{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Fantasy Football\n", "\n", "So - this is my first year participating in a fantasy football league. I enjoy football, but I typically only keep up with a few teams, so drafting an actual team was a bit daunting. So, like most things, I relied on data to help me out. I spent some time researching strategy, looking at projections, and even simulating some drafts. Draft day came and I felt pretty good about my team...but now I am currently 0-3 for the season. Ha.\n", "\n", "I started to get a bit curious about why I was doing so poorly. Typically my \"projected\" points were pretty good each week, but my team just never seemed to deliver.\n", "\n", "Thus, here is my first post looking at the biggest out performers and the biggest misses relative to ESPN projections. All data are from ESPN. So - lets get to it!" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "collapsed": true }, "outputs": [], "source": [ "import psycopg2\n", "import pandas.io.sql as psql\n", "import pandas as pd\n", "from matplotlib import pyplot as plt\n", "from __future__ import division #now division always returns a floating point number\n", "import numpy as np\n", "import seaborn as sns\n", "%matplotlib inline" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "collapsed": false }, "outputs": [], "source": [ "db = psycopg2.connect(\"dbname='fantasyfootball' host='localhost'\")\n", "\n", "def get_combined_df():\n", " actual_points = psql.read_sql(\"\"\"\n", " select name, team, position, sum(total_points) as points_actual \n", " from scoring_leaders_weekly\n", " group by name, team, position;\"\"\", con=db)\n", " predicted_points = psql.read_sql(\"\"\"\n", " select name, team, position, sum(total_points) as points_predicted\n", " from next_week_projections\n", " group by name, team, position;\"\"\", con=db)\n", " combined_df = actual_points.merge(predicted_points, \n", " on=['name', 'team', 'position'], how='left')\n", " combined_df = combined_df.dropna()\n", " combined_df = combined_df[combined_df['points_predicted'] > 0]\n", " combined_df['points_diff'] = combined_df.points_actual - combined_df.points_predicted\n", " combined_df['points_diff_pct'] = (combined_df.points_actual - combined_df.points_predicted) / combined_df.points_predicted\n", " return combined_df\n", "\n", "def get_top_bottom(df):\n", " group = df.groupby('position')\n", " top_list = []\n", " bottom_list = []\n", " for name, data in group:\n", " top = data.sort('points_diff', ascending=False)\n", " top_list.append(top.head())\n", " tail = top.tail()\n", " tail = tail.sort('points_diff')\n", " bottom_list.append(tail)\n", " top_df = pd.concat(top_list)\n", " bottom_df = pd.concat(bottom_list)\n", " return top_df, bottom_df\n", "\n", "def run_analysis():\n", " combined_df = get_combined_df()\n", " top, bottom = get_top_bottom(combined_df)\n", " return combined_df, top, bottom" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# RESULTS\n", "\n", "For the results, I decided to show the top 5 out performers and the top 5 under performers for the cumulative season based on the absolute point difference (not the percentage). First, here are the **out performers**. This is pretty interesting. Travis Benjamin, for example, has in the first 3 weeks produced an extra 33.6 fantasy points relative to expectations. Not too bad." ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "collapsed": false, "scrolled": true }, "outputs": [], "source": [ "combined_df, top_1, bottom_1 = run_analysis()" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
nameteampositionpoints_actualpoints_predictedpoints_diffpoints_diff_pct
384BroncosBroncosD5536.518.50.506849
281CardinalsCardinalsD4831.416.60.528662
490CowboysCowboysD2315.97.10.446541
248TitansTitansD2316.16.90.428571
25JetsJetsD3430.73.30.107492
255Stephen GostkowskiNEK4026.213.80.526718
309Josh BrownNYGK3932.46.60.203704
247Brandon McManusDenK3430.04.00.133333
307Steven HauschkaSeaK3430.73.30.107492
496Mason CrosbyGBK3232.1-0.1-0.003115
375Tom BradyNEQB7752.224.80.475096
204Andy DaltonCinQB6951.517.50.339806
13Marcus MariotaTenQB5745.411.60.255507
488Tyrod TaylorBufQB6453.110.90.205273
107Ryan MallettHouQB3728.68.40.293706
265Karlos WilliamsBufRB3716.920.11.189349
295Marcel ReeceOakRB203.316.75.060606
457Dion LewisNERB4024.815.20.612903
99DeAngelo WilliamsPitRB3824.213.80.570248
134Devonta FreemanAtlRB5137.213.80.370968
171Rob GronkowskiNETE5434.020.00.588235
241Anthony FasanoTenTE170.316.755.666667
219Austin Seferian-JenkinsTBTE239.014.01.555556
319Gary BarnidgeCleTE206.713.31.985075
272Travis KelceKCTE3727.99.10.326165
258Travis BenjaminCleWR5117.433.61.931034
414Larry FitzgeraldAriWR6234.127.90.818182
174Rishard MatthewsMiaWR4315.927.11.704403
250James JonesGBWR4426.217.80.679389
394Julian EdelmanNEWR3923.815.20.638655
\n", "
" ], "text/plain": [ " name team position points_actual \\\n", "384 Broncos Broncos D 55 \n", "281 Cardinals Cardinals D 48 \n", "490 Cowboys Cowboys D 23 \n", "248 Titans Titans D 23 \n", "25 Jets Jets D 34 \n", "255 Stephen Gostkowski NE K 40 \n", "309 Josh Brown NYG K 39 \n", "247 Brandon McManus Den K 34 \n", "307 Steven Hauschka Sea K 34 \n", "496 Mason Crosby GB K 32 \n", "375 Tom Brady NE QB 77 \n", "204 Andy Dalton Cin QB 69 \n", "13 Marcus Mariota Ten QB 57 \n", "488 Tyrod Taylor Buf QB 64 \n", "107 Ryan Mallett Hou QB 37 \n", "265 Karlos Williams Buf RB 37 \n", "295 Marcel Reece Oak RB 20 \n", "457 Dion Lewis NE RB 40 \n", "99 DeAngelo Williams Pit RB 38 \n", "134 Devonta Freeman Atl RB 51 \n", "171 Rob Gronkowski NE TE 54 \n", "241 Anthony Fasano Ten TE 17 \n", "219 Austin Seferian-Jenkins TB TE 23 \n", "319 Gary Barnidge Cle TE 20 \n", "272 Travis Kelce KC TE 37 \n", "258 Travis Benjamin Cle WR 51 \n", "414 Larry Fitzgerald Ari WR 62 \n", "174 Rishard Matthews Mia WR 43 \n", "250 James Jones GB WR 44 \n", "394 Julian Edelman NE WR 39 \n", "\n", " points_predicted points_diff points_diff_pct \n", "384 36.5 18.5 0.506849 \n", "281 31.4 16.6 0.528662 \n", "490 15.9 7.1 0.446541 \n", "248 16.1 6.9 0.428571 \n", "25 30.7 3.3 0.107492 \n", "255 26.2 13.8 0.526718 \n", "309 32.4 6.6 0.203704 \n", "247 30.0 4.0 0.133333 \n", "307 30.7 3.3 0.107492 \n", "496 32.1 -0.1 -0.003115 \n", "375 52.2 24.8 0.475096 \n", "204 51.5 17.5 0.339806 \n", "13 45.4 11.6 0.255507 \n", "488 53.1 10.9 0.205273 \n", "107 28.6 8.4 0.293706 \n", "265 16.9 20.1 1.189349 \n", "295 3.3 16.7 5.060606 \n", "457 24.8 15.2 0.612903 \n", "99 24.2 13.8 0.570248 \n", "134 37.2 13.8 0.370968 \n", "171 34.0 20.0 0.588235 \n", "241 0.3 16.7 55.666667 \n", "219 9.0 14.0 1.555556 \n", "319 6.7 13.3 1.985075 \n", "272 27.9 9.1 0.326165 \n", "258 17.4 33.6 1.931034 \n", "414 34.1 27.9 0.818182 \n", "174 15.9 27.1 1.704403 \n", "250 26.2 17.8 0.679389 \n", "394 23.8 15.2 0.638655 " ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "top_1" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now here are the **under performers**. These are the people you don't want to be playing...For example, C.J. Anderson comes in at a whopping 42 points under expectation. Who is my running back you ask... Drew Brees takes the cake, though, under performing by 46.2 points on the season. He was injured, though." ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
nameteampositionpoints_actualpoints_predictedpoints_diffpoints_diff_pct
45ColtsColtsD1234.4-22.4-0.651163
223DolphinsDolphinsD1436.3-22.3-0.614325
213TexansTexansD1232.1-20.1-0.626168
470LionsLionsD1427.4-13.4-0.489051
141ChargersChargersD1023.1-13.1-0.567100
366Adam VinatieriIndK531.5-26.5-0.841270
24Matt PraterDetK830.9-22.9-0.741100
461Phil DawsonSFK1335.8-22.8-0.636872
399Andrew FranksMiaK1433.6-19.6-0.583333
193Josh ScobeePitK1632.1-16.1-0.501558
21Drew BreesNOQB2874.2-46.2-0.622642
251Andrew LuckIndQB4169.5-28.5-0.410072
426Sam BradfordPhiQB2755.5-28.5-0.513514
477Teddy BridgewaterMinQB2855.5-27.5-0.495495
428Peyton ManningDenQB4367.7-24.7-0.364845
20C.J. AndersonDenRB648.0-42.0-0.875000
276Marshawn LynchSeaRB1955.0-36.0-0.654545
184DeMarco MurrayPhiRB1850.3-32.3-0.642147
311Jeremy HillCinRB2050.0-30.0-0.600000
416Lamar MillerMiaRB1542.8-27.8-0.649533
16Zach ErtzPhiTE822.0-14.0-0.636364
386Benjamin WatsonNOTE415.4-11.4-0.740260
226Mychal RiveraOakTE112.0-11.0-0.916667
136Jeff CumberlandNYJTE110.8-9.8-0.907407
227Martellus BennettChiTE1625.0-9.0-0.360000
296Alshon JefferyChiWR737.4-30.4-0.812834
123Calvin JohnsonDetWR2448.7-24.7-0.507187
32Andre JohnsonIndWR427.9-23.9-0.856631
87Demaryius ThomasDenWR3052.5-22.5-0.428571
105Mike EvansTBWR1032.4-22.4-0.691358
\n", "
" ], "text/plain": [ " name team position points_actual points_predicted \\\n", "45 Colts Colts D 12 34.4 \n", "223 Dolphins Dolphins D 14 36.3 \n", "213 Texans Texans D 12 32.1 \n", "470 Lions Lions D 14 27.4 \n", "141 Chargers Chargers D 10 23.1 \n", "366 Adam Vinatieri Ind K 5 31.5 \n", "24 Matt Prater Det K 8 30.9 \n", "461 Phil Dawson SF K 13 35.8 \n", "399 Andrew Franks Mia K 14 33.6 \n", "193 Josh Scobee Pit K 16 32.1 \n", "21 Drew Brees NO QB 28 74.2 \n", "251 Andrew Luck Ind QB 41 69.5 \n", "426 Sam Bradford Phi QB 27 55.5 \n", "477 Teddy Bridgewater Min QB 28 55.5 \n", "428 Peyton Manning Den QB 43 67.7 \n", "20 C.J. Anderson Den RB 6 48.0 \n", "276 Marshawn Lynch Sea RB 19 55.0 \n", "184 DeMarco Murray Phi RB 18 50.3 \n", "311 Jeremy Hill Cin RB 20 50.0 \n", "416 Lamar Miller Mia RB 15 42.8 \n", "16 Zach Ertz Phi TE 8 22.0 \n", "386 Benjamin Watson NO TE 4 15.4 \n", "226 Mychal Rivera Oak TE 1 12.0 \n", "136 Jeff Cumberland NYJ TE 1 10.8 \n", "227 Martellus Bennett Chi TE 16 25.0 \n", "296 Alshon Jeffery Chi WR 7 37.4 \n", "123 Calvin Johnson Det WR 24 48.7 \n", "32 Andre Johnson Ind WR 4 27.9 \n", "87 Demaryius Thomas Den WR 30 52.5 \n", "105 Mike Evans TB WR 10 32.4 \n", "\n", " points_diff points_diff_pct \n", "45 -22.4 -0.651163 \n", "223 -22.3 -0.614325 \n", "213 -20.1 -0.626168 \n", "470 -13.4 -0.489051 \n", "141 -13.1 -0.567100 \n", "366 -26.5 -0.841270 \n", "24 -22.9 -0.741100 \n", "461 -22.8 -0.636872 \n", "399 -19.6 -0.583333 \n", "193 -16.1 -0.501558 \n", "21 -46.2 -0.622642 \n", "251 -28.5 -0.410072 \n", "426 -28.5 -0.513514 \n", "477 -27.5 -0.495495 \n", "428 -24.7 -0.364845 \n", "20 -42.0 -0.875000 \n", "276 -36.0 -0.654545 \n", "184 -32.3 -0.642147 \n", "311 -30.0 -0.600000 \n", "416 -27.8 -0.649533 \n", "16 -14.0 -0.636364 \n", "386 -11.4 -0.740260 \n", "226 -11.0 -0.916667 \n", "136 -9.8 -0.907407 \n", "227 -9.0 -0.360000 \n", "296 -30.4 -0.812834 \n", "123 -24.7 -0.507187 \n", "32 -23.9 -0.856631 \n", "87 -22.5 -0.428571 \n", "105 -22.4 -0.691358 " ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "bottom_1" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Next, I wanted to take a look at the distribution of point differences by position. The below chart shows that the median player in all positions is under performing, except for TE which is pretty close to zero. There are a few break out WRs and quite a bunch of under performing running backs. The spread is also pretty wide for most of the positions." ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEZCAYAAAB1mUk3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/g94AwZYP8A9sgAAAABJRU5ErkJggg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "ax = sns.boxplot(combined_df.points_diff, groupby=combined_df.position)\n", "plt.title(\"Distribution of Point Differences by Position\")\n", "sns.despine()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "I also looked at the distribution of actual points by position. One thing you hear in fantasy is to select RBs early because they are high variance players. Meaning that you suffer more by getting a lower ranked RB than a lower ranked QB. This is also due to the fact that a lot more RBs are getting drafted than QBs. Below is the distribution for all players and provides a general sense." ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEZCAYAAAB7HPUdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEckRJX0RkRxR0hcRyRElfRGRHPk/ZUuygyS44QMAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "ax = sns.boxplot(combined_df.points_actual, groupby=combined_df.position)\n", "plt.title(\"Distribution of Actual Points by Position\")\n", "sns.despine()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To see if the high variance difference is playing out, we can look at the top 12 quarterbacks and top 36 running backs so far in the season (assume 12 team league with 1 starting QB and 3 starting RBs). You can see below that indeed the RBs standard deviation is quite a bit higher (about 7 points) than the QBs." ] }, { "cell_type": "code", "execution_count": 34, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
points_actualpoints_predictedpoints_diffpoints_diff_pct
count12.00000012.00000012.00000012.000000
mean60.50000058.4250002.0750000.052784
std10.7068378.42476112.7530480.227619
min48.00000045.400000-17.800000-0.258721
25%50.50000051.475000-5.175000-0.089687
50%60.50000057.5000000.4000000.006114
75%67.50000065.75000011.0750000.217831
max77.00000069.90000024.8000000.475096
\n", "
" ], "text/plain": [ " points_actual points_predicted points_diff points_diff_pct\n", "count 12.000000 12.000000 12.000000 12.000000\n", "mean 60.500000 58.425000 2.075000 0.052784\n", "std 10.706837 8.424761 12.753048 0.227619\n", "min 48.000000 45.400000 -17.800000 -0.258721\n", "25% 50.500000 51.475000 -5.175000 -0.089687\n", "50% 60.500000 57.500000 0.400000 0.006114\n", "75% 67.500000 65.750000 11.075000 0.217831\n", "max 77.000000 69.900000 24.800000 0.475096" ] }, "execution_count": 34, "metadata": {}, "output_type": "execute_result" } ], "source": [ "combined_df[combined_df['position'] == \"QB\"].sort('points_actual', ascending=False).head(n=12).describe()" ] }, { "cell_type": "code", "execution_count": 35, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
points_actualpoints_predictedpoints_diffpoints_diff_pct
count36.00000036.0000003.600000e+0136.000000
mean41.08333350.894444-9.811111e+00-0.186386
std17.76252115.9405491.436052e+010.272579
min9.00000014.600000-4.620000e+01-0.625000
25%28.00000044.325000-1.732500e+01-0.366491
50%41.50000054.750000-1.295000e+01-0.259642
75%49.50000062.650000-1.776357e-15-0.002554
max77.00000074.2000002.480000e+010.475096
\n", "
" ], "text/plain": [ " points_actual points_predicted points_diff points_diff_pct\n", "count 36.000000 36.000000 3.600000e+01 36.000000\n", "mean 41.083333 50.894444 -9.811111e+00 -0.186386\n", "std 17.762521 15.940549 1.436052e+01 0.272579\n", "min 9.000000 14.600000 -4.620000e+01 -0.625000\n", "25% 28.000000 44.325000 -1.732500e+01 -0.366491\n", "50% 41.500000 54.750000 -1.295000e+01 -0.259642\n", "75% 49.500000 62.650000 -1.776357e-15 -0.002554\n", "max 77.000000 74.200000 2.480000e+01 0.475096" ] }, "execution_count": 35, "metadata": {}, "output_type": "execute_result" } ], "source": [ "combined_df[combined_df['position'] == \"QB\"].sort('points_actual', ascending=False).head(n=36).describe()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# In conclusion\n", "\n", "These were just some quick analyses I did to try and get a sense of which players are doing well/poorly and how various positions are performing. \n", "\n", "If people find this interesting, I can try and update the data as the season goes on.\n", "\n", "I am hoping to find time to investigate the ESPN projections to see how sensical they really are. Based on the chart above, they seem to aim high, leading to many under performers. I would like to try and build my own projection model to see how well I can compare. Now that I think about it, here are the overall summary statistics below. It looks like on average ESPN is over projecting by about 4 points with a standard deviation of 10.5 points." ] }, { "cell_type": "code", "execution_count": 38, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
points_actualpoints_predictedpoints_diffpoints_diff_pct
count408.000000408.000000408.000000408.000000
mean17.42402021.504167-4.0801471.497506
std15.39964316.29556410.51460510.662482
min0.0000000.100000-46.200000-1.000000
25%5.0000007.875000-10.125000-0.500389
50%14.00000019.550000-3.550000-0.250522
75%25.00000030.7250001.5000000.180199
max77.00000074.20000033.600000119.000000
\n", "
" ], "text/plain": [ " points_actual points_predicted points_diff points_diff_pct\n", "count 408.000000 408.000000 408.000000 408.000000\n", "mean 17.424020 21.504167 -4.080147 1.497506\n", "std 15.399643 16.295564 10.514605 10.662482\n", "min 0.000000 0.100000 -46.200000 -1.000000\n", "25% 5.000000 7.875000 -10.125000 -0.500389\n", "50% 14.000000 19.550000 -3.550000 -0.250522\n", "75% 25.000000 30.725000 1.500000 0.180199\n", "max 77.000000 74.200000 33.600000 119.000000" ] }, "execution_count": 38, "metadata": {}, "output_type": "execute_result" } ], "source": [ "combined_df.describe()" ] } ], "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": "2.7.10" } }, "nbformat": 4, "nbformat_minor": 0 }