science4performance/TdF Oddschecker.ipynb Secret

## TdF Oddschecker.ipynb
{
  "cells": [
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "# TdF Oddschecker\n- I tried to scrape thie historic odds from oddchecker url, but this was blocked at their end. So I viewed the page source and cut & pasted the betfair histories for the three main contenders.\n- I refered to procyclingstats.com for race results"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import pandas as pd\nimport matplotlib.pylab as plt\n%matplotlib inline",
      "execution_count": 141,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "rider = 'Egan Bernal'\n\nurl = f\"https://www.oddschecker.com/cycling/tour-de-france/winner/bet-history/{rider.lower().replace(' ','-')}#all-history\"\nurl\n",
      "execution_count": 241,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 241,
          "data": {
            "text/plain": "'https://www.oddschecker.com/cycling/tour-de-france/winner/bet-history/egan-bernal#all-history'"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "names = ['Egan Bernal', 'Geraint Thomas', 'Jakob Fuglsang']",
      "execution_count": 174,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Can't read it in automatically so load pages manually into variables called Bernal, Thomas and Fuglsang"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "dfRiders = pd.DataFrame(columns=['Date','Name','DecOdds','Probability'])\n\nfor i,n in enumerate([Bernal,Thomas,Fuglsang]):\n    data = n[25:-20].replace('Date.UTC','').replace(' - 1','').replace('], [','],[')\n    data = data.split('],[')\n    x = [[pd.to_datetime('-'.join(d[1:].split('),')[0].split(',')[:3])), float(d.split('),')[1])] for d in data]\n    df = pd.DataFrame(x,columns=['Date','DecOdds'])\n    df['Probability'] = 100/df.DecOdds\n    df['Name'] = names[i]\n    \n    #df = pd.pivot_table(df,values=['DecOdds','Probability'],index=['Name','Date'],aggfunc='mean')\n    dfRiders = dfRiders.append(df)\n    print(names[i])\n\n    ",
      "execution_count": 178,
      "outputs": [
        {
          "output_type": "stream",
          "text": "/Users/Gavin/anaconda3/lib/python3.7/site-packages/pandas/core/frame.py:6692: FutureWarning: Sorting because non-concatenation axis is not aligned. A future version\nof pandas will change to not sort by default.\n\nTo accept the future behavior, pass 'sort=False'.\n\nTo retain the current behavior and silence the warning, pass 'sort=True'.\n\n  sort=sort)\n",
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": "Egan Bernal\nGeraint Thomas\nJakob Fuglsang\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "dfRiders.head()",
      "execution_count": 179,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 179,
          "data": {
            "text/plain": "        Date  DecOdds         Name  Probability\n0 2019-07-05     3.71  Egan Bernal    26.954178\n1 2019-07-05     3.71  Egan Bernal    26.954178\n2 2019-07-05     3.66  Egan Bernal    27.322404\n3 2019-07-05     3.61  Egan Bernal    27.700831\n4 2019-07-05     3.66  Egan Bernal    27.322404",
            "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>Date</th>\n      <th>DecOdds</th>\n      <th>Name</th>\n      <th>Probability</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>2019-07-05</td>\n      <td>3.71</td>\n      <td>Egan Bernal</td>\n      <td>26.954178</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>2019-07-05</td>\n      <td>3.71</td>\n      <td>Egan Bernal</td>\n      <td>26.954178</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>2019-07-05</td>\n      <td>3.66</td>\n      <td>Egan Bernal</td>\n      <td>27.322404</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>2019-07-05</td>\n      <td>3.61</td>\n      <td>Egan Bernal</td>\n      <td>27.700831</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>2019-07-05</td>\n      <td>3.66</td>\n      <td>Egan Bernal</td>\n      <td>27.322404</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "x = pd.pivot_table(dfRiders[dfRiders.Date>pd.to_datetime('2018-12-31')],\n               values=['Probability'],index=['Name','Date'],\n               aggfunc='mean')\nx.head()",
      "execution_count": 180,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 180,
          "data": {
            "text/plain": "                        Probability\nName        Date                   \nEgan Bernal 2019-01-01     4.391234\n            2019-01-03     3.769841\n            2019-01-05     4.332738\n            2019-01-08     3.769841\n            2019-01-10     2.631579",
            "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></th>\n      <th>Probability</th>\n    </tr>\n    <tr>\n      <th>Name</th>\n      <th>Date</th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th rowspan=\"5\" valign=\"top\">Egan Bernal</th>\n      <th>2019-01-01</th>\n      <td>4.391234</td>\n    </tr>\n    <tr>\n      <th>2019-01-03</th>\n      <td>3.769841</td>\n    </tr>\n    <tr>\n      <th>2019-01-05</th>\n      <td>4.332738</td>\n    </tr>\n    <tr>\n      <th>2019-01-08</th>\n      <td>3.769841</td>\n    </tr>\n    <tr>\n      <th>2019-01-10</th>\n      <td>2.631579</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "font = {'family' : 'normal',\n        'weight' : 'bold',\n        'size'   : 16}\n\nplt.rc('font', **font)",
      "execution_count": 231,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true,
        "scrolled": false
      },
      "cell_type": "code",
      "source": "fig,ax = plt.subplots(figsize=(12,8))\n\nfor n in names:\n    x.loc[n].plot(ax=ax)\nax.legend(names, loc=2);\nax.set_ylim([0, 50]);\nax.set_title('Implied probability of winning the Tour de France 2019\\n Top three contenders',\n            fontsize=20, fontweight= 'bold')\n\nplt.savefig('TdFprob.jpg')",
      "execution_count": 239,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 864x576 with 1 Axes>",
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Fuglsand spikes\n- Third in Tirreno-Adriatico 19 March with win on stage 5 on 17 March\n- Strong finish in Itzulia Basque Country 13 April\n- Not much reaction to great results (1st, 2nd and 3rd) in Ardenne Classics 21-28 April\n- Strong performance to win Critérium du Dauphiné on 16 June\n"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "x.loc['Jakob Fuglsang']['20190318':'20190630']",
      "execution_count": 221,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 221,
          "data": {
            "text/plain": "            Probability\nDate                   \n2019-03-18     2.134503\n2019-03-19    51.171617\n2019-03-20    99.009901\n2019-03-21    41.603960\n2019-03-23    51.171617\n2019-03-24     3.134804\n2019-03-25     7.610010\n2019-03-26    16.584416\n2019-03-27    10.324675\n2019-03-28    14.237013\n2019-03-29    25.974026\n2019-04-01    14.653680\n2019-04-02    14.653680\n2019-04-03    14.568564\n2019-04-04    14.653680\n2019-04-07    25.974026\n2019-04-08     3.571429\n2019-04-09     2.173913\n2019-04-12    44.879976\n2019-04-13    10.000000\n2019-04-17     9.090909\n2019-04-19     4.166667\n2019-04-20     6.628788\n2019-04-21     6.628788\n2019-04-22     6.334461\n2019-04-25     3.451013\n2019-04-26     3.814252\n2019-04-27     6.369427\n2019-04-28     6.254767\n2019-04-29     5.928567\n...                 ...\n2019-05-28     3.184268\n2019-05-30     3.774929\n2019-06-02     3.711719\n2019-06-03     3.989843\n2019-06-05     4.025765\n2019-06-06     3.609769\n2019-06-07     4.010046\n2019-06-08     4.021874\n2019-06-09     5.211395\n2019-06-10     8.858791\n2019-06-11     8.692504\n2019-06-12     8.114440\n2019-06-13    11.374088\n2019-06-14     9.845894\n2019-06-15    11.477603\n2019-06-16    11.813409\n2019-06-17    12.157287\n2019-06-18    16.701895\n2019-06-19    13.322818\n2019-06-20    15.722993\n2019-06-21    17.200840\n2019-06-22    13.075930\n2019-06-23    12.758100\n2019-06-24    13.021786\n2019-06-25    14.732238\n2019-06-26    13.755158\n2019-06-27    13.579992\n2019-06-28    13.404826\n2019-06-29    13.028945\n2019-06-30    13.238361\n\n[80 rows x 1 columns]",
            "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>Probability</th>\n    </tr>\n    <tr>\n      <th>Date</th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>2019-03-18</th>\n      <td>2.134503</td>\n    </tr>\n    <tr>\n      <th>2019-03-19</th>\n      <td>51.171617</td>\n    </tr>\n    <tr>\n      <th>2019-03-20</th>\n      <td>99.009901</td>\n    </tr>\n    <tr>\n      <th>2019-03-21</th>\n      <td>41.603960</td>\n    </tr>\n    <tr>\n      <th>2019-03-23</th>\n      <td>51.171617</td>\n    </tr>\n    <tr>\n      <th>2019-03-24</th>\n      <td>3.134804</td>\n    </tr>\n    <tr>\n      <th>2019-03-25</th>\n      <td>7.610010</td>\n    </tr>\n    <tr>\n      <th>2019-03-26</th>\n      <td>16.584416</td>\n    </tr>\n    <tr>\n      <th>2019-03-27</th>\n      <td>10.324675</td>\n    </tr>\n    <tr>\n      <th>2019-03-28</th>\n      <td>14.237013</td>\n    </tr>\n    <tr>\n      <th>2019-03-29</th>\n      <td>25.974026</td>\n    </tr>\n    <tr>\n      <th>2019-04-01</th>\n      <td>14.653680</td>\n    </tr>\n    <tr>\n      <th>2019-04-02</th>\n      <td>14.653680</td>\n    </tr>\n    <tr>\n      <th>2019-04-03</th>\n      <td>14.568564</td>\n    </tr>\n    <tr>\n      <th>2019-04-04</th>\n      <td>14.653680</td>\n    </tr>\n    <tr>\n      <th>2019-04-07</th>\n      <td>25.974026</td>\n    </tr>\n    <tr>\n      <th>2019-04-08</th>\n      <td>3.571429</td>\n    </tr>\n    <tr>\n      <th>2019-04-09</th>\n      <td>2.173913</td>\n    </tr>\n    <tr>\n      <th>2019-04-12</th>\n      <td>44.879976</td>\n    </tr>\n    <tr>\n      <th>2019-04-13</th>\n      <td>10.000000</td>\n    </tr>\n    <tr>\n      <th>2019-04-17</th>\n      <td>9.090909</td>\n    </tr>\n    <tr>\n      <th>2019-04-19</th>\n      <td>4.166667</td>\n    </tr>\n    <tr>\n      <th>2019-04-20</th>\n      <td>6.628788</td>\n    </tr>\n    <tr>\n      <th>2019-04-21</th>\n      <td>6.628788</td>\n    </tr>\n    <tr>\n      <th>2019-04-22</th>\n      <td>6.334461</td>\n    </tr>\n    <tr>\n      <th>2019-04-25</th>\n      <td>3.451013</td>\n    </tr>\n    <tr>\n      <th>2019-04-26</th>\n      <td>3.814252</td>\n    </tr>\n    <tr>\n      <th>2019-04-27</th>\n      <td>6.369427</td>\n    </tr>\n    <tr>\n      <th>2019-04-28</th>\n      <td>6.254767</td>\n    </tr>\n    <tr>\n      <th>2019-04-29</th>\n      <td>5.928567</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>2019-05-28</th>\n      <td>3.184268</td>\n    </tr>\n    <tr>\n      <th>2019-05-30</th>\n      <td>3.774929</td>\n    </tr>\n    <tr>\n      <th>2019-06-02</th>\n      <td>3.711719</td>\n    </tr>\n    <tr>\n      <th>2019-06-03</th>\n      <td>3.989843</td>\n    </tr>\n    <tr>\n      <th>2019-06-05</th>\n      <td>4.025765</td>\n    </tr>\n    <tr>\n      <th>2019-06-06</th>\n      <td>3.609769</td>\n    </tr>\n    <tr>\n      <th>2019-06-07</th>\n      <td>4.010046</td>\n    </tr>\n    <tr>\n      <th>2019-06-08</th>\n      <td>4.021874</td>\n    </tr>\n    <tr>\n      <th>2019-06-09</th>\n      <td>5.211395</td>\n    </tr>\n    <tr>\n      <th>2019-06-10</th>\n      <td>8.858791</td>\n    </tr>\n    <tr>\n      <th>2019-06-11</th>\n      <td>8.692504</td>\n    </tr>\n    <tr>\n      <th>2019-06-12</th>\n      <td>8.114440</td>\n    </tr>\n    <tr>\n      <th>2019-06-13</th>\n      <td>11.374088</td>\n    </tr>\n    <tr>\n      <th>2019-06-14</th>\n      <td>9.845894</td>\n    </tr>\n    <tr>\n      <th>2019-06-15</th>\n      <td>11.477603</td>\n    </tr>\n    <tr>\n      <th>2019-06-16</th>\n      <td>11.813409</td>\n    </tr>\n    <tr>\n      <th>2019-06-17</th>\n      <td>12.157287</td>\n    </tr>\n    <tr>\n      <th>2019-06-18</th>\n      <td>16.701895</td>\n    </tr>\n    <tr>\n      <th>2019-06-19</th>\n      <td>13.322818</td>\n    </tr>\n    <tr>\n      <th>2019-06-20</th>\n      <td>15.722993</td>\n    </tr>\n    <tr>\n      <th>2019-06-21</th>\n      <td>17.200840</td>\n    </tr>\n    <tr>\n      <th>2019-06-22</th>\n      <td>13.075930</td>\n    </tr>\n    <tr>\n      <th>2019-06-23</th>\n      <td>12.758100</td>\n    </tr>\n    <tr>\n      <th>2019-06-24</th>\n      <td>13.021786</td>\n    </tr>\n    <tr>\n      <th>2019-06-25</th>\n      <td>14.732238</td>\n    </tr>\n    <tr>\n      <th>2019-06-26</th>\n      <td>13.755158</td>\n    </tr>\n    <tr>\n      <th>2019-06-27</th>\n      <td>13.579992</td>\n    </tr>\n    <tr>\n      <th>2019-06-28</th>\n      <td>13.404826</td>\n    </tr>\n    <tr>\n      <th>2019-06-29</th>\n      <td>13.028945</td>\n    </tr>\n    <tr>\n      <th>2019-06-30</th>\n      <td>13.238361</td>\n    </tr>\n  </tbody>\n</table>\n<p>80 rows × 1 columns</p>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Froome's crash was on 11 June 2019 - Bernal and Thomas's odds shortened"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "x.swaplevel().sort_index().loc[pd.to_datetime('20190610'):pd.to_datetime('20190614')]",
      "execution_count": 212,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 212,
          "data": {
            "text/plain": "                           Probability\nDate       Name                       \n2019-06-10 Egan Bernal        5.394010\n           Geraint Thomas    21.925669\n           Jakob Fuglsang     8.858791\n2019-06-11 Egan Bernal        5.256113\n           Geraint Thomas    21.046419\n           Jakob Fuglsang     8.692504\n2019-06-12 Egan Bernal       15.945886\n           Geraint Thomas    32.543901\n           Jakob Fuglsang     8.114440\n2019-06-13 Egan Bernal       15.500936\n           Geraint Thomas    31.541214\n           Jakob Fuglsang    11.374088\n2019-06-14 Egan Bernal       13.012420\n           Geraint Thomas    29.441244\n           Jakob Fuglsang     9.845894",
            "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></th>\n      <th>Probability</th>\n    </tr>\n    <tr>\n      <th>Date</th>\n      <th>Name</th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th rowspan=\"3\" valign=\"top\">2019-06-10</th>\n      <th>Egan Bernal</th>\n      <td>5.394010</td>\n    </tr>\n    <tr>\n      <th>Geraint Thomas</th>\n      <td>21.925669</td>\n    </tr>\n    <tr>\n      <th>Jakob Fuglsang</th>\n      <td>8.858791</td>\n    </tr>\n    <tr>\n      <th rowspan=\"3\" valign=\"top\">2019-06-11</th>\n      <th>Egan Bernal</th>\n      <td>5.256113</td>\n    </tr>\n    <tr>\n      <th>Geraint Thomas</th>\n      <td>21.046419</td>\n    </tr>\n    <tr>\n      <th>Jakob Fuglsang</th>\n      <td>8.692504</td>\n    </tr>\n    <tr>\n      <th rowspan=\"3\" valign=\"top\">2019-06-12</th>\n      <th>Egan Bernal</th>\n      <td>15.945886</td>\n    </tr>\n    <tr>\n      <th>Geraint Thomas</th>\n      <td>32.543901</td>\n    </tr>\n    <tr>\n      <th>Jakob Fuglsang</th>\n      <td>8.114440</td>\n    </tr>\n    <tr>\n      <th rowspan=\"3\" valign=\"top\">2019-06-13</th>\n      <th>Egan Bernal</th>\n      <td>15.500936</td>\n    </tr>\n    <tr>\n      <th>Geraint Thomas</th>\n      <td>31.541214</td>\n    </tr>\n    <tr>\n      <th>Jakob Fuglsang</th>\n      <td>11.374088</td>\n    </tr>\n    <tr>\n      <th rowspan=\"3\" valign=\"top\">2019-06-14</th>\n      <th>Egan Bernal</th>\n      <td>13.012420</td>\n    </tr>\n    <tr>\n      <th>Geraint Thomas</th>\n      <td>29.441244</td>\n    </tr>\n    <tr>\n      <th>Jakob Fuglsang</th>\n      <td>9.845894</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Dumoulin's crash was on 15 May 2019 - no impact"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "x.swaplevel().sort_index().loc[pd.to_datetime('20190513'):pd.to_datetime('20190518')]",
      "execution_count": 226,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 226,
          "data": {
            "text/plain": "                           Probability\nDate       Name                       \n2019-05-13 Egan Bernal       11.098674\n           Geraint Thomas    23.429585\n2019-05-14 Egan Bernal       13.889762\n           Geraint Thomas    22.886291\n           Jakob Fuglsang     2.631579\n2019-05-15 Egan Bernal        6.785147\n           Geraint Thomas    22.172949\n           Jakob Fuglsang     3.846154\n2019-05-16 Egan Bernal        6.108690\n           Geraint Thomas    21.262654\n2019-05-17 Geraint Thomas    21.276596\n           Jakob Fuglsang     4.096990\n2019-05-18 Jakob Fuglsang     4.347826",
            "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></th>\n      <th>Probability</th>\n    </tr>\n    <tr>\n      <th>Date</th>\n      <th>Name</th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th rowspan=\"2\" valign=\"top\">2019-05-13</th>\n      <th>Egan Bernal</th>\n      <td>11.098674</td>\n    </tr>\n    <tr>\n      <th>Geraint Thomas</th>\n      <td>23.429585</td>\n    </tr>\n    <tr>\n      <th rowspan=\"3\" valign=\"top\">2019-05-14</th>\n      <th>Egan Bernal</th>\n      <td>13.889762</td>\n    </tr>\n    <tr>\n      <th>Geraint Thomas</th>\n      <td>22.886291</td>\n    </tr>\n    <tr>\n      <th>Jakob Fuglsang</th>\n      <td>2.631579</td>\n    </tr>\n    <tr>\n      <th rowspan=\"3\" valign=\"top\">2019-05-15</th>\n      <th>Egan Bernal</th>\n      <td>6.785147</td>\n    </tr>\n    <tr>\n      <th>Geraint Thomas</th>\n      <td>22.172949</td>\n    </tr>\n    <tr>\n      <th>Jakob Fuglsang</th>\n      <td>3.846154</td>\n    </tr>\n    <tr>\n      <th rowspan=\"2\" valign=\"top\">2019-05-16</th>\n      <th>Egan Bernal</th>\n      <td>6.108690</td>\n    </tr>\n    <tr>\n      <th>Geraint Thomas</th>\n      <td>21.262654</td>\n    </tr>\n    <tr>\n      <th rowspan=\"2\" valign=\"top\">2019-05-17</th>\n      <th>Geraint Thomas</th>\n      <td>21.276596</td>\n    </tr>\n    <tr>\n      <th>Jakob Fuglsang</th>\n      <td>4.096990</td>\n    </tr>\n    <tr>\n      <th>2019-05-18</th>\n      <th>Jakob Fuglsang</th>\n      <td>4.347826</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Geraint Thomas \n- Boosted a bit by decent performance 3rd in Tour of Romandie\n- Crash in Tour of Switzerland on 18 June 2019"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "\nx.loc['Geraint Thomas']['20190428':'20190630']",
      "execution_count": 223,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 223,
          "data": {
            "text/plain": "            Probability\nDate                   \n2019-04-28    22.624434\n2019-04-29    23.394405\n2019-04-30    24.510686\n2019-05-01    29.585799\n2019-05-02    28.218530\n2019-05-03    24.196569\n2019-05-06    22.398692\n2019-05-07    21.484285\n2019-05-08    23.640662\n2019-05-09    23.148148\n2019-05-10    23.640662\n2019-05-12    23.350839\n2019-05-13    23.429585\n2019-05-14    22.886291\n2019-05-15    22.172949\n2019-05-16    21.262654\n2019-05-17    21.276596\n2019-05-19    20.040080\n2019-05-20    21.061297\n2019-05-21    20.981087\n2019-05-23    20.641616\n2019-05-26    20.244989\n2019-05-27    19.475692\n2019-05-28    18.621974\n2019-05-30    19.122207\n2019-05-31    18.307535\n2019-06-01    17.667845\n2019-06-02    18.307535\n2019-06-03    18.885551\n2019-06-04    18.098623\n2019-06-05    18.955462\n2019-06-06    18.413341\n2019-06-07    17.812703\n2019-06-08    19.058871\n2019-06-09    19.417841\n2019-06-10    21.925669\n2019-06-11    21.046419\n2019-06-12    32.543901\n2019-06-13    31.541214\n2019-06-14    29.441244\n2019-06-15    30.800820\n2019-06-16    29.226069\n2019-06-17    30.361810\n2019-06-18    24.899569\n2019-06-19    25.324326\n2019-06-20    22.785551\n2019-06-21    25.825977\n2019-06-22    25.759938\n2019-06-23    25.429566\n2019-06-24    25.736700\n2019-06-25    27.603150\n2019-06-26    27.859484\n2019-06-27    27.597718\n2019-06-28    27.567127\n2019-06-29    26.670271\n2019-06-30    25.054070",
            "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>Probability</th>\n    </tr>\n    <tr>\n      <th>Date</th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>2019-04-28</th>\n      <td>22.624434</td>\n    </tr>\n    <tr>\n      <th>2019-04-29</th>\n      <td>23.394405</td>\n    </tr>\n    <tr>\n      <th>2019-04-30</th>\n      <td>24.510686</td>\n    </tr>\n    <tr>\n      <th>2019-05-01</th>\n      <td>29.585799</td>\n    </tr>\n    <tr>\n      <th>2019-05-02</th>\n      <td>28.218530</td>\n    </tr>\n    <tr>\n      <th>2019-05-03</th>\n      <td>24.196569</td>\n    </tr>\n    <tr>\n      <th>2019-05-06</th>\n      <td>22.398692</td>\n    </tr>\n    <tr>\n      <th>2019-05-07</th>\n      <td>21.484285</td>\n    </tr>\n    <tr>\n      <th>2019-05-08</th>\n      <td>23.640662</td>\n    </tr>\n    <tr>\n      <th>2019-05-09</th>\n      <td>23.148148</td>\n    </tr>\n    <tr>\n      <th>2019-05-10</th>\n      <td>23.640662</td>\n    </tr>\n    <tr>\n      <th>2019-05-12</th>\n      <td>23.350839</td>\n    </tr>\n    <tr>\n      <th>2019-05-13</th>\n      <td>23.429585</td>\n    </tr>\n    <tr>\n      <th>2019-05-14</th>\n      <td>22.886291</td>\n    </tr>\n    <tr>\n      <th>2019-05-15</th>\n      <td>22.172949</td>\n    </tr>\n    <tr>\n      <th>2019-05-16</th>\n      <td>21.262654</td>\n    </tr>\n    <tr>\n      <th>2019-05-17</th>\n      <td>21.276596</td>\n    </tr>\n    <tr>\n      <th>2019-05-19</th>\n      <td>20.040080</td>\n    </tr>\n    <tr>\n      <th>2019-05-20</th>\n      <td>21.061297</td>\n    </tr>\n    <tr>\n      <th>2019-05-21</th>\n      <td>20.981087</td>\n    </tr>\n    <tr>\n      <th>2019-05-23</th>\n      <td>20.641616</td>\n    </tr>\n    <tr>\n      <th>2019-05-26</th>\n      <td>20.244989</td>\n    </tr>\n    <tr>\n      <th>2019-05-27</th>\n      <td>19.475692</td>\n    </tr>\n    <tr>\n      <th>2019-05-28</th>\n      <td>18.621974</td>\n    </tr>\n    <tr>\n      <th>2019-05-30</th>\n      <td>19.122207</td>\n    </tr>\n    <tr>\n      <th>2019-05-31</th>\n      <td>18.307535</td>\n    </tr>\n    <tr>\n      <th>2019-06-01</th>\n      <td>17.667845</td>\n    </tr>\n    <tr>\n      <th>2019-06-02</th>\n      <td>18.307535</td>\n    </tr>\n    <tr>\n      <th>2019-06-03</th>\n      <td>18.885551</td>\n    </tr>\n    <tr>\n      <th>2019-06-04</th>\n      <td>18.098623</td>\n    </tr>\n    <tr>\n      <th>2019-06-05</th>\n      <td>18.955462</td>\n    </tr>\n    <tr>\n      <th>2019-06-06</th>\n      <td>18.413341</td>\n    </tr>\n    <tr>\n      <th>2019-06-07</th>\n      <td>17.812703</td>\n    </tr>\n    <tr>\n      <th>2019-06-08</th>\n      <td>19.058871</td>\n    </tr>\n    <tr>\n      <th>2019-06-09</th>\n      <td>19.417841</td>\n    </tr>\n    <tr>\n      <th>2019-06-10</th>\n      <td>21.925669</td>\n    </tr>\n    <tr>\n      <th>2019-06-11</th>\n      <td>21.046419</td>\n    </tr>\n    <tr>\n      <th>2019-06-12</th>\n      <td>32.543901</td>\n    </tr>\n    <tr>\n      <th>2019-06-13</th>\n      <td>31.541214</td>\n    </tr>\n    <tr>\n      <th>2019-06-14</th>\n      <td>29.441244</td>\n    </tr>\n    <tr>\n      <th>2019-06-15</th>\n      <td>30.800820</td>\n    </tr>\n    <tr>\n      <th>2019-06-16</th>\n      <td>29.226069</td>\n    </tr>\n    <tr>\n      <th>2019-06-17</th>\n      <td>30.361810</td>\n    </tr>\n    <tr>\n      <th>2019-06-18</th>\n      <td>24.899569</td>\n    </tr>\n    <tr>\n      <th>2019-06-19</th>\n      <td>25.324326</td>\n    </tr>\n    <tr>\n      <th>2019-06-20</th>\n      <td>22.785551</td>\n    </tr>\n    <tr>\n      <th>2019-06-21</th>\n      <td>25.825977</td>\n    </tr>\n    <tr>\n      <th>2019-06-22</th>\n      <td>25.759938</td>\n    </tr>\n    <tr>\n      <th>2019-06-23</th>\n      <td>25.429566</td>\n    </tr>\n    <tr>\n      <th>2019-06-24</th>\n      <td>25.736700</td>\n    </tr>\n    <tr>\n      <th>2019-06-25</th>\n      <td>27.603150</td>\n    </tr>\n    <tr>\n      <th>2019-06-26</th>\n      <td>27.859484</td>\n    </tr>\n    <tr>\n      <th>2019-06-27</th>\n      <td>27.597718</td>\n    </tr>\n    <tr>\n      <th>2019-06-28</th>\n      <td>27.567127</td>\n    </tr>\n    <tr>\n      <th>2019-06-29</th>\n      <td>26.670271</td>\n    </tr>\n    <tr>\n      <th>2019-06-30</th>\n      <td>25.054070</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Bernal \n- Win in Tour de Suisse on 23 June\n- 3rd Volta Ciclista a Catalunya 31 March\n- Win Paris Nice 17 March"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "x.loc['Egan Bernal']['20190501':'20190630']",
      "execution_count": 217,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 217,
          "data": {
            "text/plain": "            Probability\nDate                   \n2019-05-02     3.409962\n2019-05-04     4.187431\n2019-05-05    11.310299\n2019-05-06     5.847953\n2019-05-08     6.385619\n2019-05-09     6.285147\n2019-05-10     6.900454\n2019-05-11     6.681217\n2019-05-13    11.098674\n2019-05-14    13.889762\n2019-05-15     6.785147\n2019-05-16     6.108690\n2019-05-21     5.681818\n2019-05-22     5.524862\n2019-05-24     5.263158\n2019-05-26     5.000000\n2019-05-27     5.263158\n2019-05-28     4.545455\n2019-05-29     5.263158\n2019-05-31     5.131579\n2019-06-02    23.853714\n2019-06-03     7.778417\n2019-06-04     7.255902\n2019-06-06     6.874882\n2019-06-07     6.433008\n2019-06-08     5.988024\n2019-06-10     5.394010\n2019-06-11     5.256113\n2019-06-12    15.945886\n2019-06-13    15.500936\n2019-06-14    13.012420\n2019-06-15    14.221056\n2019-06-16    13.768230\n2019-06-17    11.994521\n2019-06-18    23.164923\n2019-06-19    17.527134\n2019-06-20    23.418266\n2019-06-21    21.309115\n2019-06-22    28.341976\n2019-06-23    26.674899\n2019-06-24    25.780988\n2019-06-25    23.524021\n2019-06-26    22.974396\n2019-06-27    23.183786\n2019-06-28    23.893892\n2019-06-29    25.476340\n2019-06-30    27.173861",
            "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>Probability</th>\n    </tr>\n    <tr>\n      <th>Date</th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>2019-05-02</th>\n      <td>3.409962</td>\n    </tr>\n    <tr>\n      <th>2019-05-04</th>\n      <td>4.187431</td>\n    </tr>\n    <tr>\n      <th>2019-05-05</th>\n      <td>11.310299</td>\n    </tr>\n    <tr>\n      <th>2019-05-06</th>\n      <td>5.847953</td>\n    </tr>\n    <tr>\n      <th>2019-05-08</th>\n      <td>6.385619</td>\n    </tr>\n    <tr>\n      <th>2019-05-09</th>\n      <td>6.285147</td>\n    </tr>\n    <tr>\n      <th>2019-05-10</th>\n      <td>6.900454</td>\n    </tr>\n    <tr>\n      <th>2019-05-11</th>\n      <td>6.681217</td>\n    </tr>\n    <tr>\n      <th>2019-05-13</th>\n      <td>11.098674</td>\n    </tr>\n    <tr>\n      <th>2019-05-14</th>\n      <td>13.889762</td>\n    </tr>\n    <tr>\n      <th>2019-05-15</th>\n      <td>6.785147</td>\n    </tr>\n    <tr>\n      <th>2019-05-16</th>\n      <td>6.108690</td>\n    </tr>\n    <tr>\n      <th>2019-05-21</th>\n      <td>5.681818</td>\n    </tr>\n    <tr>\n      <th>2019-05-22</th>\n      <td>5.524862</td>\n    </tr>\n    <tr>\n      <th>2019-05-24</th>\n      <td>5.263158</td>\n    </tr>\n    <tr>\n      <th>2019-05-26</th>\n      <td>5.000000</td>\n    </tr>\n    <tr>\n      <th>2019-05-27</th>\n      <td>5.263158</td>\n    </tr>\n    <tr>\n      <th>2019-05-28</th>\n      <td>4.545455</td>\n    </tr>\n    <tr>\n      <th>2019-05-29</th>\n      <td>5.263158</td>\n    </tr>\n    <tr>\n      <th>2019-05-31</th>\n      <td>5.131579</td>\n    </tr>\n    <tr>\n      <th>2019-06-02</th>\n      <td>23.853714</td>\n    </tr>\n    <tr>\n      <th>2019-06-03</th>\n      <td>7.778417</td>\n    </tr>\n    <tr>\n      <th>2019-06-04</th>\n      <td>7.255902</td>\n    </tr>\n    <tr>\n      <th>2019-06-06</th>\n      <td>6.874882</td>\n    </tr>\n    <tr>\n      <th>2019-06-07</th>\n      <td>6.433008</td>\n    </tr>\n    <tr>\n      <th>2019-06-08</th>\n      <td>5.988024</td>\n    </tr>\n    <tr>\n      <th>2019-06-10</th>\n      <td>5.394010</td>\n    </tr>\n    <tr>\n      <th>2019-06-11</th>\n      <td>5.256113</td>\n    </tr>\n    <tr>\n      <th>2019-06-12</th>\n      <td>15.945886</td>\n    </tr>\n    <tr>\n      <th>2019-06-13</th>\n      <td>15.500936</td>\n    </tr>\n    <tr>\n      <th>2019-06-14</th>\n      <td>13.012420</td>\n    </tr>\n    <tr>\n      <th>2019-06-15</th>\n      <td>14.221056</td>\n    </tr>\n    <tr>\n      <th>2019-06-16</th>\n      <td>13.768230</td>\n    </tr>\n    <tr>\n      <th>2019-06-17</th>\n      <td>11.994521</td>\n    </tr>\n    <tr>\n      <th>2019-06-18</th>\n      <td>23.164923</td>\n    </tr>\n    <tr>\n      <th>2019-06-19</th>\n      <td>17.527134</td>\n    </tr>\n    <tr>\n      <th>2019-06-20</th>\n      <td>23.418266</td>\n    </tr>\n    <tr>\n      <th>2019-06-21</th>\n      <td>21.309115</td>\n    </tr>\n    <tr>\n      <th>2019-06-22</th>\n      <td>28.341976</td>\n    </tr>\n    <tr>\n      <th>2019-06-23</th>\n      <td>26.674899</td>\n    </tr>\n    <tr>\n      <th>2019-06-24</th>\n      <td>25.780988</td>\n    </tr>\n    <tr>\n      <th>2019-06-25</th>\n      <td>23.524021</td>\n    </tr>\n    <tr>\n      <th>2019-06-26</th>\n      <td>22.974396</td>\n    </tr>\n    <tr>\n      <th>2019-06-27</th>\n      <td>23.183786</td>\n    </tr>\n    <tr>\n      <th>2019-06-28</th>\n      <td>23.893892</td>\n    </tr>\n    <tr>\n      <th>2019-06-29</th>\n      <td>25.476340</td>\n    </tr>\n    <tr>\n      <th>2019-06-30</th>\n      <td>27.173861</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3",
      "language": "python"
    },
    "language_info": {
      "name": "python",
      "version": "3.7.2",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "gist": {
      "id": "15eb59739ae07d95021043865df31aff",
      "data": {
        "description": "Strava/TdF Oddschecker.ipynb",
        "public": false
      }
    },
    "_draft": {
      "nbviewer_url": "https://gist.github.com/15eb59739ae07d95021043865df31aff"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 2
}