Skip to content

Instantly share code, notes, and snippets.

@jsvine

jsvine/warriors-streak-odds-2015-12-09.ipynb Secret

Created December 9, 2015 18:15
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save jsvine/2420508ffd35c10a0a95 to your computer and use it in GitHub Desktop.
Save jsvine/2420508ffd35c10a0a95 to your computer and use it in GitHub Desktop.
Download ZIP
How Long Will The Warriors' Streak Last?
Display the source blob
Display the rendered blob
Raw
warriors-streak-odds-2015-12-09.ipynb
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# How Long Will The Warriors' Streak Last?\n",
"\n",
"The code below estimates the Golden State Warriors' odds of matching or beating the NBA's win-streak record of 33 games. To do so, it uses per-game odds produced by [FiveThirtyEight's \"CARM-Elo\" projections](http://projects.fivethirtyeight.com/2016-nba-picks/).\n",
"\n",
"The [longest regular-season win streak belongs](https://en.wikipedia.org/wiki/List_of_National_Basketball_Association_longest_winning_streaks#Regular_season) to the '71-'72 Los Angeles Lakers, who won 33 games straight in a single season. The Warriors would tie the streak if they win their first 29 games this season, adding to a four-game streak carried over from the end of last regular season."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import requests\n",
"import random\n",
"import re\n",
"import yaml\n",
"import pandas as pd\n",
"import matplotlib as mpl\n",
"import seaborn as sb\n",
"%matplotlib inline"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Fetch the projections from FiveThirtyEight"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"URL = \"http://projects.fivethirtyeight.com/2016-nba-picks/js/bundle.js\"\n",
"js = requests.get(URL, params={\"r\": random.random()}).content.decode(\"utf-8\")"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def fix_json(json_str):\n",
" json_str = re.sub(r\"([,:])\", r\"\\1 \", json_str)\n",
" json_str = re.sub(r\"!1\", \"false\", json_str)\n",
" json_str = re.sub(r\"!0\", \"true\", json_str)\n",
" return json_str"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"game_strings = re.findall(r\"\\{[^}[]+team1[^}[]+\\}\", fix_json(js))[:82 * 15]"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"games = pd.DataFrame([ yaml.load(game) for game in game_strings ])"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"# Make sure every team has 41 home games.\n",
"assert((games[\"team1\"].value_counts() == 41).sum() == 30)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>0</th>\n",
" <th>1</th>\n",
" <th>2</th>\n",
" <th>3</th>\n",
" <th>4</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>carmelo1_post</th>\n",
" <td>1570.47</td>\n",
" <td>1521.81</td>\n",
" <td>1734.34</td>\n",
" <td>1442.24</td>\n",
" <td>1467.79</td>\n",
" </tr>\n",
" <tr>\n",
" <th>carmelo1_pre</th>\n",
" <td>1564.37</td>\n",
" <td>1542.66</td>\n",
" <td>1730.51</td>\n",
" <td>1447.01</td>\n",
" <td>1460.24</td>\n",
" </tr>\n",
" <tr>\n",
" <th>carmelo2_post</th>\n",
" <td>1725.92</td>\n",
" <td>1472.68</td>\n",
" <td>1551.3</td>\n",
" <td>1502.45</td>\n",
" <td>1469.15</td>\n",
" </tr>\n",
" <tr>\n",
" <th>carmelo2_pre</th>\n",
" <td>1732.03</td>\n",
" <td>1451.83</td>\n",
" <td>1555.13</td>\n",
" <td>1497.68</td>\n",
" <td>1476.71</td>\n",
" </tr>\n",
" <tr>\n",
" <th>carmelo_prob1</th>\n",
" <td>0.403078</td>\n",
" <td>0.755572</td>\n",
" <td>0.833364</td>\n",
" <td>0.567454</td>\n",
" <td>0.612829</td>\n",
" </tr>\n",
" <tr>\n",
" <th>carmelo_prob2</th>\n",
" <td>0.596922</td>\n",
" <td>0.244428</td>\n",
" <td>0.166636</td>\n",
" <td>0.432546</td>\n",
" <td>0.387171</td>\n",
" </tr>\n",
" <tr>\n",
" <th>carmelo_spread</th>\n",
" <td>2.5</td>\n",
" <td>-7</td>\n",
" <td>-10</td>\n",
" <td>-1.5</td>\n",
" <td>-3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>date</th>\n",
" <td>2015-10-27</td>\n",
" <td>2015-10-27</td>\n",
" <td>2015-10-27</td>\n",
" <td>2015-10-28</td>\n",
" <td>2015-10-28</td>\n",
" </tr>\n",
" <tr>\n",
" <th>id</th>\n",
" <td>400827889</td>\n",
" <td>400827888</td>\n",
" <td>400827890</td>\n",
" <td>400827891</td>\n",
" <td>400827895</td>\n",
" </tr>\n",
" <tr>\n",
" <th>neutral</th>\n",
" <td>False</td>\n",
" <td>False</td>\n",
" <td>False</td>\n",
" <td>False</td>\n",
" <td>False</td>\n",
" </tr>\n",
" <tr>\n",
" <th>playoff</th>\n",
" <td>None</td>\n",
" <td>None</td>\n",
" <td>None</td>\n",
" <td>None</td>\n",
" <td>None</td>\n",
" </tr>\n",
" <tr>\n",
" <th>score1</th>\n",
" <td>97</td>\n",
" <td>94</td>\n",
" <td>111</td>\n",
" <td>87</td>\n",
" <td>104</td>\n",
" </tr>\n",
" <tr>\n",
" <th>score2</th>\n",
" <td>95</td>\n",
" <td>106</td>\n",
" <td>95</td>\n",
" <td>88</td>\n",
" <td>94</td>\n",
" </tr>\n",
" <tr>\n",
" <th>status</th>\n",
" <td>post</td>\n",
" <td>post</td>\n",
" <td>post</td>\n",
" <td>post</td>\n",
" <td>post</td>\n",
" </tr>\n",
" <tr>\n",
" <th>team1</th>\n",
" <td>CHI</td>\n",
" <td>ATL</td>\n",
" <td>GS</td>\n",
" <td>ORL</td>\n",
" <td>MIA</td>\n",
" </tr>\n",
" <tr>\n",
" <th>team2</th>\n",
" <td>CLE</td>\n",
" <td>DET</td>\n",
" <td>NO</td>\n",
" <td>WSH</td>\n",
" <td>CHA</td>\n",
" </tr>\n",
" <tr>\n",
" <th>time_et</th>\n",
" <td>20: 00</td>\n",
" <td>20: 00</td>\n",
" <td>22: 30</td>\n",
" <td>19: 00</td>\n",
" <td>19: 30</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" 0 1 2 3 4\n",
"carmelo1_post 1570.47 1521.81 1734.34 1442.24 1467.79\n",
"carmelo1_pre 1564.37 1542.66 1730.51 1447.01 1460.24\n",
"carmelo2_post 1725.92 1472.68 1551.3 1502.45 1469.15\n",
"carmelo2_pre 1732.03 1451.83 1555.13 1497.68 1476.71\n",
"carmelo_prob1 0.403078 0.755572 0.833364 0.567454 0.612829\n",
"carmelo_prob2 0.596922 0.244428 0.166636 0.432546 0.387171\n",
"carmelo_spread 2.5 -7 -10 -1.5 -3\n",
"date 2015-10-27 2015-10-27 2015-10-27 2015-10-28 2015-10-28\n",
"id 400827889 400827888 400827890 400827891 400827895\n",
"neutral False False False False False\n",
"playoff None None None None None\n",
"score1 97 94 111 87 104\n",
"score2 95 106 95 88 94\n",
"status post post post post post\n",
"team1 CHI ATL GS ORL MIA\n",
"team2 CLE DET NO WSH CHA\n",
"time_et 20: 00 20: 00 22: 30 19: 00 19: 30"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"games.head().T"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Select the Warriors' games\n",
"\n",
"... and add a few extra columns for ease of computation."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>date</th>\n",
" <th>status</th>\n",
" <th>team1</th>\n",
" <th>team2</th>\n",
" <th>carmelo_prob1</th>\n",
" <th>carmelo_prob2</th>\n",
" <th>opponent</th>\n",
" </tr>\n",
" <tr>\n",
" <th>game_no</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>1</th>\n",
" <td>2015-10-27</td>\n",
" <td>post</td>\n",
" <td>GS</td>\n",
" <td>NO</td>\n",
" <td>0.833364</td>\n",
" <td>0.166636</td>\n",
" <td>NO</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>2015-10-30</td>\n",
" <td>post</td>\n",
" <td>HOU</td>\n",
" <td>GS</td>\n",
" <td>0.445619</td>\n",
" <td>0.554381</td>\n",
" <td>HOU</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>2015-10-31</td>\n",
" <td>post</td>\n",
" <td>NO</td>\n",
" <td>GS</td>\n",
" <td>0.384816</td>\n",
" <td>0.615184</td>\n",
" <td>NO</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>2015-11-02</td>\n",
" <td>post</td>\n",
" <td>GS</td>\n",
" <td>MEM</td>\n",
" <td>0.812844</td>\n",
" <td>0.187156</td>\n",
" <td>MEM</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>2015-11-04</td>\n",
" <td>post</td>\n",
" <td>GS</td>\n",
" <td>LAC</td>\n",
" <td>0.735165</td>\n",
" <td>0.264835</td>\n",
" <td>LAC</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" date status team1 team2 carmelo_prob1 carmelo_prob2 opponent\n",
"game_no \n",
"1 2015-10-27 post GS NO 0.833364 0.166636 NO\n",
"2 2015-10-30 post HOU GS 0.445619 0.554381 HOU\n",
"3 2015-10-31 post NO GS 0.384816 0.615184 NO\n",
"4 2015-11-02 post GS MEM 0.812844 0.187156 MEM\n",
"5 2015-11-04 post GS LAC 0.735165 0.264835 LAC"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gs = games[\n",
" ((games[\"team1\"] == \"GS\") |\n",
" (games[\"team2\"] == \"GS\"))\n",
"][[ \"date\", \"status\", \"team1\", \"team2\", \"carmelo_prob1\", \"carmelo_prob2\" ]]\\\n",
" .sort_values(\"date\").copy()\n",
"gs[\"game_no\"] = list(range(1, 83))\n",
"gs = gs.set_index(\"game_no\")\n",
"gs.loc[(gs[\"team1\"] == \"GS\"), \"opponent\"] = gs[\"team2\"]\n",
"gs.loc[(gs[\"team2\"] == \"GS\"), \"opponent\"] = gs[\"team1\"]\n",
"gs.head()"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"gs[\"gs_prob\"] = (gs[\"team1\"] == \"GS\") * gs[\"carmelo_prob1\"] \\\n",
" + (gs[\"team2\"] == \"GS\") * gs[\"carmelo_prob2\"]"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# Set all finished games' gs_prob to 1, because the Warriors won those\n",
"gs.loc[gs[\"status\"] == \"post\", \"gs_prob\"] = 1"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Calculate the odds that the Warriors extend their streak through a given game"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"gs[\"cumulative_prob\"] = gs[\"gs_prob\"].cumprod()"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"23"
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"SEASON_WINS = (gs[\"status\"] == \"post\").sum()\n",
"SEASON_WINS"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"PREV_WINS = 4"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"STREAK_WINS = SEASON_WINS + PREV_WINS"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"STREAK_RECORD = SEASON_RECORD = 33"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"29"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"STREAK_TIE_GAME = SEASON_WINS + (STREAK_RECORD - STREAK_WINS)\n",
"STREAK_TIE_GAME"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"STREAK_BREAK_GAME = STREAK_TIE_GAME + 1"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"SEASON_TIE_GAME = SEASON_RECORD\n",
"SEASON_BREAK_GAME = SEASON_TIE_GAME + 1"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## The previous game, plus the next 10 games:\n",
"\n",
"Note: `game_no` refers to the sequence of Warriors games *this* regular season, not the game number in the streak."
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>date</th>\n",
" <th>status</th>\n",
" <th>team1</th>\n",
" <th>team2</th>\n",
" <th>carmelo_prob1</th>\n",
" <th>carmelo_prob2</th>\n",
" <th>opponent</th>\n",
" <th>gs_prob</th>\n",
" <th>cumulative_prob</th>\n",
" </tr>\n",
" <tr>\n",
" <th>game_no</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>23</th>\n",
" <td>2015-12-08</td>\n",
" <td>post</td>\n",
" <td>IND</td>\n",
" <td>GS</td>\n",
" <td>0.252509</td>\n",
" <td>0.747491</td>\n",
" <td>IND</td>\n",
" <td>1.000000</td>\n",
" <td>1.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>24</th>\n",
" <td>2015-12-11</td>\n",
" <td>pre</td>\n",
" <td>BOS</td>\n",
" <td>GS</td>\n",
" <td>0.304816</td>\n",
" <td>0.695184</td>\n",
" <td>BOS</td>\n",
" <td>0.695184</td>\n",
" <td>0.695184</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25</th>\n",
" <td>2015-12-12</td>\n",
" <td>pre</td>\n",
" <td>MIL</td>\n",
" <td>GS</td>\n",
" <td>0.127148</td>\n",
" <td>0.872852</td>\n",
" <td>MIL</td>\n",
" <td>0.872852</td>\n",
" <td>0.606793</td>\n",
" </tr>\n",
" <tr>\n",
" <th>26</th>\n",
" <td>2015-12-16</td>\n",
" <td>pre</td>\n",
" <td>GS</td>\n",
" <td>PHX</td>\n",
" <td>0.927174</td>\n",
" <td>0.072826</td>\n",
" <td>PHX</td>\n",
" <td>0.927174</td>\n",
" <td>0.562602</td>\n",
" </tr>\n",
" <tr>\n",
" <th>27</th>\n",
" <td>2015-12-18</td>\n",
" <td>pre</td>\n",
" <td>GS</td>\n",
" <td>MIL</td>\n",
" <td>0.954676</td>\n",
" <td>0.045324</td>\n",
" <td>MIL</td>\n",
" <td>0.954676</td>\n",
" <td>0.537103</td>\n",
" </tr>\n",
" <tr>\n",
" <th>28</th>\n",
" <td>2015-12-23</td>\n",
" <td>pre</td>\n",
" <td>GS</td>\n",
" <td>UTA</td>\n",
" <td>0.913321</td>\n",
" <td>0.086679</td>\n",
" <td>UTA</td>\n",
" <td>0.913321</td>\n",
" <td>0.490547</td>\n",
" </tr>\n",
" <tr>\n",
" <th>29</th>\n",
" <td>2015-12-25</td>\n",
" <td>pre</td>\n",
" <td>GS</td>\n",
" <td>CLE</td>\n",
" <td>0.815078</td>\n",
" <td>0.184922</td>\n",
" <td>CLE</td>\n",
" <td>0.815078</td>\n",
" <td>0.399834</td>\n",
" </tr>\n",
" <tr>\n",
" <th>30</th>\n",
" <td>2015-12-28</td>\n",
" <td>pre</td>\n",
" <td>GS</td>\n",
" <td>SAC</td>\n",
" <td>0.950854</td>\n",
" <td>0.049146</td>\n",
" <td>SAC</td>\n",
" <td>0.950854</td>\n",
" <td>0.380184</td>\n",
" </tr>\n",
" <tr>\n",
" <th>31</th>\n",
" <td>2015-12-30</td>\n",
" <td>pre</td>\n",
" <td>DAL</td>\n",
" <td>GS</td>\n",
" <td>0.248551</td>\n",
" <td>0.751449</td>\n",
" <td>DAL</td>\n",
" <td>0.751449</td>\n",
" <td>0.285689</td>\n",
" </tr>\n",
" <tr>\n",
" <th>32</th>\n",
" <td>2015-12-31</td>\n",
" <td>pre</td>\n",
" <td>HOU</td>\n",
" <td>GS</td>\n",
" <td>0.248239</td>\n",
" <td>0.751761</td>\n",
" <td>HOU</td>\n",
" <td>0.751761</td>\n",
" <td>0.214770</td>\n",
" </tr>\n",
" <tr>\n",
" <th>33</th>\n",
" <td>2016-01-02</td>\n",
" <td>pre</td>\n",
" <td>GS</td>\n",
" <td>DEN</td>\n",
" <td>0.960504</td>\n",
" <td>0.039496</td>\n",
" <td>DEN</td>\n",
" <td>0.960504</td>\n",
" <td>0.206288</td>\n",
" </tr>\n",
" <tr>\n",
" <th>34</th>\n",
" <td>2016-01-04</td>\n",
" <td>pre</td>\n",
" <td>GS</td>\n",
" <td>CHA</td>\n",
" <td>0.915446</td>\n",
" <td>0.084554</td>\n",
" <td>CHA</td>\n",
" <td>0.915446</td>\n",
" <td>0.188845</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" date status team1 team2 carmelo_prob1 carmelo_prob2 opponent \\\n",
"game_no \n",
"23 2015-12-08 post IND GS 0.252509 0.747491 IND \n",
"24 2015-12-11 pre BOS GS 0.304816 0.695184 BOS \n",
"25 2015-12-12 pre MIL GS 0.127148 0.872852 MIL \n",
"26 2015-12-16 pre GS PHX 0.927174 0.072826 PHX \n",
"27 2015-12-18 pre GS MIL 0.954676 0.045324 MIL \n",
"28 2015-12-23 pre GS UTA 0.913321 0.086679 UTA \n",
"29 2015-12-25 pre GS CLE 0.815078 0.184922 CLE \n",
"30 2015-12-28 pre GS SAC 0.950854 0.049146 SAC \n",
"31 2015-12-30 pre DAL GS 0.248551 0.751449 DAL \n",
"32 2015-12-31 pre HOU GS 0.248239 0.751761 HOU \n",
"33 2016-01-02 pre GS DEN 0.960504 0.039496 DEN \n",
"34 2016-01-04 pre GS CHA 0.915446 0.084554 CHA \n",
"\n",
" gs_prob cumulative_prob \n",
"game_no \n",
"23 1.000000 1.000000 \n",
"24 0.695184 0.695184 \n",
"25 0.872852 0.606793 \n",
"26 0.927174 0.562602 \n",
"27 0.954676 0.537103 \n",
"28 0.913321 0.490547 \n",
"29 0.815078 0.399834 \n",
"30 0.950854 0.380184 \n",
"31 0.751449 0.285689 \n",
"32 0.751761 0.214770 \n",
"33 0.960504 0.206288 \n",
"34 0.915446 0.188845 "
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gs.ix[SEASON_WINS:SEASON_WINS + 11]"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"0.39983449666561749"
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gs.ix[STREAK_TIE_GAME][\"cumulative_prob\"]"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"0.38018407428889595"
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gs.ix[STREAK_BREAK_GAME][\"cumulative_prob\"]"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"0.2062875354050733"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gs.ix[SEASON_TIE_GAME][\"cumulative_prob\"]"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"0.18884502260758246"
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gs.ix[SEASON_BREAK_GAME][\"cumulative_prob\"]"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAABDcAAAHmCAYAAABnF2X6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x114f2f278>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"next_20 = gs.ix[SEASON_WINS:SEASON_WINS+21]\n",
"ax = next_20[\"cumulative_prob\"].plot(kind=\"bar\", figsize=(18, 6), width=0.9,\n",
" color=[ \"#006BB6\", \"#FDB927\" ],\n",
" lw=0)\n",
"\n",
"ax.set_title(\"How Long Will The Warriors' Streak Last?\\n\", linespacing=0.5, fontsize=24, fontweight=\"bold\")\n",
"\n",
"ax.set_xlabel(\"\\nStreak Length and Opponent\", fontsize=16)\n",
"ax.set_xticklabels([ \"{0}\\n{1}\".format(wins, opp)\n",
" for wins, opp in zip(next_20.index + PREV_WINS, next_20[\"opponent\"]) ],\n",
" rotation=0, fontsize=14)\n",
"ax.set_ylabel(\"Probability\", fontsize=16)\n",
"ax.set_yticks(pd.np.arange(0, 1.1, 0.1))\n",
"ax.set_yticklabels([ \"{0:.0f}%\".format(y * 100) for y in ax.get_yticks() ], fontsize=14)\n",
"ax.figure.text(0.5, -0.1,\n",
" s=\"Per-game odds: FiveThirtyEight's \\\"CARM-Elo\\\" model. / Chart: BuzzFeed News\",\n",
" fontsize=14,\n",
" ha=\"center\")\n",
"pass"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Color codes from [teamcolorcodes.com](http://teamcolorcodes.com/golden-state-warriors-color-codes/)."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"---\n",
"\n",
"---\n",
"\n",
"---"
]
}
],
"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": "3.4.3"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment