except KeyboardInterrupt:\n",
    "        raise\n",
    "    except:\n",
    "        return ([1500 for i in range(N)],0)\n",
    "def generate_ratings(dset,teams,weight_recent = 0):\n",
    "    N = len(teams)\n",
    "    X = []\n",
    "    Y = []\n",
    "    for t1,games in dset.items():\n",
    "        for t2,tW in games:\n",
    "            v = np.zeros(N)\n",
    "            v[teams[t1]] = 1\n",
    "            v[teams[t2]] = -1\n",
    "            #v[-1] = 70/173\n",
    "            Y.append(tW)\n",
    "            X.append(v)\n",
    "    try:\n",
    "        clf = linear_model.LogisticRegression()\n",
    "        if weight_recent != 0:\n",
    "            w = [np.exp(-(i*i/weight_recent)) for i in range(len(Y))]\n",
    "            w = np.array(w)[::-1]\n",
    "            w /= w.mean()\n",
    "            clf.fit(X,Y,w)\n",
    "        else:\n",
    "            clf.fit(X,Y)\n",
    "        return (clf.coef_[0] * 173 + 1500,clf.intercept_[0])\n",
    "    except:\n",
    "        return ([1500 for i in range(N)],0)\n",
    "hfa_vals = []\n",
    "\n",
    "def get_lr_ratings(df_local,start_year = 2011,games_std_dev=0):\n",
    "    correct_r = 0\n",
    "    total_r = 0\n",
    "    MAX_HOME_GAMES_PER_TEAM = 30 # 30\n",
    "    year_id = start_year-1\n",
    "  \n",
    "    df_local = df_local[df_local.year_id >= start_year]\n",
    "    team_year_elo = defaultdict(lambda : defaultdict(list))\n",
    "    \n",
    "    dataset = defaultdict(lambda : deque(maxlen=MAX_HOME_GAMES_PER_TEAM))\n",
    "    \n",
    "    for i,row in enumerate(df_local.itertuples()):\n",
    "        # skip duplicates\n",
    "        if row[3] != 0:\n",
    "            continue\n",
    "        # update the year\n",
    "        if row[4] != year_id:\n",
    "            year_id +=1\n",
    "            \n",
    "            teams1 = set(list(df[df.year_id == year_id-1].team_id.unique()) + list(df[df.year_id == year_id-1].opp_id.unique()))\n",
    "            teams2 = set(list(df[df.year_id == year_id].team_id.unique()) + list(df[df.year_id == year_id].opp_id.unique()))\n",
    "            old_to_new = {}\n",
    "            for t in [_ for _ in teams2 if not _ in teams1]:\n",
    "                if t in from_new_to_old_team:\n",
    "                    old_team = from_new_to_old_team[t]\n",
    "                    team_year_elo[t].update(team_year_elo[old_team])\n",
    "                    dataset[t] = dataset[old_team]\n",
    "                    old_to_new[old_team] = t\n",
    "            # remove the old team\n",
    "            for t in [t for t in dataset if not t in teams2]:\n",
    "                del dataset[t]\n",
    "            # correct the entries\n",
    "            for team in dataset:\n",
    "                replaced_teams = [(old_to_new.get(g[0],g[0]),g[1]) for g in dataset[team]]\n",
    "                dataset[team] = deque(replaced_teams,maxlen=MAX_HOME_GAMES_PER_TEAM)\n",
    "            teams = {t: i for i,t in enumerate(sorted(teams2))}\n",
    "\n",
    "        t1,t2=row[8],row[14]\n",
    "        p1,p2 = row[10],row[16]\n",
    "        \n",
    "        # generate a rating\n",
    "        rate,hfa = generate_ratings(dataset,teams,games_std_dev)\n",
    "        hfa_vals.append(hfa)\n",
    "        team_year_elo[t1][year_id].append(rate[teams[t1]])\n",
    "        team_year_elo[t2][year_id].append(rate[teams[t2]])\n",
    "        \n",
    "        if year_id >= 2013:\n",
    "            correct_r += int(((rate[teams[t1]] - rate[teams[t2]] +hfa) > 0) == (p1 > p2))\n",
    "            total_r += 1\n",
    "        # just store the other team and the result\n",
    "        dataset[t1].append((t2,p1>p2))\n",
    "        #print('.',end='')\n",
    "    print(correct_r/total_r)\n",
    "    return team_year_elo,dataset\n",
    "team_year_elo_reg,dataset = get_lr_ratings(df,2010,0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Regressed HCA (Last 30 Games weighted)')"
      ]
     },
     "execution_count": 6,
     "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": [
    "#_ = plt.plot([np.exp(-(i*i/(200**2))) for i in range(100)])\n",
    "#plt.figure()\n",
    "_ = plt.plot(173*np.array(hfa_vals)[100*10:])\n",
    "plt.xlabel('Game Number')\n",
    "plt.ylabel('Elo Home Court Advantage')\n",
    "plt.title('Regressed HCA (Last 30 Games weighted)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, '76.487')"
      ]
     },
     "execution_count": 7,
     "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": [
    "_ = plt.hist(173*np.array(hfa_vals[41*15:]),50,density=True)\n",
    "plt.title('{:.3f}'.format(173*np.array(hfa_vals[41*15:]).mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6675963488843814\n",
      "0.6635395537525355\n"
     ]
    }
   ],
   "source": [
    "for ds,hfa in zip([team_year_elo_reg,team_year_elo],[75,100]):\n",
    "    correct = 0\n",
    "    total = 0\n",
    "    for yr in range(2013,2016):\n",
    "        ds = {_:ds[_] for _ in ds if yr in ds[_]}\n",
    "        counter = {k:0 for k in ds}\n",
    "        for i,row in enumerate(df[df.year_id == yr].itertuples()):\n",
    "            # skip duplicates\n",
    "            if row[3] != 0:\n",
    "                continue\n",
    "            t1,t2=row[8],row[14]\n",
    "            p1,p2 = row[10],row[16]\n",
    "\n",
    "            elo1 = ds[t1][yr][counter[t1]]\n",
    "            elo2 = ds[t2][yr][counter[t2]]\n",
    "            correct += int(((elo1-elo2 + hfa) > 0) == (p1 > p2))\n",
    "            total += 1\n",
    "\n",
    "            counter[t1] += 1\n",
    "            counter[t2] += 1\n",
    "    print(correct/total)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1a21a755c0>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from scipy.ndimage.filters import gaussian_filter1d\n",
    "\n",
    "teams = ['GSW','HOU','CLE','MIL','SAS','LAL']\n",
    "plt.figure(figsize=(10,5))\n",
    "plt.subplot(2,1,1)\n",
    "plt.vlines(82,0,2500,linestyles='dashed',color=(0.1,0.1,0.1),lw=3,alpha=0.8)\n",
    "plt.vlines(82*2,0,2500,linestyles='dashed',color=(0.1,0.1,0.1),lw=3,alpha=0.8)\n",
    "plt.hlines(1500,0,82*3,linestyles='dashed',color=(0.1,0.1,0.1),lw=3,alpha=0.8)\n",
    "\n",
    "for t in teams:\n",
    "    plt.plot(np.concatenate([team_year_elo[t][y][:82] for y in range(2013,2016)]),label=t)\n",
    "\n",
    "plt.xticks([42,82+42,42+82*2],['12-13','13-14','14-15'])\n",
    "plt.xlim(0,82*3)\n",
    "plt.ylabel('538 Elo')\n",
    "plt.xlim(0,82*3)\n",
    "plt.ylim(1250,1800)\n",
    "\n",
    "plt.subplot(2,1,2)\n",
    "plt.vlines(82,0,2500,linestyles='dashed',color=(0.1,0.1,0.1),lw=3,alpha=0.8)\n",
    "plt.vlines(82*2,0,2500,linestyles='dashed',color=(0.1,0.1,0.1),lw=3,alpha=0.8)\n",
    "plt.hlines(1500,0,82*3,linestyles='dashed',color=(0.1,0.1,0.1),lw=3,alpha=0.8)\n",
    "\n",
    "for t in teams:\n",
    "    plt.plot(np.concatenate([team_year_elo_reg[t][y][:82] for y in range(2013,2016)]),label=t)\n",
    "plt.xticks([42,82+42,42+82*2],['12-13','13-14','14-15'])\n",
    "plt.xlim(0,82*3)\n",
    "plt.ylim(1200,1800)\n",
    "plt.ylabel('Regressed Elo (30 G)')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.legend(loc=8,ncol=3,borderaxespad=0,bbox_to_anchor=(0,1.0,1,0),frameon=True,fontsize=10)\n",
    "#plt.savefig('elo4.png',edgecolor='w',facecolor='w')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "77.76946663133926\n",
      "| Variable | Coeff |\n",
      "|----------|-------|\n",
      "|GSW                      |1749.10|\n",
      "|CLE                      |1699.11|\n",
      "|SAS                      |1655.38|\n",
      "|LAC                      |1647.26|\n",
      "|HOU                      |1644.81|\n",
      "|ATL                      |1633.64|\n",
      "|MEM                      |1627.77|\n",
      "|OKC                      |1583.05|\n",
      "|DAL                      |1559.62|\n",
      "|CHI                      |1552.56|\n",
      "|POR                      |1551.88|\n",
      "|UTA                      |1537.73|\n",
      "|NOP                      |1533.68|\n",
      "|WAS                      |1519.72|\n",
      "|BOS                      |1506.27|\n",
      "|TOR                      |1502.36|\n",
      "|IND                      |1494.39|\n",
      "|MIL                      |1485.54|\n",
      "|BRK                      |1476.20|\n",
      "|PHO                      |1475.12|\n",
      "|DET                      |1472.54|\n",
      "|MIA                      |1457.91|\n",
      "|CHO                      |1452.37|\n",
      "|DEN                      |1389.98|\n",
      "|SAC                      |1350.62|\n",
      "|ORL                      |1319.00|\n",
      "|PHI                      |1305.26|\n",
      "|LAL                      |1300.38|\n",
      "|NYK                      |1260.42|\n",
      "|MIN                      |1256.31|\n"
     ]
    }
   ],
   "source": [
    "teams = {k:i for i,k in enumerate(sorted([_ for _ in team_year_elo if 2015 in team_year_elo[_]]))}\n",
    "rate,hfa = generate_ratings(dataset,teams,0)\n",
    "coeffs = rate\n",
    "col_names = sorted(dataset.keys())\n",
    "v = np.argsort(abs(coeffs))[::-1]\n",
    "coeffs2 = [(coeffs[i2],col_names[i2]) for i2 in v]\n",
    "\n",
    "print(hfa*173.0)\n",
    "print('| Variable | Coeff |')\n",
    "print('|----------|-------|')\n",
    "for v,n in sorted(coeffs2,reverse=True):\n",
    "    print('|{:25s}|{:.2f}|'.format(n, v))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>gameorder</th>\n",
       "      <th>lg_id</th>\n",
       "      <th>_iscopy</th>\n",
       "      <th>year_id</th>\n",
       "      <th>date_game</th>\n",
       "      <th>seasongame</th>\n",
       "      <th>is_playoffs</th>\n",
       "      <th>team_id</th>\n",
       "      <th>fran_id</th>\n",
       "      <th>pts</th>\n",
       "      <th>...</th>\n",
       "      <th>win_equiv</th>\n",
       "      <th>opp_id</th>\n",
       "      <th>opp_fran</th>\n",
       "      <th>opp_pts</th>\n",
       "      <th>opp_elo_i</th>\n",
       "      <th>opp_elo_n</th>\n",
       "      <th>game_location</th>\n",
       "      <th>game_result</th>\n",
       "      <th>forecast</th>\n",
       "      <th>notes</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>game_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <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",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>201506090CLE</th>\n",
       "      <td>63154</td>\n",
       "      <td>NBA</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>6/9/2015</td>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>CLE</td>\n",
       "      <td>Cavaliers</td>\n",
       "      <td>96</td>\n",
       "      <td>...</td>\n",
       "      <td>61.826408</td>\n",
       "      <td>GSW</td>\n",
       "      <td>Warriors</td>\n",
       "      <td>91</td>\n",
       "      <td>1797.5032</td>\n",
       "      <td>1790.9591</td>\n",
       "      <td>H</td>\n",
       "      <td>W</td>\n",
       "      <td>0.527843</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201506110CLE</th>\n",
       "      <td>63155</td>\n",
       "      <td>NBA</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>6/11/2015</td>\n",
       "      <td>100</td>\n",
       "      <td>1</td>\n",
       "      <td>CLE</td>\n",
       "      <td>Cavaliers</td>\n",
       "      <td>82</td>\n",
       "      <td>...</td>\n",
       "      <td>60.309792</td>\n",
       "      <td>GSW</td>\n",
       "      <td>Warriors</td>\n",
       "      <td>103</td>\n",
       "      <td>1790.9591</td>\n",
       "      <td>1809.9791</td>\n",
       "      <td>H</td>\n",
       "      <td>L</td>\n",
       "      <td>0.546572</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201506140GSW</th>\n",
       "      <td>63156</td>\n",
       "      <td>NBA</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>6/14/2015</td>\n",
       "      <td>102</td>\n",
       "      <td>1</td>\n",
       "      <td>GSW</td>\n",
       "      <td>Warriors</td>\n",
       "      <td>104</td>\n",
       "      <td>...</td>\n",
       "      <td>68.013329</td>\n",
       "      <td>CLE</td>\n",
       "      <td>Cavaliers</td>\n",
       "      <td>91</td>\n",
       "      <td>1704.3949</td>\n",
       "      <td>1700.7391</td>\n",
       "      <td>H</td>\n",
       "      <td>W</td>\n",
       "      <td>0.765565</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201506170CLE</th>\n",
       "      <td>63157</td>\n",
       "      <td>NBA</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>6/16/2015</td>\n",
       "      <td>102</td>\n",
       "      <td>1</td>\n",
       "      <td>CLE</td>\n",
       "      <td>Cavaliers</td>\n",
       "      <td>97</td>\n",
       "      <td>...</td>\n",
       "      <td>59.290245</td>\n",
       "      <td>GSW</td>\n",
       "      <td>Warriors</td>\n",
       "      <td>105</td>\n",
       "      <td>1813.6349</td>\n",
       "      <td>1822.2881</td>\n",
       "      <td>H</td>\n",
       "      <td>L</td>\n",
       "      <td>0.481450</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              gameorder lg_id  _iscopy  year_id  date_game  seasongame  \\\n",
       "game_id                                                                  \n",
       "201506090CLE      63154   NBA        0     2015   6/9/2015          99   \n",
       "201506110CLE      63155   NBA        0     2015  6/11/2015         100   \n",
       "201506140GSW      63156   NBA        0     2015  6/14/2015         102   \n",
       "201506170CLE      63157   NBA        0     2015  6/16/2015         102   \n",
       "\n",
       "              is_playoffs team_id    fran_id  pts  ...  win_equiv  opp_id  \\\n",
       "game_id                                            ...                      \n",
       "201506090CLE            1     CLE  Cavaliers   96  ...  61.826408     GSW   \n",
       "201506110CLE            1     CLE  Cavaliers   82  ...  60.309792     GSW   \n",
       "201506140GSW            1     GSW   Warriors  104  ...  68.013329     CLE   \n",
       "201506170CLE            1     CLE  Cavaliers   97  ...  59.290245     GSW   \n",
       "\n",
       "               opp_fran opp_pts  opp_elo_i  opp_elo_n  game_location  \\\n",
       "game_id                                                                \n",
       "201506090CLE   Warriors      91  1797.5032  1790.9591              H   \n",
       "201506110CLE   Warriors     103  1790.9591  1809.9791              H   \n",
       "201506140GSW  Cavaliers      91  1704.3949  1700.7391              H   \n",
       "201506170CLE   Warriors     105  1813.6349  1822.2881              H   \n",
       "\n",
       "              game_result  forecast notes  \n",
       "game_id                                    \n",
       "201506090CLE            W  0.527843   NaN  \n",
       "201506110CLE            L  0.546572   NaN  \n",
       "201506140GSW            W  0.765565   NaN  \n",
       "201506170CLE            L  0.481450   NaN  \n",
       "\n",
       "[4 rows x 22 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[-4:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "75.0\n",
      "| Variable | Coeff |\n",
      "|----------|-------|\n",
      "|GSW                      |1780.67|\n",
      "|CLE                      |1720.58|\n",
      "|SAS                      |1675.20|\n",
      "|LAC                      |1667.05|\n",
      "|HOU                      |1664.00|\n",
      "|ATL                      |1645.98|\n",
      "|MEM                      |1644.10|\n",
      "|OKC                      |1591.79|\n",
      "|DAL                      |1569.49|\n",
      "|POR                      |1559.28|\n",
      "|CHI                      |1558.85|\n",
      "|UTA                      |1542.78|\n",
      "|NOP                      |1538.96|\n",
      "|WAS                      |1521.82|\n",
      "|BOS                      |1505.14|\n",
      "|TOR                      |1500.75|\n",
      "|IND                      |1491.56|\n",
      "|MIL                      |1482.98|\n",
      "|PHO                      |1475.21|\n",
      "|BRK                      |1474.54|\n",
      "|DET                      |1468.82|\n",
      "|MIA                      |1452.55|\n",
      "|CHO                      |1445.49|\n",
      "|DEN                      |1377.89|\n",
      "|SAC                      |1334.34|\n",
      "|ORL                      |1297.18|\n",
      "|PHI                      |1281.79|\n",
      "|LAL                      |1275.94|\n",
      "|NYK                      |1228.70|\n",
      "|MIN                      |1224.77|\n"
     ]
    }
   ],
   "source": [
    "if False:\n",
    "    import autograd.numpy as np\n",
    "    import autograd\n",
    "\n",
    "    def generate_ratings2(dset,teams,weight_recent = 0,HFA=75/173):\n",
    "        N = len(teams)\n",
    "        X = []\n",
    "        Y = []\n",
    "        for t1,games in dset.items():\n",
    "            for t2,tW in games:\n",
    "                v = np.zeros(N)\n",
    "                v[teams[t1]] = 1\n",
    "                v[teams[t2]] = -1\n",
    "                #v[-1] = 70/173\n",
    "                Y.append(tW)\n",
    "                X.append(v)\n",
    "        try:\n",
    "            w = np.ones_like(Y)\n",
    "            if weight_recent != 0:\n",
    "                w = [np.exp(-(i*i/weight_recent)) for i in range(len(Y))]\n",
    "                w = np.array(w)[::-1]\n",
    "                w /= w.mean()\n",
    "            X = np.array(X).astype(np.float)\n",
    "            Y = np.array(Y).astype(np.float)\n",
    "            W = 1e-2*np.random.normal(size=(X.shape[1]))\n",
    "            def loss_f(xt):\n",
    "                p = 1/(1+np.exp(- (X @ xt + HFA)))\n",
    "                loss = np.mean(w*(-Y * np.log(p) - (1-Y) *np.log(1-p+1e-12)))\n",
    "                return loss\n",
    "            GRAD = autograd.grad(loss_f)\n",
    "            alpha = 3e-1\n",
    "            v = np.zeros_like(W)\n",
    "            for i in range(1000):\n",
    "                loss = loss_f(W)\n",
    "                v = 0.9*v - GRAD(W)\n",
    "                W += (alpha) * v\n",
    "                #if (i%1000== 0):\n",
    "                #    break\n",
    "                if (np.linalg.norm(v) < 1e-2):\n",
    "                    print('early')\n",
    "                    break\n",
    "                if (i%50) == 0:\n",
    "                    print(loss,)\n",
    "            return (W * 173 + 1500,HFA)\n",
    "        except:\n",
    "            return ([1500 for i in range(N)],0)\n",
    "import scipy.optimize as opt\n",
    "def generate_ratings2(dset,teams,weight_recent = 0,HFA=75/173):\n",
    "    N = len(teams)\n",
    "    X = []\n",
    "    Y = []\n",
    "    for t1,games in dset.items():\n",
    "        for t2,tW in games:\n",
    "            v = np.zeros(N)\n",
    "            v[teams[t1]] = 1\n",
    "            v[teams[t2]] = -1\n",
    "            #v[-1] = 70/173\n",
    "            Y.append(tW)\n",
    "            X.append(v)\n",
    "    try:\n",
    "        w = np.ones_like(Y)\n",
    "        if weight_recent != 0:\n",
    "            w = [np.exp(-(i*i/weight_recent)) for i in range(len(Y))]\n",
    "            w = np.array(w)[::-1]\n",
    "            w /= w.mean()\n",
    "        X = np.array(X).astype(np.float)\n",
    "        Y = np.array(Y).astype(np.float)\n",
    "        W = 1e-2*np.random.normal(size=(X.shape[1]))\n",
    "        def loss_f(xt):\n",
    "            V = X @ xt + HFA\n",
    "            loss = np.mean(w*(-Y * log_logit(V) - (1-Y) *log_logit(-V)))\n",
    "            return loss\n",
    "        res = opt.minimize(loss_f,W)\n",
    "        W = res.x\n",
    "        return (W * 173 + 1500,HFA)\n",
    "    except:\n",
    "        return ([1500 for i in range(N)],0)\n",
    "teams = {k:i for i,k in enumerate(sorted([_ for _ in team_year_elo if 2015 in team_year_elo[_]]))}\n",
    "rate,hfa = generate_ratings2(dataset,teams,0)\n",
    "coeffs = rate\n",
    "col_names = sorted(dataset.keys())\n",
    "v = np.argsort(abs(coeffs))[::-1]\n",
    "coeffs2 = [(coeffs[i2],col_names[i2]) for i2 in v]\n",
    "\n",
    "print(hfa*173.0)\n",
    "print('| Variable | Coeff |')\n",
    "print('|----------|-------|')\n",
    "for v,n in sorted(coeffs2,reverse=True):\n",
    "    print('|{:25s}|{:.2f}|'.format(n, v))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}