psthomas/voting_wisconsin.ipynb

## voting_wisconsin.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualizing the Presidential Election Results \n",
    "\n",
    "\n",
    "Like many people, I'm still recovering from the results of the election.  A big part of the process for me is coming up with an accurate description of what actually happened, so this post attempts to do so by building a few visualizations.\n",
    "\n",
    "Building a graphic that clearly communicates the election results is difficult because you need to do a few things simultaneously:\n",
    "\n",
    "1. Use a high enough data resolution to show sub-state trends.  County level data does a good job here.  \n",
    "\n",
    "2. Communicate the size of the vote in each county so we know how consequential it was for the election.  If the data is at the national level, weight the vote totals by their contribution to the electoral college.\n",
    "\n",
    "3. Show both the voter turnout and margins for each party.  Either can swing an election.\n",
    "\n",
    "4. Compare the outcomes with those of past elections so we can get an idea of what changed this time.  Ideally, compare with the average of the past few elections so you're not comparing against a single candidate or a single point in time.\n",
    "\n",
    "The best graphics I have seen come from a Washington Post [article looking at the urban/rural divide](https://www.washingtonpost.com/graphics/politics/2016-election/urban-rural-vote-swing/), which gets pretty close to meeting all of the above requirements.  Unfortunately, their graphics communicate changes in voter turnout using line thickness, which makes it hard to discern differences over time.  \n",
    "\n",
    "Below I attempt to accomplish most of those objectives using county level data from Wisconsin.  \n",
    "\n",
    "\n",
    "## Getting the data\n",
    "\n",
    "The county level data for the 2000-2016 elections is from David Leip's [Atlas of Presidential Elections](http://uselectionatlas.org/).  There's a script for getting the state level data below. \n",
    "\n",
    "The voting age population (VAP) data is from the [American Community Survey](http://www.census.gov/rdo/data/voting_age_population_by_citizenship_and_race_cvap.html).  I downloaded county level data averaged over 2005-2009, and 2010-2014 to try to account for population changes.  I calculated turnout by dividing the sum of the votes by the voting age population for each county.  All of the input data is available [here](https://www.dropbox.com/s/6ubdl8mblq1lzpr/election_data.zip?dl=1).  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import pandas as pd\n",
    "import os \n",
    "import glob\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "\n",
    "matplotlib.style.use('ggplot')  #'ggplot' 'fivethirtyeight'  'seaborn-paper'\n",
    "\n",
    "#pd.set_option('max_colwidth', 400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Optional script for getting state level data:\n",
    "'''\n",
    "import requests\n",
    "from bs4 import BeautifulSoup\n",
    "import csv\n",
    "\n",
    "years = [2000, 2004, 2008, 2012, 2016]\n",
    "\n",
    "#Iterate through years 2000-2016, by 4\n",
    "base_url = 'http://uselectionatlas.org/RESULTS/datagraph.php?year={year}&fips=55&f=1&off=0&elect=0'\n",
    "\n",
    "row_list = []\n",
    "\n",
    "for year in years:\n",
    "\n",
    "    url = base_url.format(year=year)\n",
    "    r = requests.get(url) #Encoding is UTF-8\n",
    "\n",
    "    if r.status_code == 200:\n",
    "\n",
    "        soup = BeautifulSoup(r.text, 'html.parser')\n",
    "        tables = soup.findAll('table')[:-1]  #Ignore extra table at end\n",
    "        county = None\n",
    "\n",
    "        for table in tables:\n",
    "            rows = table.findAll('tr')\n",
    "\n",
    "            for row in rows:\n",
    "                data = row.findAll('td')\n",
    "\n",
    "                for element in data:\n",
    "                    if element.get('style') == 'width:100px':\n",
    "                        county = element.get_text(strip=True)\n",
    "                        data.remove(element)\n",
    "\n",
    "                candidate = data[0].get_text(strip=True)\n",
    "                pct = data[1].get_text(strip=True)\n",
    "                num = data[2].get_text(strip=True)\n",
    "\n",
    "                row = {'year': year, 'county': county, 'candidate': candidate, 'pct': pct, 'num':num}\n",
    "                row_list.append(row)\n",
    "                \n",
    "        print 'Completed {0}.'.format(year)\n",
    "\n",
    "    else:\n",
    "        print \"Failed to access url: \" + url\n",
    "\n",
    "    \n",
    "with open('county_data_wi_2000-2016.csv', 'wb') as csvfile:\n",
    "    fieldnames = ['year', 'county', 'candidate', 'pct', 'num']\n",
    "    writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\n",
    "    writer.writeheader()\n",
    "    writer.writerows(row_list)\n",
    "''' "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>year</th>\n",
       "      <th>county</th>\n",
       "      <th>candidate</th>\n",
       "      <th>pct</th>\n",
       "      <th>num</th>\n",
       "      <th>party</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2000</td>\n",
       "      <td>Adams</td>\n",
       "      <td>Gore</td>\n",
       "      <td>52.9</td>\n",
       "      <td>4826.0</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2000</td>\n",
       "      <td>Adams</td>\n",
       "      <td>Bush</td>\n",
       "      <td>43.0</td>\n",
       "      <td>3920.0</td>\n",
       "      <td>R</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2000</td>\n",
       "      <td>Adams</td>\n",
       "      <td>Nader</td>\n",
       "      <td>2.4</td>\n",
       "      <td>217.0</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2000</td>\n",
       "      <td>Adams</td>\n",
       "      <td>Other</td>\n",
       "      <td>1.7</td>\n",
       "      <td>153.0</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2000</td>\n",
       "      <td>Ashland</td>\n",
       "      <td>Gore</td>\n",
       "      <td>55.2</td>\n",
       "      <td>4356.0</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2000</td>\n",
       "      <td>Ashland</td>\n",
       "      <td>Bush</td>\n",
       "      <td>38.5</td>\n",
       "      <td>3038.0</td>\n",
       "      <td>R</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2000</td>\n",
       "      <td>Ashland</td>\n",
       "      <td>Nader</td>\n",
       "      <td>5.6</td>\n",
       "      <td>440.0</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2000</td>\n",
       "      <td>Ashland</td>\n",
       "      <td>Other</td>\n",
       "      <td>0.7</td>\n",
       "      <td>56.0</td>\n",
       "      <td>O</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2000</td>\n",
       "      <td>Barron</td>\n",
       "      <td>Gore</td>\n",
       "      <td>44.9</td>\n",
       "      <td>8928.0</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2000</td>\n",
       "      <td>Barron</td>\n",
       "      <td>Bush</td>\n",
       "      <td>49.5</td>\n",
       "      <td>9848.0</td>\n",
       "      <td>R</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year   county candidate   pct     num party\n",
       "0  2000    Adams      Gore  52.9  4826.0     D\n",
       "1  2000    Adams      Bush  43.0  3920.0     R\n",
       "2  2000    Adams     Nader   2.4   217.0     O\n",
       "3  2000    Adams     Other   1.7   153.0     O\n",
       "4  2000  Ashland      Gore  55.2  4356.0     D\n",
       "5  2000  Ashland      Bush  38.5  3038.0     R\n",
       "6  2000  Ashland     Nader   5.6   440.0     O\n",
       "7  2000  Ashland     Other   0.7    56.0     O\n",
       "8  2000   Barron      Gore  44.9  8928.0     D\n",
       "9  2000   Barron      Bush  49.5  9848.0     R"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Voting behavior by county in WI\n",
    "# Source: http://uselectionatlas.org/RESULTS/datagraph.php?year=2016&fips=55&f=1&off=0&elect=0\n",
    "\n",
    "voting_df = pd.read_csv('./data/voting/county_data_wi_2000-2016.csv')\n",
    "\n",
    "#print len(voting_df['county'].drop_duplicates())\n",
    "# 72 Counties\n",
    "\n",
    "rs = ['Bush', 'McCain', 'Romney', 'Trump']\n",
    "ds = ['Gore', 'Kerry', 'Obama', 'Clinton']\n",
    "\n",
    "def party(element):\n",
    "    if element in rs:\n",
    "        return 'R'\n",
    "    elif element in ds:\n",
    "        return 'D'\n",
    "    else:\n",
    "        return 'O'\n",
    "\n",
    "voting_df['party'] = voting_df['candidate'].apply(party)\n",
    "\n",
    "voting_df.replace({'%': '', ',': ''}, inplace=True, regex=True)\n",
    "voting_df[['pct','num']] = voting_df[['pct','num']].astype(float)  #.apply(pd.to_numeric)\n",
    "#voting_df.dtypes\n",
    "voting_df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>county</th>\n",
       "      <th>CVAP_EST</th>\n",
       "      <th>CVAP_MOE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>39624</th>\n",
       "      <td>Adams</td>\n",
       "      <td>16640</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39637</th>\n",
       "      <td>Ashland</td>\n",
       "      <td>12420</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39650</th>\n",
       "      <td>Barron</td>\n",
       "      <td>35260</td>\n",
       "      <td>94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39663</th>\n",
       "      <td>Bayfield</td>\n",
       "      <td>11945</td>\n",
       "      <td>51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39676</th>\n",
       "      <td>Brown</td>\n",
       "      <td>176420</td>\n",
       "      <td>491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39689</th>\n",
       "      <td>Buffalo</td>\n",
       "      <td>10470</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39702</th>\n",
       "      <td>Burnett</td>\n",
       "      <td>12925</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39715</th>\n",
       "      <td>Calumet</td>\n",
       "      <td>31990</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39728</th>\n",
       "      <td>Chippewa</td>\n",
       "      <td>45520</td>\n",
       "      <td>82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39741</th>\n",
       "      <td>Clark</td>\n",
       "      <td>23525</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         county  CVAP_EST  CVAP_MOE\n",
       "39624     Adams     16640        81\n",
       "39637   Ashland     12420        45\n",
       "39650    Barron     35260        94\n",
       "39663  Bayfield     11945        51\n",
       "39676     Brown    176420       491\n",
       "39689   Buffalo     10470        32\n",
       "39702   Burnett     12925        26\n",
       "39715   Calumet     31990       123\n",
       "39728  Chippewa     45520        82\n",
       "39741     Clark     23525        40"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Voting age population dataframe\n",
    "# Source: http://www.census.gov/rdo/data/voting_age_population_by_citizenship_and_race_cvap.html\n",
    "\n",
    "# 2010-2014 Averaged data\n",
    "population2014_df = pd.read_csv('./data/CVAP_CSV_Format_2010-2014/County.csv')\n",
    "\n",
    "population2014_df = population2014_df[(population2014_df['LNTITLE'] == 'Total') & \n",
    "                              (population2014_df['GEONAME'].str.contains('Wisconsin'))]\n",
    "population2014_df = population2014_df[['GEONAME', 'CVAP_EST', 'CVAP_MOE']]\n",
    "population2014_df.replace({' County, Wisconsin': ''}, inplace=True, regex=True)\n",
    "population2014_df.rename(columns={'GEONAME': 'county'}, inplace=True)\n",
    "                        \n",
    "#print len(population_df['GEONAME'].drop_duplicates())\n",
    "# 72 counties\n",
    "\n",
    "# 2005 to 2009 average\n",
    "population2009_df = pd.read_csv('./data/CVAP_CSV_Format_2005-2009/County.csv')\n",
    "population2009_df = population2009_df[(population2009_df['LNTITLE'] == 'Total') & \n",
    "                              (population2009_df['GEONAME'].str.contains('Wisconsin'))]\n",
    "population2009_df = population2009_df[['GEONAME', 'CVAP_EST', 'CVAP_MOE']]\n",
    "population2009_df.replace({' County, Wisconsin': ''}, inplace=True, regex=True)\n",
    "population2009_df.rename(columns={'GEONAME': 'county'}, inplace=True)\n",
    "\n",
    "# print len(population2008_df['GEONAME'].drop_duplicates())\n",
    "# 72\n",
    "\n",
    "population2009_df.head(10)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>county</th>\n",
       "      <th>num_2016</th>\n",
       "      <th>turnout_2016</th>\n",
       "      <th>turnout_2000_2012</th>\n",
       "      <th>turnout_change</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>Milwaukee</td>\n",
       "      <td>440698.0</td>\n",
       "      <td>0.654155</td>\n",
       "      <td>0.703174</td>\n",
       "      <td>-0.049019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Dane</td>\n",
       "      <td>309096.0</td>\n",
       "      <td>0.830200</td>\n",
       "      <td>0.757048</td>\n",
       "      <td>0.073152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>Waukesha</td>\n",
       "      <td>237588.0</td>\n",
       "      <td>0.808631</td>\n",
       "      <td>0.801177</td>\n",
       "      <td>0.007454</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Brown</td>\n",
       "      <td>128965.0</td>\n",
       "      <td>0.706793</td>\n",
       "      <td>0.680866</td>\n",
       "      <td>0.025927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>Outagamie</td>\n",
       "      <td>95162.0</td>\n",
       "      <td>0.724051</td>\n",
       "      <td>0.686042</td>\n",
       "      <td>0.038009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>Racine</td>\n",
       "      <td>94921.0</td>\n",
       "      <td>0.666838</td>\n",
       "      <td>0.686983</td>\n",
       "      <td>-0.020146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70</th>\n",
       "      <td>Winnebago</td>\n",
       "      <td>87140.0</td>\n",
       "      <td>0.667714</td>\n",
       "      <td>0.678759</td>\n",
       "      <td>-0.011046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>Washington</td>\n",
       "      <td>77551.0</td>\n",
       "      <td>0.776092</td>\n",
       "      <td>0.741324</td>\n",
       "      <td>0.034768</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Kenosha</td>\n",
       "      <td>76894.0</td>\n",
       "      <td>0.636830</td>\n",
       "      <td>0.639971</td>\n",
       "      <td>-0.003141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>Rock</td>\n",
       "      <td>76056.0</td>\n",
       "      <td>0.649524</td>\n",
       "      <td>0.670892</td>\n",
       "      <td>-0.021368</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        county  num_2016  turnout_2016  turnout_2000_2012  turnout_change\n",
       "40   Milwaukee  440698.0      0.654155           0.703174       -0.049019\n",
       "12        Dane  309096.0      0.830200           0.757048        0.073152\n",
       "67    Waukesha  237588.0      0.808631           0.801177        0.007454\n",
       "4        Brown  128965.0      0.706793           0.680866        0.025927\n",
       "44   Outagamie   95162.0      0.724051           0.686042        0.038009\n",
       "51      Racine   94921.0      0.666838           0.686983       -0.020146\n",
       "70   Winnebago   87140.0      0.667714           0.678759       -0.011046\n",
       "66  Washington   77551.0      0.776092           0.741324        0.034768\n",
       "29     Kenosha   76894.0      0.636830           0.639971       -0.003141\n",
       "53        Rock   76056.0      0.649524           0.670892       -0.021368"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "# Calculate voter turnouts by county\n",
    "turnout_df = pd.pivot_table(voting_df, values='num', index=['year', 'county'], aggfunc=np.sum)\n",
    "turnout_df = turnout_df.reset_index()\n",
    "\n",
    "turnout_df = pd.merge(turnout_df, population2009_df, on='county', suffixes=('', '_2009'))\n",
    "turnout_df = pd.merge(turnout_df, population2014_df, on='county', suffixes=('', '_2014'))\n",
    "turnout_df.rename(columns={'CVAP_EST': 'CVAP_EST_2009',\n",
    "                           'CVAP_MOE': 'CVAP_MOE_2009'}, inplace=True)\n",
    "\n",
    "def turnout_frac(row):\n",
    "    if row['year'] < 2012:\n",
    "        return row['num']/row['CVAP_EST_2009']\n",
    "    else:\n",
    "        return row['num']/row['CVAP_EST_2014']\n",
    "\n",
    "turnout_df['turnout'] = turnout_df.apply(turnout_frac, axis=1)\n",
    "\n",
    "# Mean turnout, 2000-2012\n",
    "turnout2012_df = turnout_df[turnout_df['year'] < 2016].groupby(['county'], as_index=False\n",
    "                        ).agg({'turnout':'mean'})\n",
    "\n",
    "# Turnout, 2016\n",
    "turnout2016_df = turnout_df[turnout_df['year'] == 2016]\n",
    "\n",
    "turnout_df = pd.merge(turnout2012_df, turnout2016_df, on='county', suffixes=('_2000_2012', '_2016'))\n",
    "turnout_df.rename(columns={'num': 'num_2016'}, inplace=True)\n",
    "turnout_df = turnout_df[['county', 'num_2016', 'turnout_2016', 'turnout_2000_2012']]\n",
    "turnout_df['turnout_change'] = turnout_df['turnout_2016'] - turnout_df['turnout_2000_2012']\n",
    "    \n",
    "turnout_df.sort_values(by='num_2016', ascending=False).head(10)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>county</th>\n",
       "      <th>pct_clinton</th>\n",
       "      <th>pct_trump</th>\n",
       "      <th>dem_lead_2016</th>\n",
       "      <th>num_clinton</th>\n",
       "      <th>num_trump</th>\n",
       "      <th>pct_dem</th>\n",
       "      <th>pct_repub</th>\n",
       "      <th>dem_lead_2000_2012</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Menominee</td>\n",
       "      <td>78.4</td>\n",
       "      <td>21.0</td>\n",
       "      <td>57.4</td>\n",
       "      <td>1003.0</td>\n",
       "      <td>269.0</td>\n",
       "      <td>83.225</td>\n",
       "      <td>15.200</td>\n",
       "      <td>68.025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Dane</td>\n",
       "      <td>70.4</td>\n",
       "      <td>23.1</td>\n",
       "      <td>47.3</td>\n",
       "      <td>217506.0</td>\n",
       "      <td>71270.0</td>\n",
       "      <td>67.725</td>\n",
       "      <td>29.725</td>\n",
       "      <td>38.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Milwaukee</td>\n",
       "      <td>65.6</td>\n",
       "      <td>28.6</td>\n",
       "      <td>37.0</td>\n",
       "      <td>288986.0</td>\n",
       "      <td>126091.0</td>\n",
       "      <td>63.675</td>\n",
       "      <td>34.500</td>\n",
       "      <td>29.175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Iowa</td>\n",
       "      <td>54.5</td>\n",
       "      <td>39.3</td>\n",
       "      <td>15.2</td>\n",
       "      <td>6669.0</td>\n",
       "      <td>4809.0</td>\n",
       "      <td>60.900</td>\n",
       "      <td>37.200</td>\n",
       "      <td>23.700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Rock</td>\n",
       "      <td>51.7</td>\n",
       "      <td>41.4</td>\n",
       "      <td>10.3</td>\n",
       "      <td>39336.0</td>\n",
       "      <td>31483.0</td>\n",
       "      <td>60.050</td>\n",
       "      <td>38.150</td>\n",
       "      <td>21.900</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      county  pct_clinton  pct_trump  dem_lead_2016  num_clinton  num_trump  \\\n",
       "0  Menominee         78.4       21.0           57.4       1003.0      269.0   \n",
       "1       Dane         70.4       23.1           47.3     217506.0    71270.0   \n",
       "2  Milwaukee         65.6       28.6           37.0     288986.0   126091.0   \n",
       "3       Iowa         54.5       39.3           15.2       6669.0     4809.0   \n",
       "4       Rock         51.7       41.4           10.3      39336.0    31483.0   \n",
       "\n",
       "   pct_dem  pct_repub  dem_lead_2000_2012  \n",
       "0   83.225     15.200              68.025  \n",
       "1   67.725     29.725              38.000  \n",
       "2   63.675     34.500              29.175  \n",
       "3   60.900     37.200              23.700  \n",
       "4   60.050     38.150              21.900  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate 2016 Democrate-Republican Difference, by county\n",
    "clinton2016_df = voting_df[(voting_df['year'] == 2016) & \n",
    "                           (voting_df['candidate'] == 'Clinton')]\n",
    "\n",
    "trump2016_df = voting_df[(voting_df['year'] == 2016) & \n",
    "                         (voting_df['candidate'] == 'Trump')]\n",
    "\n",
    "spread2016_df = pd.merge(trump2016_df, clinton2016_df, on='county', suffixes=('_trump', '_clinton'))\n",
    "\n",
    "spread2016_df['dem_lead'] = spread2016_df['pct_clinton'] - spread2016_df['pct_trump']\n",
    "\n",
    "spread2016_df = spread2016_df[['county', 'pct_clinton', 'pct_trump', \n",
    "                               'dem_lead', 'num_clinton', 'num_trump']\n",
    "                             ].sort_values(by='dem_lead', ascending=False)\n",
    "\n",
    "\n",
    "# Avg 2000-2012 Democrat-Republican Difference by county \n",
    "dem_df = voting_df[(voting_df['year'] < 2016) & (voting_df['party'] == 'D')]\n",
    "dem_df = dem_df.groupby(['county'], as_index=False\n",
    "                        ).agg({'pct':'mean'})\n",
    "\n",
    "repub_df = voting_df[(voting_df['year'] < 2016) & (voting_df['party'] == 'R')]\n",
    "repub_df = repub_df.groupby(['county'], as_index=False\n",
    "                            ).agg({'pct':'mean'})\n",
    "\n",
    "spread2000_df = pd.merge(dem_df, repub_df[['pct','county']],\n",
    "                         on='county', suffixes=('_dem', '_repub'))\n",
    "spread2000_df['dem_lead'] = spread2000_df['pct_dem'] - spread2000_df['pct_repub']\n",
    "\n",
    "\n",
    "#Merge both together\n",
    "spreadall_df = pd.merge(spread2016_df, spread2000_df, on='county', suffixes=('_2016', '_2000_2012'))\n",
    "spreadall_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The Final Dataframe\n",
    "\n",
    "After all the calculations above, this is the final dataframe.  The relevant columns for the graphics are the `turnout` columns which show the fraction of voting age people that voted in the relevant years, and the `dem_lead` columns which show the margins between parties for various elections.  The `demlead_change` column shows the difference between the Democratic party lead over the 2000-2012 time period and the lead in 2016. "
   ]
  },
  {
   "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>county</th>\n",
       "      <th>num_2016</th>\n",
       "      <th>turnout_2016</th>\n",
       "      <th>turnout_2000_2012</th>\n",
       "      <th>turnout_change</th>\n",
       "      <th>pct_clinton</th>\n",
       "      <th>pct_trump</th>\n",
       "      <th>dem_lead_2016</th>\n",
       "      <th>num_clinton</th>\n",
       "      <th>num_trump</th>\n",
       "      <th>pct_dem</th>\n",
       "      <th>pct_repub</th>\n",
       "      <th>dem_lead_2000_2012</th>\n",
       "      <th>demlead_change</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>Milwaukee</td>\n",
       "      <td>440698.0</td>\n",
       "      <td>0.654155</td>\n",
       "      <td>0.703174</td>\n",
       "      <td>-0.049019</td>\n",
       "      <td>65.6</td>\n",
       "      <td>28.6</td>\n",
       "      <td>37.0</td>\n",
       "      <td>288986.0</td>\n",
       "      <td>126091.0</td>\n",
       "      <td>63.675</td>\n",
       "      <td>34.500</td>\n",
       "      <td>29.175</td>\n",
       "      <td>7.825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Dane</td>\n",
       "      <td>309096.0</td>\n",
       "      <td>0.830200</td>\n",
       "      <td>0.757048</td>\n",
       "      <td>0.073152</td>\n",
       "      <td>70.4</td>\n",
       "      <td>23.1</td>\n",
       "      <td>47.3</td>\n",
       "      <td>217506.0</td>\n",
       "      <td>71270.0</td>\n",
       "      <td>67.725</td>\n",
       "      <td>29.725</td>\n",
       "      <td>38.000</td>\n",
       "      <td>9.300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>Waukesha</td>\n",
       "      <td>237588.0</td>\n",
       "      <td>0.808631</td>\n",
       "      <td>0.801177</td>\n",
       "      <td>0.007454</td>\n",
       "      <td>33.3</td>\n",
       "      <td>60.0</td>\n",
       "      <td>-26.7</td>\n",
       "      <td>79199.0</td>\n",
       "      <td>142519.0</td>\n",
       "      <td>33.125</td>\n",
       "      <td>65.425</td>\n",
       "      <td>-32.300</td>\n",
       "      <td>5.600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Brown</td>\n",
       "      <td>128965.0</td>\n",
       "      <td>0.706793</td>\n",
       "      <td>0.680866</td>\n",
       "      <td>0.025927</td>\n",
       "      <td>41.4</td>\n",
       "      <td>52.1</td>\n",
       "      <td>-10.7</td>\n",
       "      <td>53358.0</td>\n",
       "      <td>67192.0</td>\n",
       "      <td>48.150</td>\n",
       "      <td>49.975</td>\n",
       "      <td>-1.825</td>\n",
       "      <td>-8.875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>Outagamie</td>\n",
       "      <td>95162.0</td>\n",
       "      <td>0.724051</td>\n",
       "      <td>0.686042</td>\n",
       "      <td>0.038009</td>\n",
       "      <td>40.1</td>\n",
       "      <td>54.2</td>\n",
       "      <td>-14.1</td>\n",
       "      <td>38117.0</td>\n",
       "      <td>51579.0</td>\n",
       "      <td>47.750</td>\n",
       "      <td>49.950</td>\n",
       "      <td>-2.200</td>\n",
       "      <td>-11.900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>Racine</td>\n",
       "      <td>94921.0</td>\n",
       "      <td>0.666838</td>\n",
       "      <td>0.686983</td>\n",
       "      <td>-0.020146</td>\n",
       "      <td>44.8</td>\n",
       "      <td>49.1</td>\n",
       "      <td>-4.3</td>\n",
       "      <td>42506.0</td>\n",
       "      <td>46620.0</td>\n",
       "      <td>49.675</td>\n",
       "      <td>48.625</td>\n",
       "      <td>1.050</td>\n",
       "      <td>-5.350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70</th>\n",
       "      <td>Winnebago</td>\n",
       "      <td>87140.0</td>\n",
       "      <td>0.667714</td>\n",
       "      <td>0.678759</td>\n",
       "      <td>-0.011046</td>\n",
       "      <td>42.5</td>\n",
       "      <td>49.9</td>\n",
       "      <td>-7.4</td>\n",
       "      <td>37054.0</td>\n",
       "      <td>43447.0</td>\n",
       "      <td>49.200</td>\n",
       "      <td>48.350</td>\n",
       "      <td>0.850</td>\n",
       "      <td>-8.250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>Washington</td>\n",
       "      <td>77551.0</td>\n",
       "      <td>0.776092</td>\n",
       "      <td>0.741324</td>\n",
       "      <td>0.034768</td>\n",
       "      <td>26.9</td>\n",
       "      <td>66.7</td>\n",
       "      <td>-39.8</td>\n",
       "      <td>20854.0</td>\n",
       "      <td>51729.0</td>\n",
       "      <td>30.700</td>\n",
       "      <td>67.625</td>\n",
       "      <td>-36.925</td>\n",
       "      <td>-2.875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Kenosha</td>\n",
       "      <td>76894.0</td>\n",
       "      <td>0.636830</td>\n",
       "      <td>0.639971</td>\n",
       "      <td>-0.003141</td>\n",
       "      <td>46.5</td>\n",
       "      <td>46.9</td>\n",
       "      <td>-0.4</td>\n",
       "      <td>35770.0</td>\n",
       "      <td>36025.0</td>\n",
       "      <td>54.275</td>\n",
       "      <td>43.800</td>\n",
       "      <td>10.475</td>\n",
       "      <td>-10.875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>Rock</td>\n",
       "      <td>76056.0</td>\n",
       "      <td>0.649524</td>\n",
       "      <td>0.670892</td>\n",
       "      <td>-0.021368</td>\n",
       "      <td>51.7</td>\n",
       "      <td>41.4</td>\n",
       "      <td>10.3</td>\n",
       "      <td>39336.0</td>\n",
       "      <td>31483.0</td>\n",
       "      <td>60.050</td>\n",
       "      <td>38.150</td>\n",
       "      <td>21.900</td>\n",
       "      <td>-11.600</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        county  num_2016  turnout_2016  turnout_2000_2012  turnout_change  \\\n",
       "40   Milwaukee  440698.0      0.654155           0.703174       -0.049019   \n",
       "12        Dane  309096.0      0.830200           0.757048        0.073152   \n",
       "67    Waukesha  237588.0      0.808631           0.801177        0.007454   \n",
       "4        Brown  128965.0      0.706793           0.680866        0.025927   \n",
       "44   Outagamie   95162.0      0.724051           0.686042        0.038009   \n",
       "51      Racine   94921.0      0.666838           0.686983       -0.020146   \n",
       "70   Winnebago   87140.0      0.667714           0.678759       -0.011046   \n",
       "66  Washington   77551.0      0.776092           0.741324        0.034768   \n",
       "29     Kenosha   76894.0      0.636830           0.639971       -0.003141   \n",
       "53        Rock   76056.0      0.649524           0.670892       -0.021368   \n",
       "\n",
       "    pct_clinton  pct_trump  dem_lead_2016  num_clinton  num_trump  pct_dem  \\\n",
       "40         65.6       28.6           37.0     288986.0   126091.0   63.675   \n",
       "12         70.4       23.1           47.3     217506.0    71270.0   67.725   \n",
       "67         33.3       60.0          -26.7      79199.0   142519.0   33.125   \n",
       "4          41.4       52.1          -10.7      53358.0    67192.0   48.150   \n",
       "44         40.1       54.2          -14.1      38117.0    51579.0   47.750   \n",
       "51         44.8       49.1           -4.3      42506.0    46620.0   49.675   \n",
       "70         42.5       49.9           -7.4      37054.0    43447.0   49.200   \n",
       "66         26.9       66.7          -39.8      20854.0    51729.0   30.700   \n",
       "29         46.5       46.9           -0.4      35770.0    36025.0   54.275   \n",
       "53         51.7       41.4           10.3      39336.0    31483.0   60.050   \n",
       "\n",
       "    pct_repub  dem_lead_2000_2012  demlead_change  \n",
       "40     34.500              29.175           7.825  \n",
       "12     29.725              38.000           9.300  \n",
       "67     65.425             -32.300           5.600  \n",
       "4      49.975              -1.825          -8.875  \n",
       "44     49.950              -2.200         -11.900  \n",
       "51     48.625               1.050          -5.350  \n",
       "70     48.350               0.850          -8.250  \n",
       "66     67.625             -36.925          -2.875  \n",
       "29     43.800              10.475         -10.875  \n",
       "53     38.150              21.900         -11.600  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Join with turnout_df\n",
    "turnout_df = pd.merge(turnout_df, spreadall_df, on='county')\n",
    "\n",
    "#Calculate change in Democratic margin, 2016 - average 2000 to 2012\n",
    "turnout_df['demlead_change'] = turnout_df['dem_lead_2016'] - turnout_df['dem_lead_2000_2012']\n",
    "\n",
    "turnout_df.sort_values(by='num_2016', ascending=False, inplace=True)\n",
    "turnout_df.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# An Initial Attempt\n",
    "\n",
    "This was my first attempt at a comprehensive plot that meets all the requirements.  Average `2000-2012` turnout is on the x-axis, and `2016` turnout is on the y-axis.  Any counties above/below the `45 degree` line had better/worse turnout in `2016` relative to the average.  The area of each of the circles is proportional to the number of votes cast, and the color corresponds with the vote margin. \n",
    "\n",
    "It's clear that Milwaukee underperformed in this election.  If Milwaukee voted at it's recent historical average, Wisconsin would probably be a blue state.  \n",
    "\n",
    "Unfortunately, this plot doesn't do good job of communicating the shift in the vote margin relative to the past.  For that, I created the next plot.   \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAucAAAItCAYAAACadNQaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nERG5CCbnREREREQu4v8HIYr4fe7PjEAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x104b03f10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "#http://stackoverflow.com/questions/37401872/custom-continuous-color-map-in-matplotlib\n",
    "import matplotlib.colors as clr\n",
    "\n",
    "#Find midpoint of data on 0-1 scale:\n",
    "vmin = turnout_df['dem_lead_2016'].min()\n",
    "vmax = turnout_df['dem_lead_2016'].max()\n",
    "mid = 1 - vmax/(vmax + abs(vmin))\n",
    "\n",
    "#Construct colormap using midpoint:\n",
    "cmap = clr.LinearSegmentedColormap.from_list('red_blue', \n",
    "                        [(0, '#EF3B2C'), (mid, '#FFFFFF'), (1,'#08519C')], N=256) \n",
    "\n",
    "fig, ax = plt.subplots(figsize=(13,9))  \n",
    "\n",
    "#label='_nolegend_'\n",
    "plt.scatter(x=turnout_df['turnout_2000_2012'], y=turnout_df['turnout_2016'], \n",
    "            s=turnout_df['num_2016']/100, marker='o', alpha=1.0, label='Votes', \n",
    "            c=turnout_df['dem_lead_2016'], cmap=cmap, edgecolors='gray') \n",
    "\n",
    "ax.set_xlim(0.55, 0.85)\n",
    "ax.set_ylim(0.55, 0.85)\n",
    "#s=20\n",
    "\n",
    "# Create X points\n",
    "x = pd.DataFrame({'line': np.linspace(0, 1, 100)})\n",
    "plt.plot(x, x, 'k--', alpha=0.9, label='Equal', color='gray')\n",
    "\n",
    "A = turnout_df['turnout_2000_2012']\n",
    "B = turnout_df['turnout_2016']\n",
    "C = turnout_df['county']\n",
    "D = turnout_df['num_2016']\n",
    "\n",
    "for a,b,c,d in zip(A, B, C, D):\n",
    "    if d > 70000: #Annotate large counties\n",
    "        ax.annotate('%s' % c, xy=(a,b), textcoords='data') \n",
    "        \n",
    "plt.xlabel('Average Voter Turnout, 2000-2012 (% VAP)')\n",
    "plt.ylabel('Voter Turnout, 2016 (% VAP)')\n",
    "\n",
    "legend = plt.legend(loc='upper left')\n",
    "legend.legendHandles[1]._sizes = [40]\n",
    "\n",
    "plt.colorbar(shrink=0.5, pad=0.03, label='Democratic Margin', format='%.0f%%')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# A Better Approach \n",
    "\n",
    "Ok, here is a different approach that matches all of my initial criteria for a visualization.\n",
    "\n",
    "It's easy to build a graphic that shows a difference relative to Obama, but that's not very informative. Wisconsin has voted for the Democratic nominee in every election since 1984, so Trump's win should represent a deviation from deeper historical averages, not just a deviation from Obama's results.  That's what the axes of this graphic try to show, with the x and y axes showing changes from the 2000-2012 averages.  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAu0AAAInCAYAAAAszWe1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1076984d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(13.5,9))  #figsize=(12,10)\n",
    "\n",
    "plt.scatter(x=turnout_df['demlead_change'], y=turnout_df['turnout_change']*100, \n",
    "            s=turnout_df['num_2016']/100, marker='o', alpha=1.0, label='Votes', \n",
    "            c=turnout_df['dem_lead_2016'], cmap=cmap, edgecolors='gray')\n",
    "\n",
    "#http://stackoverflow.com/questions/31556446\n",
    "# Move left y-axis and bottim x-axis to zero:\n",
    "ax.spines['bottom'].set_position('zero')\n",
    "ax.spines['left'].set_position('zero')\n",
    "\n",
    "# Eliminate upper and right axes\n",
    "ax.spines['right'].set_color('none')\n",
    "ax.spines['top'].set_color('none')\n",
    "\n",
    "# Show ticks in the left and lower axes only\n",
    "ax.xaxis.set_ticks_position('bottom')\n",
    "ax.yaxis.set_ticks_position('left')\n",
    "\n",
    "ax.set_ylim(-10, 10)\n",
    "\n",
    "\n",
    "A = turnout_df['demlead_change']\n",
    "B = turnout_df['turnout_change']*100\n",
    "C = turnout_df['county']\n",
    "D = turnout_df['num_2016']\n",
    "\n",
    "for a,b,c,d in zip(A, B, C, D):\n",
    "    if d > 70000: #Annotate large counties  >90000\n",
    "        ax.annotate('%s' % c, xy=(a,b), textcoords='data') \n",
    "        \n",
    "plt.xlabel('Change from Average 2000-2012 Democratic Margin (%)', labelpad=250)\n",
    "plt.ylabel('Change from Average 2000-2012 Turnout (%)', labelpad=400)\n",
    "\n",
    "legend = plt.legend(loc='upper left')\n",
    "legend.legendHandles[0]._sizes = [40]\n",
    "\n",
    "#http://stackoverflow.com/questions/5306756\n",
    "plt.colorbar(shrink=0.5, pad=0.03, label='Democratic Margin', format='%.0f%%') #orientation='horizontal'\n",
    "\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# So What Happened?\n",
    "\n",
    "The size and colors of the circles do a fairly good job of communicating what happened this election.  The area of the circle is proportional to the votes cast, and the color communicates the margin for the Democrats so it's pretty easy to get an idea of the crucial counties and the role they played for each party.  \n",
    "\n",
    "The axes in this plot show what changed relative to the past.  On the x-axis, I have the change from the average democratic margin, which does a good job of communicating the shift in support within a county.  This axis makes it clear that there was a massive shift to the Republican party in small rural counties.  There was even a shift towards Trump in many larger suburban counties, even if they were still won by Clinton.    \n",
    "\n",
    "Interestingly, Waukesha and Ozaukee counties, which are fairly rich, well educated suburban counties of Milwaukee, actually saw `5-10%` shifts towards the Democrats.  This fits with Nate Silver's analysis that suggested [level of education was the most important predictor](http://fivethirtyeight.com/features/education-not-income-predicted-who-would-vote-for-trump/) of voting behavior in this election.  Waukesha county is one of the most conservative counties in the country, but it's dominated by establishment conservatives that may have clashed with Trump's views.     \n",
    "\n",
    "The y-axis shows `change in voter turnout` relative to the `2000-2012` average.  It's clear that Milwaukee really underperformed here, with a `5%` drop in turnout relative to the average.  Part of this is probably because Obama wasn't on the ballot, but the historical average includes the Gore and Kerry elections as well, which should help mitigate the Obama effect.  \n",
    "\n",
    "One thing I haven't heard anyone discuss is the fact that the Democrat's margin in Milwaukee was up by `7.8%` compared with the 2000-2012 average.  A margin increase of `7.8%` would lead to a larger net gain in votes than an increase in turnout of the same percent, so this margin increase probably canceled out the effects of lower turnout (although it would be ideal to have increases in both).          \n",
    "\n",
    "Dane County had an amazing `7.3%` increase in turnout along with a large `10%` increase in Democratic support.  This was almost enough to cancel out the poorer performance elsewhere in the state.  But in the end, the large shift to Trump in rural counties overwhelmed the Dane County effect. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Why did Dane County do so much better?\n",
    "\n",
    "Dane County performed much better than Milwaukee County when it comes to voter turnout and margins.  It might make sense to try to apply lessons learned from Dane County to Milwaukee, although they are two very different places.  There are much higher levels of poverty in Milwaukee, which might magnify the effects of the recent voter ID laws.  Anecdotally, there seemed to be less enthusiasm for [Clinton](http://www.nytimes.com/2016/11/21/us/many-in-milwaukee-neighborhood-didnt-vote-and-dont-regret-it.html) as well.     \n",
    "\n",
    "Madison is a younger, more progressive city, and probably had more resources to put into get out the vote efforts.  It also has a large student population and one of the top public universities in the world.  For more thoughts on the causes of the voting disparities, see this [article](http://host.madison.com/ct/news/local/govt-and-politics/election-matters/why-did-wisconsin-see-its-lowest-presidential-election-voter-turnout/article_6dd2887f-e1fc-5ed8-a454-284d37204669.html). \n",
    "\n",
    "Whatever the cause, I think it makes sense to study the differences between the two counties to look for ways to improve in the future.  Things are only going to get more difficult over the coming years with more voter suppression \n",
    "efforts likely at the state and now national level. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "\n",
    "#Data output script for interactive visualization.  \n",
    "\n",
    "'''\n",
    "# Calculate voter turnouts by county\n",
    "out_df = pd.pivot_table(voting_df, values='num', index=['year', 'county'], aggfunc=np.sum)\n",
    "out_df = out_df.reset_index()\n",
    "\n",
    "out_df = pd.merge(out_df, population2009_df, on='county', suffixes=('', '_2009'))\n",
    "out_df = pd.merge(out_df, population2014_df, on='county', suffixes=('', '_2014'))\n",
    "out_df.rename(columns={'CVAP_EST': 'CVAP_EST_2009',\n",
    "                       'num': 'county_num',\n",
    "                       'CVAP_MOE': 'CVAP_MOE_2009'}, inplace=True)\n",
    "\n",
    "def turnout_frac(row):\n",
    "    if row['year'] < 2012:\n",
    "        return row['county_num']/row['CVAP_EST_2009']\n",
    "    else:\n",
    "        return row['county_num']/row['CVAP_EST_2014']\n",
    "\n",
    "out_df['turnout'] = out_df.apply(turnout_frac, axis=1)\n",
    "\n",
    "out_df = out_df[['year', 'county', 'county_num', 'turnout']]\n",
    "\n",
    "#out_df.head(30)\n",
    "\n",
    "#out_df.head(30)\n",
    "\n",
    "#Calculate 2016 Democrate-Republican Difference, by county\n",
    "demout_df = voting_df[(voting_df['party'] == 'D')]\n",
    "\n",
    "repout_df = voting_df[(voting_df['party'] == 'R')]\n",
    "\n",
    "spreadout_df = pd.merge(repout_df, demout_df, on=['year', 'county'], suffixes=('_rep', '_dem'))\n",
    "\n",
    "\n",
    "#spreadout_df.head(20)\n",
    "\n",
    "spreadout_df.head()\n",
    "\n",
    "spreadout_df['dem_lead'] = spreadout_df['pct_dem'] - spreadout_df['pct_rep']\n",
    "\n",
    "spreadout_df = spreadout_df[['year', 'county', 'pct_dem', 'pct_rep', \n",
    "                               'dem_lead', 'num_dem', 'num_rep']\n",
    "                             ]  #.sort_values(by='dem_lead', ascending=False)\n",
    "\n",
    "out_df = pd.merge(out_df, spreadout_df, on=['year', 'county'])\n",
    "#out_df.round(4) #{'turnout':4}\n",
    "#np.round(out_df, decimals = 4)\n",
    "#Round isn't working. . . \n",
    "out_df['turnout'] = out_df['turnout'].apply(lambda x: round(x, 4) )\n",
    "\n",
    "#out_df.head()  #.dtypes #.head()\n",
    "\n",
    "#out_df = out_df.groupby(['year', 'county'])\n",
    "#out_df.head(30)\n",
    "#out_df.tail(30)\n",
    "\n",
    "#Output to current directory:\n",
    "out_df.to_csv('./county_results_20002016.csv', index=False)\n",
    "'''\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}