AustinRochford/pp-for-sports-analytics-part-2.ipynb

## pp-for-sports-analytics-part-2.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "A Dockerfile that will produce a container with all the dependencies necessary to run this notebook is available [here](https://github.com/AustinRochford/notebooks)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "import datetime\n",
    "import logging\n",
    "from warnings import filterwarnings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING (theano.configdefaults): install mkl with `conda install mkl-service`: No module named 'mkl'\n"
     ]
    }
   ],
   "source": [
    "import arviz as az\n",
    "from matplotlib import pyplot as plt\n",
    "from matplotlib.ticker import StrMethodFormatter\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import pymc3 as pm\n",
    "import scipy as sp\n",
    "import seaborn as sns\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from theano import pprint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "sns.set(color_codes=True)\n",
    "pct_formatter = StrMethodFormatter('{x:.1%}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "# configure pyplot for readability when rendered as a slideshow and projected\n",
    "FIG_WIDTH, FIG_HEIGHT = 8, 6\n",
    "plt.rc('figure', figsize=(FIG_WIDTH, FIG_HEIGHT))\n",
    "\n",
    "LABELSIZE = 14\n",
    "plt.rc('axes', labelsize=LABELSIZE)\n",
    "plt.rc('axes', titlesize=LABELSIZE)\n",
    "plt.rc('figure', titlesize=LABELSIZE)\n",
    "plt.rc('legend', fontsize=LABELSIZE)\n",
    "plt.rc('xtick', labelsize=LABELSIZE)\n",
    "plt.rc('ytick', labelsize=LABELSIZE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "filterwarnings('ignore', 'findfont')\n",
    "filterwarnings('ignore', \"Conversion of the second argument of issubdtype\")\n",
    "filterwarnings('ignore', \"Set changed size during iteration\")\n",
    "\n",
    "# keep theano from complaining about compile locks for small models\n",
    "(logging.getLogger('theano.gof.compilelock')\n",
    "        .setLevel(logging.CRITICAL))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "SEED = 54902 # from random.org, for reproducibility\n",
    "\n",
    "np.random.seed(SEED)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# NBA Foul Calls\n",
    "\n",
    "<center><img src=\"https://upload.wikimedia.org/wikipedia/commons/7/76/Kyrie_Irving_Free_Throw.jpg\" width=700></center>\n",
    "\n",
    "**Question:** Is (not) committing and/or drawing fouls a measurable player skill?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "See this [talk](http://austinrochford.com/resources/talks/nba-fouls-pydata-nyc-2017.slides.html) or this [post](http://austinrochford.com/posts/2018-02-04-nba-irt-2.html) for more information on the data, expanded models, and conclusions from this case study."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "%%bash\n",
    "DATA_URI=https://raw.githubusercontent.com/polygraph-cool/last-two-minute-report/32f1c43dfa06c2e7652cc51ea65758007f2a1a01/output/all_games.csv\n",
    "DATA_DEST=/tmp/all_games.csv\n",
    "\n",
    "if [[ ! -e $DATA_DEST ]];\n",
    "then\n",
    "    wget -q -O $DATA_DEST $DATA_URI\n",
    "fi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "USECOLS = [\n",
    "    'period',\n",
    "    'seconds_left',\n",
    "    'call_type',\n",
    "    'committing_player',\n",
    "    'disadvantaged_player',\n",
    "    'review_decision',\n",
    "    'play_id',\n",
    "    'away',\n",
    "    'home',\n",
    "    'date',\n",
    "    'score_away',\n",
    "    'score_home',\n",
    "    'disadvantaged_team',\n",
    "    'committing_team'\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "orig_df = pd.read_csv(\n",
    "    '/tmp/all_games.csv',\n",
    "    usecols=USECOLS,\n",
    "    index_col='play_id',\n",
    "    parse_dates=['date']\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "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>play_id</th>\n",
       "      <th>20150301CLEHOU-0</th>\n",
       "      <th>20150301CLEHOU-1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>period</th>\n",
       "      <td>Q4</td>\n",
       "      <td>Q4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>seconds_left</th>\n",
       "      <td>112</td>\n",
       "      <td>103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>call_type</th>\n",
       "      <td>Foul: Shooting</td>\n",
       "      <td>Foul: Shooting</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>committing_player</th>\n",
       "      <td>Josh Smith</td>\n",
       "      <td>J.R. Smith</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>disadvantaged_player</th>\n",
       "      <td>Kevin Love</td>\n",
       "      <td>James Harden</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_decision</th>\n",
       "      <td>CNC</td>\n",
       "      <td>CC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>away</th>\n",
       "      <td>CLE</td>\n",
       "      <td>CLE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>home</th>\n",
       "      <td>HOU</td>\n",
       "      <td>HOU</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <td>2015-03-01 00:00:00</td>\n",
       "      <td>2015-03-01 00:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>score_away</th>\n",
       "      <td>103</td>\n",
       "      <td>103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>score_home</th>\n",
       "      <td>105</td>\n",
       "      <td>105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>disadvantaged_team</th>\n",
       "      <td>CLE</td>\n",
       "      <td>HOU</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>committing_team</th>\n",
       "      <td>HOU</td>\n",
       "      <td>CLE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "play_id                  20150301CLEHOU-0     20150301CLEHOU-1\n",
       "period                                 Q4                   Q4\n",
       "seconds_left                          112                  103\n",
       "call_type                  Foul: Shooting       Foul: Shooting\n",
       "committing_player              Josh Smith           J.R. Smith\n",
       "disadvantaged_player           Kevin Love         James Harden\n",
       "review_decision                       CNC                   CC\n",
       "away                                  CLE                  CLE\n",
       "home                                  HOU                  HOU\n",
       "date                  2015-03-01 00:00:00  2015-03-01 00:00:00\n",
       "score_away                            103                  103\n",
       "score_home                            105                  105\n",
       "disadvantaged_team                    CLE                  HOU\n",
       "committing_team                       HOU                  CLE"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "orig_df.head(n=2).T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "foul_df = orig_df[\n",
    "    orig_df.call_type\n",
    "           .fillna(\"UNKNOWN\")\n",
    "           .str.startswith(\"Foul\")\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "FOULS = [\n",
    "    f\"Foul: {foul_type}\"\n",
    "    for foul_type in [\n",
    "        \"Personal\",\n",
    "        \"Shooting\",\n",
    "        \"Offensive\",\n",
    "        \"Loose Ball\",\n",
    "        \"Away from Play\"\n",
    "    ]\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "TEAM_MAP = {\n",
    "    \"NKY\": \"NYK\",\n",
    "    \"COS\": \"BOS\",\n",
    "    \"SAT\": \"SAS\",\n",
    "    \"CHi\": \"CHI\",\n",
    "    \"LA)\": \"LAC\",\n",
    "    \"AT)\": \"ATL\",\n",
    "    \"ARL\": \"ATL\"\n",
    "}\n",
    "\n",
    "def correct_team_name(col):\n",
    "    def _correct_team_name(df):\n",
    "        return df[col].apply(lambda team_name: TEAM_MAP.get(team_name, team_name))\n",
    "    \n",
    "    return _correct_team_name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "def date_to_season(date):\n",
    "    if date >= datetime.datetime(2017, 10, 17):\n",
    "        return '2017-2018'\n",
    "    elif date >= datetime.datetime(2016, 10, 25):\n",
    "        return '2016-2017'\n",
    "    elif date >= datetime.datetime(2015, 10, 27):\n",
    "        return '2015-2016'\n",
    "    else:\n",
    "        return '2014-2015'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "clean_df = (foul_df.where(lambda df: df.period == \"Q4\")\n",
    "                   .where(lambda df: (df.date.between(datetime.datetime(2016, 10, 25),\n",
    "                                                      datetime.datetime(2017, 4, 12))\n",
    "                                      | df.date.between(datetime.datetime(2015, 10, 27),\n",
    "                                                       datetime.datetime(2016, 5, 30)))\n",
    "                         )\n",
    "                   .assign(\n",
    "                       review_decision=lambda df: df.review_decision.fillna(\"INC\"),\n",
    "                       committing_team=correct_team_name('committing_team'),\n",
    "                       disadvantged_team=correct_team_name('disadvantaged_team'),\n",
    "                       away=correct_team_name('away'),\n",
    "                       home=correct_team_name('home'),\n",
    "                       season=lambda df: df.date.apply(date_to_season)\n",
    "                    )\n",
    "                   .where(lambda df: df.call_type.isin(FOULS))\n",
    "                   .dropna()\n",
    "                   .drop('period', axis=1)\n",
    "                   .assign(call_type=lambda df: (df.call_type\n",
    "                                                   .str.split(': ', expand=True)  \n",
    "                                                   .iloc[:, 1])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "player_enc = LabelEncoder().fit(\n",
    "    np.concatenate((\n",
    "        clean_df.committing_player,\n",
    "        clean_df.disadvantaged_player\n",
    "    ))\n",
    ")\n",
    "n_player = player_enc.classes_.size\n",
    "\n",
    "season_enc = LabelEncoder().fit(\n",
    "    clean_df.season\n",
    ")\n",
    "n_season = season_enc.classes_.size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "df = (clean_df[['seconds_left']]\n",
    "              .round(0)\n",
    "              .assign(\n",
    "                foul_called=1. * clean_df.review_decision.isin(['CC', 'INC']),\n",
    "                player_committing=player_enc.transform(clean_df.committing_player),\n",
    "                player_disadvantaged=player_enc.transform(clean_df.disadvantaged_player),\n",
    "                score_committing=clean_df.score_home.where(\n",
    "                    clean_df.committing_team == clean_df.home,\n",
    "                    clean_df.score_away\n",
    "                ),\n",
    "                score_disadvantaged=clean_df.score_home.where(\n",
    "                    clean_df.disadvantaged_team == clean_df.home,\n",
    "                    clean_df.score_away\n",
    "                ),\n",
    "                season=season_enc.transform(clean_df.season)\n",
    "              ))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "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>seconds_left</th>\n",
       "      <th>foul_called</th>\n",
       "      <th>player_committing</th>\n",
       "      <th>player_disadvantaged</th>\n",
       "      <th>score_committing</th>\n",
       "      <th>score_disadvantaged</th>\n",
       "      <th>season</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>play_id</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>20151028INDTOR-1</th>\n",
       "      <td>89.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>162</td>\n",
       "      <td>98</td>\n",
       "      <td>99.0</td>\n",
       "      <td>106.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20151028INDTOR-2</th>\n",
       "      <td>73.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>36</td>\n",
       "      <td>358</td>\n",
       "      <td>106.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20151028INDTOR-3</th>\n",
       "      <td>38.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>229</td>\n",
       "      <td>222</td>\n",
       "      <td>99.0</td>\n",
       "      <td>106.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20151028INDTOR-4</th>\n",
       "      <td>30.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>229</td>\n",
       "      <td>98</td>\n",
       "      <td>99.0</td>\n",
       "      <td>106.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20151028INDTOR-6</th>\n",
       "      <td>24.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>229</td>\n",
       "      <td>100</td>\n",
       "      <td>99.0</td>\n",
       "      <td>106.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  seconds_left  foul_called  player_committing  \\\n",
       "play_id                                                          \n",
       "20151028INDTOR-1          89.0          1.0                162   \n",
       "20151028INDTOR-2          73.0          0.0                 36   \n",
       "20151028INDTOR-3          38.0          1.0                229   \n",
       "20151028INDTOR-4          30.0          0.0                229   \n",
       "20151028INDTOR-6          24.0          0.0                229   \n",
       "\n",
       "                  player_disadvantaged  score_committing  score_disadvantaged  \\\n",
       "play_id                                                                         \n",
       "20151028INDTOR-1                    98              99.0                106.0   \n",
       "20151028INDTOR-2                   358             106.0                 99.0   \n",
       "20151028INDTOR-3                   222              99.0                106.0   \n",
       "20151028INDTOR-4                    98              99.0                106.0   \n",
       "20151028INDTOR-6                   100              99.0                106.0   \n",
       "\n",
       "                  season  \n",
       "play_id                   \n",
       "20151028INDTOR-1       0  \n",
       "20151028INDTOR-2       0  \n",
       "20151028INDTOR-3       0  \n",
       "20151028INDTOR-4       0  \n",
       "20151028INDTOR-6       0  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "player_committing = df.player_committing.values\n",
    "player_disadvantaged = df.player_disadvantaged.values\n",
    "\n",
    "season = df.season.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "def hierarchical_normal(name, shape):\n",
    "    Δ = pm.Normal(f'Δ_{name}', 0., 1., shape=shape)\n",
    "    σ = pm.HalfNormal(f'σ_{name}', 5.)\n",
    "    \n",
    "    return pm.Deterministic(name, Δ * σ)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Model outline\n",
    "\n",
    "$$\n",
    "\\operatorname{log-odds}(\\textrm{Foul}) \\\n",
    "    \\sim \\textrm{Season factor} + \\left(\\textrm{Disadvantaged skill} - \\textrm{Committing skill}\\right)\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "with pm.Model() as irt_model:\n",
    "    β_season = pm.Normal('β_season', 0., 2.5, shape=n_season)\n",
    "    \n",
    "    θ = hierarchical_normal('θ', n_player) # disadvantaged skill\n",
    "    b = hierarchical_normal('b', n_player) # committing skill\n",
    "\n",
    "    p = pm.math.sigmoid(\n",
    "        β_season[season] + θ[player_disadvantaged] - b[player_committing]\n",
    "    )\n",
    "    \n",
    "    obs = pm.Bernoulli(\n",
    "        'obs', p,\n",
    "        observed=df['foul_called'].values\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "CHAINS = 3\n",
    "\n",
    "SAMPLE_KWARGS = {\n",
    "    'chains': CHAINS,\n",
    "    'cores': CHAINS,\n",
    "    'random_seed': list(SEED + np.arange(CHAINS))\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Auto-assigning NUTS sampler...\n",
      "Initializing NUTS using jitter+adapt_diag...\n",
      "Multiprocess sampling (3 chains in 3 jobs)\n",
      "NUTS: [σ_b, Δ_b, σ_θ, Δ_θ, β_season]\n",
      "Sampling 3 chains: 100%|██████████| 3000/3000 [02:07<00:00, 23.58draws/s]\n"
     ]
    }
   ],
   "source": [
    "with irt_model:\n",
    "    trace = pm.sample(500, **SAMPLE_KWARGS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "az.plot_energy(trace);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<xarray.Dataset>\n",
       "Dimensions:   ()\n",
       "Data variables:\n",
       "    β_season  float64 1.0\n",
       "    Δ_θ       float64 1.01\n",
       "    Δ_b       float64 1.0\n",
       "    σ_θ       float64 1.01\n",
       "    θ         float64 1.01\n",
       "    σ_b       float64 1.0\n",
       "    b         float64 1.0"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "az.rhat(trace).max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Basketball strategy leads to more complexity\n",
    "\n",
    "<center>\n",
    "    <img src=\"https://cdn10.phillymag.com/wp-content/uploads/2016/12/joel-embiid-2016123101.jpg\" width=800>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "df['trailing_committing'] = (df.score_committing\n",
    "                               .lt(df.score_disadvantaged)\n",
    "                               .mul(1.)\n",
    "                               .astype(np.int64))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "def make_foul_rate_yaxis(ax, label=\"Observed foul call rate\"):\n",
    "    ax.yaxis.set_major_formatter(pct_formatter)\n",
    "    ax.set_ylabel(label)\n",
    "    \n",
    "    return ax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "def make_time_axes(ax,\n",
    "                   xlabel=\"Seconds remaining in game\",\n",
    "                   ylabel=\"Observed foul call rate\"):\n",
    "    ax.invert_xaxis()\n",
    "    ax.set_xlabel(xlabel)\n",
    "    \n",
    "    return make_foul_rate_yaxis(ax, label=ylabel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = make_time_axes(\n",
    "    df.pivot_table('foul_called', 'seconds_left', 'trailing_committing')\n",
    "      .rolling(20).mean()\n",
    "      .rename(columns={0: \"No\", 1: \"Yes\"})\n",
    "      .rename_axis(\"Committing team is trailing\", axis=1)\n",
    "      .plot()\n",
    ").figure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fig"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### The shot clock\n",
    "\n",
    "<center><img src=\"https://cdn-s3.si.com/s3fs-public/styles/marquee_large_2x/public/2017/08/22/nba-shot-clock-invention.jpg\" width=700></center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "<center>\n",
    "    <img src=\"https://austinrochford.com/resources/nba_irt2/An%20Improved%20Analysis%20of%20NBA%20Foul%20Calls%20with%20Python_88_0.png\">\n",
    "    <a href=\"https://austinrochford.com/posts/2018-02-04-nba-irt-2.html\">Full model reference</a>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "### Are we measuring a skill?\n",
    "\n",
    "<center>\n",
    "    <img src=\"https://austinrochford.com/resources/nba_irt2/An%20Improved%20Analysis%20of%20NBA%20Foul%20Calls%20with%20Python_157_0.png\">\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "<center>\n",
    "    <table>\n",
    "        <tr>\n",
    "            <td>\n",
    "                <img src=\"https://austinrochford.com/resources/nba_irt2/An%20Improved%20Analysis%20of%20NBA%20Foul%20Calls%20with%20Python_161_0.png\">\n",
    "            </td>\n",
    "            <td>\n",
    "                <img src=\"https://austinrochford.com/resources/nba_irt2/An%20Improved%20Analysis%20of%20NBA%20Foul%20Calls%20with%20Python_163_0.png\">\n",
    "            </td>\n",
    "        </tr>\n",
    "    </table>\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "<center>\n",
    "    <img src=\"https://austinrochford.com/resources/nba_irt2/An%20Improved%20Analysis%20of%20NBA%20Foul%20Calls%20with%20Python_171_0.png\">\n",
    "</center>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Future Work\n",
    "\n",
    "* Advanced baseball stats\n",
    "    * <a href=\"https://en.wikipedia.org/wiki/On-base_plus_slugging\">OPS</a>\n",
    "    * <a href=\"https://en.wikipedia.org/wiki/Wins_Above_Replacement\">WAR</a>\n",
    "    * <a href=\"https://gist.github.com/AustinRochford/3ce28da1b7205c123366f1ac66811cfc\">Park factors</a>\n",
    "* Hockey\n",
    "    * Faceoff skills\n",
    "    * <a href=\"https://www.hockeyfights.com/\">Fighting ability</a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Thank you!\n",
    "\n",
    "<center>\n",
    "    <table>\n",
    "        <tr>\n",
    "            <td>\n",
    "                <img src='https://technical.ly/philly/wp-content/uploads/sites/2/2014/10/monetate-sign.jpg' width=400>\n",
    "                <a href=\"https://www.monetate.com/about/careers\">monetate.com/about/careers</a>\n",
    "            </td>\n",
    "            <td>\n",
    "                <img src=\"https://camo.githubusercontent.com/3d42ba7345f7b32431f6ed23847599a0ec80ec17/68747470733a2f2f6d65646961312e67697068792e636f6d2f6d656469612f78494a4c674f3672697a554a692f67697068792e676966\">\n",
    "            </td>\n",
    "        </tr>\n",
    "    </table>\n",
    "</center>\n",
    "\n",
    "### [@AustinRochford](https://twitter.com/AustinRochford) &#8226; [arochford@monetate.com](mailto:arochford@monetate.com) &#8226; [austin.rochford@gmail.com](mailto:austin.rochford@gmail.com)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[NbConvertApp] Converting notebook ./basketball-irt.ipynb to slides\n",
      "[NbConvertApp] Writing 323016 bytes to ./pp-sports-analytics-part-2.slides.html\n"
     ]
    }
   ],
   "source": [
    "%%bash\n",
    "jupyter nbconvert \\\n",
    "    --to=slides \\\n",
    "    --reveal-prefix=https://cdnjs.cloudflare.com/ajax/libs/reveal.js/3.2.0/ \\\n",
    "    --output=pp-sports-analytics-part-2 \\\n",
    "    ./basketball-irt.ipynb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "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": "3.6.7"
  },
  "widgets": {
   "state": {},
   "version": "1.1.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}