{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Positional Spending Analysis\n", "**Name:** Jaime Avendaño \n", "**Date:** 5/21/2021 \n", "
\n", "This notebook takes the positional spending data and does feature engineering by normalizing the spending. All values are compared to the NFL Salary Cap for the given year. \n", "
\n", "The salary cap for each team can vary slightly, since there is an amount that can be rolled over. Also, these numbers don't count the dead money from old contracts or released players. But by comparing it to a single number for a year, we can get a relative percentage that can be compared across teams and years." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "from matplotlib.offsetbox import OffsetImage, AnnotationBbox\n", "import matplotlib.ticker as mtick\n", "import seaborn as sns\n", "\n", "import plotly.graph_objects as go\n", "from plotly.colors import n_colors\n", "\n", "import janitor\n", "\n", "from ipywidgets import interact, interactive, fixed, interact_manual\n", "import ipywidgets as widgets" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "((288, 15), (256, 5), (32, 4))" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df = pd.read_parquet('teams_spending_df.parquet')\n", "details_df = pd.read_parquet('teams_detail_df.parquet')\n", "logos_df = pd.read_parquet('teams_logos_df.parquet')\n", "df.shape, details_df.shape, logos_df.shape" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "teams = df.team.unique().astype('str')\n", "teams.sort()" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Index(['qb', 'rb', 'wr', 'te', 'ol', 'offense', 'idl', 'edge', 'lb', 's', 'cb',\n", " 'defense'],\n", " dtype='object')" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "positions = df.columns[1:-2]\n", "positions" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['qb_pct',\n", " 'rb_pct',\n", " 'wr_pct',\n", " 'te_pct',\n", " 'ol_pct',\n", " 'idl_pct',\n", " 'edge_pct',\n", " 'lb_pct',\n", " 's_pct',\n", " 'cb_pct']" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "positions_pct = [f'{pos}_pct' for pos in positions if pos not in ['defense', 'offense']]\n", "positions_pct" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "def getImage(path): \n", " return OffsetImage(plt.imread(path), zoom=0.35)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Feature Engineering\n", "Preparing the totals and percentate of cap columns." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "df['total'] = df['offense'] + df['defense']" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "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", "
teamqbrbwrteoloffenseidledgelb...wr_pctte_pctol_pctoffense_pctidl_pctedge_pctlb_pcts_pctcb_pctdefense_pct
0Eagles1338513710203112192419895509036256648997400417352646661024110110004817...0.1564390.0447890.2086580.6016600.0428020.0832610.0813400.0562610.0490470.312711
1Seahawks155708510799653168314231277878827955261699222107701509250138327377232...0.1368410.1038930.2272790.5684730.0626140.2033640.0599770.0697530.0285190.424228
2Titans6336958153760981268689669795002672198468101436649652887762935740835...0.1031450.0567440.2172520.5536700.0528170.0713520.0466730.0876180.0780690.336530
3Broncos18716295507063210120554801390224318758662401416880138531355412951882...0.0822810.0651540.1977130.5385380.0559360.0432000.1053000.0479220.1427580.395115
4Giants219984005036739867762629989132423590062947578511899595238137916847...0.0705500.0243810.1970400.5117690.0416180.0774290.0643650.1032930.0910740.377778
\n", "

5 rows × 28 columns

\n", "
" ], "text/plain": [ " team qb rb wr te ol offense \\\n", "0 Eagles 13385137 10203112 19241989 5509036 25664899 74004173 \n", "1 Seahawks 1557085 10799653 16831423 12778788 27955261 69922210 \n", "2 Titans 6336958 15376098 12686896 6979500 26721984 68101436 \n", "3 Broncos 18716295 5070632 10120554 8013902 24318758 66240141 \n", "4 Giants 21998400 5036739 8677626 2998913 24235900 62947578 \n", "\n", " idl edge lb ... wr_pct te_pct ol_pct \\\n", "0 5264666 10241101 10004817 ... 0.156439 0.044789 0.208658 \n", "1 7701509 25013832 7377232 ... 0.136841 0.103893 0.227279 \n", "2 6496528 8776293 5740835 ... 0.103145 0.056744 0.217252 \n", "3 6880138 5313554 12951882 ... 0.082281 0.065154 0.197713 \n", "4 5118995 9523813 7916847 ... 0.070550 0.024381 0.197040 \n", "\n", " offense_pct idl_pct edge_pct lb_pct s_pct cb_pct defense_pct \n", "0 0.601660 0.042802 0.083261 0.081340 0.056261 0.049047 0.312711 \n", "1 0.568473 0.062614 0.203364 0.059977 0.069753 0.028519 0.424228 \n", "2 0.553670 0.052817 0.071352 0.046673 0.087618 0.078069 0.336530 \n", "3 0.538538 0.055936 0.043200 0.105300 0.047922 0.142758 0.395115 \n", "4 0.511769 0.041618 0.077429 0.064365 0.103293 0.091074 0.377778 \n", "\n", "[5 rows x 28 columns]" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "for pos in positions:\n", " df[f'{pos}_pct'] = df[pos] / df.cap\n", "df.head()" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "C:\\Users\\jaime.avendano\\Anaconda3\\lib\\site-packages\\traitlets\\traitlets.py:586: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n", " silent = bool(old_value == new_value)\n" ] }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "285efc06d5be4e658af358a71efe5554", "version_major": 2, "version_minor": 0 }, "text/plain": [ "interactive(children=(Dropdown(description='Select team: ', options=('49ers', 'Bears', 'Bengals', 'Bills', 'Br…" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "@interact(team=widgets.Dropdown(options=teams, description='Select team: '))\n", "def f(team):\n", " df_long = df.loc[df.team == team, ['team', 'year','offense_pct','defense_pct']]\n", " df_long = df_long.melt(id_vars=['team', 'year'], value_vars=['offense_pct', 'defense_pct'],\n", " var_name='position', value_name='cap_pct')\n", " plt.figure(figsize=(15, 10))\n", " sns.lineplot(data=df_long, x='year', y='cap_pct', hue='position')\n", " return None;" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "def get_single_year_data(df, year):\n", " df_year = df[df.year == year][['team', 'offense_pct', 'defense_pct']]\n", " df_year = df_year.join(logos_df.set_index('team'), on='team')\n", " return df_year" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "688d15eb613d41c4835d95cfa8da91e3", "version_major": 2, "version_minor": 0 }, "text/plain": [ "interactive(children=(Dropdown(description='Select year: ', options=(2013, 2014, 2015, 2016, 2017, 2018, 2019,…" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "@interact(plot_year=widgets.Dropdown(options=range(2013, 2021), description='Select year: '))\n", "def f(plot_year):\n", " plot_df = get_single_year_data(df, plot_year)\n", "\n", " plt.figure(figsize=(12, 12))\n", " sns.set_style('darkgrid')\n", " ax = sns.scatterplot(data=df, x='offense_pct', y='defense_pct', s=4)\n", " ax.xaxis.set_major_formatter(mtick.PercentFormatter(1.0))\n", " ax.yaxis.set_major_formatter(mtick.PercentFormatter(1.0))\n", " plt.xlim((0.2, 0.6))\n", " plt.ylim((0.2, 0.6))\n", "\n", "\n", " for x0, y0, path in zip(plot_df.offense_pct, plot_df.defense_pct, plot_df.path):\n", " ab = AnnotationBbox(getImage(path), (x0, y0), frameon=False, fontsize=4)\n", " ax.add_artist(ab)\n", "\n", " plt.suptitle(f'NFL - Cap Spending ({ plot_year } highlighted)', fontsize=20, y=0.93)\n", " plt.title('Does not include dead money for the year.', fontsize=15)\n", " plt.xlabel('% of Cap spent on Offense', fontsize=18)\n", " plt.ylabel('% of Cap spent on Defense', fontsize=18)\n", " plt.show()\n", " return None;" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "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", "
offense_pctdefense_pct
count32.00000032.000000
mean0.4484160.418107
std0.0327130.035482
min0.3780940.343014
25%0.4342560.392952
50%0.4492470.427205
75%0.4685990.443904
max0.5087750.479979
\n", "
" ], "text/plain": [ " offense_pct defense_pct\n", "count 32.000000 32.000000\n", "mean 0.448416 0.418107\n", "std 0.032713 0.035482\n", "min 0.378094 0.343014\n", "25% 0.434256 0.392952\n", "50% 0.449247 0.427205\n", "75% 0.468599 0.443904\n", "max 0.508775 0.479979" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "team_avg_spending = df[['team', 'offense_pct', 'defense_pct']].groupby('team').mean().reset_index()\n", "team_avg_spending.describe()" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "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", "
teamoffense_pctdefense_pct
24Ravens0.3780940.441465
3Bills0.3880250.450515
12Dolphins0.3905190.453997
16Jaguars0.4025570.440993
4Broncos0.4081360.478347
\n", "
" ], "text/plain": [ " team offense_pct defense_pct\n", "24 Ravens 0.378094 0.441465\n", "3 Bills 0.388025 0.450515\n", "12 Dolphins 0.390519 0.453997\n", "16 Jaguars 0.402557 0.440993\n", "4 Broncos 0.408136 0.478347" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "team_avg_spending.sort_values('offense_pct').head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## KMeans Analysis" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "from sklearn.cluster import KMeans\n", "from sklearn.metrics import silhouette_score, davies_bouldin_score,v_measure_score" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['qb_pct',\n", " 'rb_pct',\n", " 'wr_pct',\n", " 'te_pct',\n", " 'ol_pct',\n", " 'idl_pct',\n", " 'edge_pct',\n", " 'lb_pct',\n", " 's_pct',\n", " 'cb_pct']" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "positions_pct" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "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", "
qb_pctrb_pctwr_pctte_pctol_pctidl_pctedge_pctlb_pcts_pctcb_pct
00.1088220.0829520.1564390.0447890.2086580.0428020.0832610.0813400.0562610.049047
10.0126590.0878020.1368410.1038930.2272790.0626140.2033640.0599770.0697530.028519
20.0515200.1250090.1031450.0567440.2172520.0528170.0713520.0466730.0876180.078069
30.1521650.0412250.0822810.0651540.1977130.0559360.0432000.1053000.0479220.142758
40.1788490.0409490.0705500.0243810.1970400.0416180.0774290.0643650.1032930.091074
\n", "
" ], "text/plain": [ " qb_pct rb_pct wr_pct te_pct ol_pct idl_pct edge_pct \\\n", "0 0.108822 0.082952 0.156439 0.044789 0.208658 0.042802 0.083261 \n", "1 0.012659 0.087802 0.136841 0.103893 0.227279 0.062614 0.203364 \n", "2 0.051520 0.125009 0.103145 0.056744 0.217252 0.052817 0.071352 \n", "3 0.152165 0.041225 0.082281 0.065154 0.197713 0.055936 0.043200 \n", "4 0.178849 0.040949 0.070550 0.024381 0.197040 0.041618 0.077429 \n", "\n", " lb_pct s_pct cb_pct \n", "0 0.081340 0.056261 0.049047 \n", "1 0.059977 0.069753 0.028519 \n", "2 0.046673 0.087618 0.078069 \n", "3 0.105300 0.047922 0.142758 \n", "4 0.064365 0.103293 0.091074 " ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "#kmeans_data = df.join(details_df.set_index(['year', 'team']), on=['year', 'team'], how='inner')[positions_pct + ['win_pct']].copy()\n", "#kmeans_data = df.join(details_df.set_index(['year', 'team']), on=['year', 'team'], how='inner')[['offense_pct', 'defense_pct']].copy()\n", "kmeans_data = df.join(details_df.set_index(['year', 'team']), on=['year', 'team'], how='inner')[positions_pct].copy()\n", "kmeans_data.head()" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [], "source": [ "kmeans_data['skill_pos_pct'] = kmeans_data['rb_pct'] + kmeans_data['wr_pct'] + kmeans_data['te_pct']\n", "kmeans_data['def_front_pct'] = kmeans_data['idl_pct'] + kmeans_data['edge_pct'] + kmeans_data['lb_pct']\n", "kmeans_data['def_backs_pct'] = kmeans_data['s_pct'] + kmeans_data['cb_pct']\n", "kmeans_data = kmeans_data.drop(columns=['rb_pct', 'wr_pct', 'te_pct', 'idl_pct', 'edge_pct', 'lb_pct', 's_pct', 'cb_pct'])" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "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", "
qb_pctol_pctskill_pos_pctdef_front_pctdef_backs_pct
count256.000000256.000000256.000000256.000000256.000000
mean0.0946150.1588920.1843580.2593300.149787
std0.0483100.0412570.0452280.0596170.049788
min0.0076660.0406230.0767440.0894530.050437
25%0.0531080.1306710.1568960.2155150.111904
50%0.0959760.1544960.1847880.2595320.151131
75%0.1333720.1901640.2120910.3058860.182531
max0.2376670.2812360.3285350.4049490.332771
\n", "
" ], "text/plain": [ " qb_pct ol_pct skill_pos_pct def_front_pct def_backs_pct\n", "count 256.000000 256.000000 256.000000 256.000000 256.000000\n", "mean 0.094615 0.158892 0.184358 0.259330 0.149787\n", "std 0.048310 0.041257 0.045228 0.059617 0.049788\n", "min 0.007666 0.040623 0.076744 0.089453 0.050437\n", "25% 0.053108 0.130671 0.156896 0.215515 0.111904\n", "50% 0.095976 0.154496 0.184788 0.259532 0.151131\n", "75% 0.133372 0.190164 0.212091 0.305886 0.182531\n", "max 0.237667 0.281236 0.328535 0.404949 0.332771" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "kmeans_data.describe()" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [], "source": [ "kmeans_kwargs = {\n", " 'init': 'k-means++',\n", " 'n_init': 10,\n", " 'max_iter': 100,\n", " 'random_state': 42\n", "}" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [], "source": [ "km_scores= []\n", "km_silhouette = []\n", "db_score = []\n", "\n", "for k in range(2, 11):\n", " km = KMeans(n_clusters=k, **kmeans_kwargs)\n", " km.fit(kmeans_data)\n", " preds = km.predict(kmeans_data)\n", " \n", " km_scores.append(-km.score(kmeans_data))\n", " km_silhouette.append(silhouette_score(kmeans_data, preds))\n", " db_score.append(davies_bouldin_score(kmeans_data, preds))" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, axs = plt.subplots(3, sharex=True)\n", "fig.set_figheight(10)\n", "fig.set_figwidth(15)\n", "\n", "axs[0].scatter(x=range(2, 11), y=km_scores)\n", "axs[0].set_title('The elbow method with Score')\n", "axs[0].set(ylabel='WCSS')\n", "\n", "axs[1].scatter(x=range(2, 11), y=km_silhouette)\n", "axs[1].set_title('KMeans with Silhouette Score')\n", "axs[1].set(ylabel='Silhouette Score')\n", "\n", "axs[2].scatter(x=range(2, 11), y=db_score)\n", "axs[2].set_title('KMeans with Davies-Bouldin Score')\n", "axs[2].set(ylabel='DB Score')\n", "\n", "plt.show();" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [], "source": [ "km = KMeans(n_clusters=6, **kmeans_kwargs)\n", "nfl_pred = km.fit_predict(kmeans_data)" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [], "source": [ "df_grouped = df[df.year <= 2020].copy()\n", "df_grouped['group'] = nfl_pred" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "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", "
groupteams
02Packers, Chargers, Giants
14Buccaneers, Rams, Bills
30Washington, Raiders, Titans, Lions, Patriots, ...
41Bengals, Browns, Cowboys, Colts, Chiefs, Eagle...
7549ers, Saints, Falcons, Seahawks, Steelers, Pa...
133Vikings, Texans, Bears
\n", "
" ], "text/plain": [ " group teams\n", "0 2 Packers, Chargers, Giants\n", "1 4 Buccaneers, Rams, Bills\n", "3 0 Washington, Raiders, Titans, Lions, Patriots, ...\n", "4 1 Bengals, Browns, Cowboys, Colts, Chiefs, Eagle...\n", "7 5 49ers, Saints, Falcons, Seahawks, Steelers, Pa...\n", "13 3 Vikings, Texans, Bears" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_grouped['teams'] = df_grouped[df_grouped.year == 2019][['group', 'team']].reset_index().groupby(['group'])['team'].transform(lambda x: ', '.join(x))\n", "group_2019_df = df_grouped[df_grouped.year == 2019][['group', 'teams']].drop_duplicates()\n", "group_2019_df\n" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "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", "
qb_pctol_pctskill_pos_pctdef_front_pctdef_backs_pct
00.1195130.1509840.1582270.1865860.196191
10.0443670.1917410.2017020.2324520.155091
20.1351940.1614790.2007060.2224260.116409
30.0714220.1301030.1892870.3019820.183293
40.0687240.1654480.2173530.3262460.097044
50.1194170.1720830.1272780.2994090.134926
\n", "
" ], "text/plain": [ " qb_pct ol_pct skill_pos_pct def_front_pct def_backs_pct\n", "0 0.119513 0.150984 0.158227 0.186586 0.196191\n", "1 0.044367 0.191741 0.201702 0.232452 0.155091\n", "2 0.135194 0.161479 0.200706 0.222426 0.116409\n", "3 0.071422 0.130103 0.189287 0.301982 0.183293\n", "4 0.068724 0.165448 0.217353 0.326246 0.097044\n", "5 0.119417 0.172083 0.127278 0.299409 0.134926" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "km_centers_df = pd.DataFrame(km.cluster_centers_, columns=['qb_pct', 'ol_pct', 'skill_pos_pct', 'def_front_pct', 'def_backs_pct'])\n", "km_centers_df" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "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", "
qb_pctol_pctskill_pos_pctdef_front_pctdef_backs_pctteamsqb_rankol_rankskill_pos_rankdef_front_rankdef_backs_rank
00.1195130.1509840.1582270.1865860.196191Washington, Raiders, Titans, Lions, Patriots, ...41105
10.0443670.1917410.2017020.2324520.155091Bengals, Browns, Cowboys, Colts, Chiefs, Eagle...05423
20.1351940.1614790.2007060.2224260.116409Packers, Chargers, Giants52311
30.0714220.1301030.1892870.3019820.183293Vikings, Texans, Bears20244
40.0687240.1654480.2173530.3262460.097044Buccaneers, Rams, Bills13550
50.1194170.1720830.1272780.2994090.13492649ers, Saints, Falcons, Seahawks, Steelers, Pa...34032
\n", "
" ], "text/plain": [ " qb_pct ol_pct skill_pos_pct def_front_pct def_backs_pct \\\n", "0 0.119513 0.150984 0.158227 0.186586 0.196191 \n", "1 0.044367 0.191741 0.201702 0.232452 0.155091 \n", "2 0.135194 0.161479 0.200706 0.222426 0.116409 \n", "3 0.071422 0.130103 0.189287 0.301982 0.183293 \n", "4 0.068724 0.165448 0.217353 0.326246 0.097044 \n", "5 0.119417 0.172083 0.127278 0.299409 0.134926 \n", "\n", " teams qb_rank ol_rank \\\n", "0 Washington, Raiders, Titans, Lions, Patriots, ... 4 1 \n", "1 Bengals, Browns, Cowboys, Colts, Chiefs, Eagle... 0 5 \n", "2 Packers, Chargers, Giants 5 2 \n", "3 Vikings, Texans, Bears 2 0 \n", "4 Buccaneers, Rams, Bills 1 3 \n", "5 49ers, Saints, Falcons, Seahawks, Steelers, Pa... 3 4 \n", "\n", " skill_pos_rank def_front_rank def_backs_rank \n", "0 1 0 5 \n", "1 4 2 3 \n", "2 3 1 1 \n", "3 2 4 4 \n", "4 5 5 0 \n", "5 0 3 2 " ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "km_table = km_centers_df.join(group_2019_df.set_index('group'))\n", "km_table['qb_rank'] = km_table.qb_pct.rank(method='first', ascending=True).astype('int') - 1\n", "km_table['ol_rank'] = km_table.ol_pct.rank(method='first', ascending=True).astype('int') - 1\n", "km_table['skill_pos_rank'] = km_table.skill_pos_pct.rank(method='first', ascending=True).astype('int') - 1\n", "km_table['def_front_rank'] = km_table.def_front_pct.rank(method='first', ascending=True).astype('int') - 1\n", "km_table['def_backs_rank'] = km_table.def_backs_pct.rank(method='first', ascending=True).astype('int') - 1\n", "km_table" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "data": { "text/html": [ " \n", " " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "cells": { "align": [ "left", "left", "center" ], "fill": { "color": [ [ "rgb(225.0, 225.0, 255.0)", "rgb(205.0, 205.0, 255.0)", "rgb(185.0, 185.0, 255.0)", "rgb(165.0, 165.0, 255.0)", "rgb(145.0, 145.0, 255.0)", "rgb(125.0, 125.0, 255.0)" ], [ "rgb(225.0, 225.0, 255.0)", "rgb(205.0, 205.0, 255.0)", "rgb(185.0, 185.0, 255.0)", "rgb(165.0, 165.0, 255.0)", "rgb(145.0, 145.0, 255.0)", "rgb(125.0, 125.0, 255.0)" ], [ "rgb(145.0, 145.0, 255.0)", "rgb(225.0, 225.0, 255.0)", "rgb(125.0, 125.0, 255.0)", "rgb(185.0, 185.0, 255.0)", "rgb(205.0, 205.0, 255.0)", "rgb(165.0, 165.0, 255.0)" ], [ "rgb(205.0, 205.0, 255.0)", "rgb(125.0, 125.0, 255.0)", "rgb(185.0, 185.0, 255.0)", "rgb(225.0, 225.0, 255.0)", "rgb(165.0, 165.0, 255.0)", "rgb(145.0, 145.0, 255.0)" ], [ "rgb(205.0, 205.0, 255.0)", "rgb(145.0, 145.0, 255.0)", "rgb(165.0, 165.0, 255.0)", "rgb(185.0, 185.0, 255.0)", "rgb(125.0, 125.0, 255.0)", "rgb(225.0, 225.0, 255.0)" ], [ "rgb(225.0, 225.0, 255.0)", "rgb(185.0, 185.0, 255.0)", "rgb(205.0, 205.0, 255.0)", "rgb(145.0, 145.0, 255.0)", "rgb(125.0, 125.0, 255.0)", "rgb(165.0, 165.0, 255.0)" ], [ "rgb(125.0, 125.0, 255.0)", "rgb(165.0, 165.0, 255.0)", "rgb(205.0, 205.0, 255.0)", "rgb(145.0, 145.0, 255.0)", "rgb(225.0, 225.0, 255.0)", "rgb(185.0, 185.0, 255.0)" ] ] }, "font": { "color": "black", "size": 13 }, "format": [ null, null, ".3p" ], "line": { "color": "black" }, "values": [ [ 0, 1, 2, 3, 4, 5 ], [ "Washington, Raiders, Titans, Lions, Patriots, Ravens, Dolphins", "Bengals, Browns, Cowboys, Colts, Chiefs, Eagles, Jets, Cardinals, Jaguars", "Packers, Chargers, Giants", "Vikings, Texans, Bears", "Buccaneers, Rams, Bills", "49ers, Saints, Falcons, Seahawks, Steelers, Panthers, Broncos" ], [ 0.119512558971068, 0.04436741479336779, 0.13519390941537898, 0.07142230113054876, 0.06872411503292912, 0.11941728163984384 ], [ 0.15098438766698014, 0.19174051295346534, 0.16147939202057246, 0.13010292394979675, 0.1654483737949208, 0.17208344678170717 ], [ 0.15822668871155568, 0.20170237058803203, 0.20070587467714196, 0.1892867443821866, 0.21735340491864077, 0.12727820732248582 ], [ 0.18658598366041151, 0.232452011159666, 0.22242619858875373, 0.3019820741186818, 0.3262456629011091, 0.29940853265373374 ], [ 0.19619058276005985, 0.15509086753905982, 0.11640887215306006, 0.18329310712092123, 0.09704432699988078, 0.13492571571715822 ] ] }, "columnwidth": [ 5, 30, 10, 10, 10, 10, 10 ], "header": { "align": [ "left", "left", "center" ], "fill": { "color": "white" }, "font": { "color": "black", "size": 15 }, "line": { "color": "black" }, "values": [ "Cluster", "Teams", "QB %", "OL %", "Skill Pos %", "Def Front %", "Def Backs %" ] }, "type": "table" } ], "layout": { "height": 250, "margin": { "b": 20, "l": 20, "r": 20, "t": 55 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "title": { "text": "NFL Cap Spending for 2019", "y": 0.95 } } }, "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "colors = np.array(n_colors('rgb(225, 225, 255)', 'rgb(125, 125, 255)', 6, colortype='rgb'))\n", "\n", "fig = go.Figure(data=[go.Table(\n", " header=dict(\n", " values=['Cluster', 'Teams', 'QB %', 'OL %', 'Skill Pos %', 'Def Front %', 'Def Backs %'],\n", " align=['left', 'left', 'center'],\n", " font=dict(color='black', size=15),\n", " fill_color='white',\n", " line_color='black'\n", " ),\n", " columnwidth=[5,30,10,10,10,10,10],\n", " cells=dict(\n", " values=[km_table.index.tolist(), km_table.teams, km_table.qb_pct, km_table.ol_pct, km_table.skill_pos_pct, km_table.def_front_pct, km_table.def_backs_pct],\n", " format=[None, None, '.3p'],\n", " line_color='black',\n", " fill_color=[colors[km_table.index.tolist()], \n", " colors[km_table.index.tolist()], \n", " colors[km_table.qb_rank], \n", " colors[km_table.ol_rank], \n", " colors[km_table.skill_pos_rank], \n", " colors[km_table.def_front_rank], \n", " colors[km_table.def_backs_rank]],\n", " align=['left', 'left', 'center'],\n", " font=dict(color='black', size=13)\n", " )),\n", "])\n", "fig.update_layout(title=dict(text='NFL Cap Spending for 2019', y=0.95),\n", " margin=dict(l=20, r=20, t=55, b=20),\n", " height=250)\n", "\n", "fig.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Relationship between spending and Win Percentage" ] }, { "cell_type": "code", "execution_count": 142, "metadata": {}, "outputs": [], "source": [ "from sklearn.linear_model import LinearRegression\n", "from sklearn.metrics import r2_score, mean_squared_error\n", "from sklearn.preprocessing import PolynomialFeatures\n", "from sklearn.tree import DecisionTreeRegressor" ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(288, 28)" ] }, "execution_count": 35, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.shape" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "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", "
teamyearqb_pctrb_pctwr_pctte_pctol_pctidl_pctedge_pctlb_pcts_pctcb_pctwin_pctpoints_forpoints_against
0Eagles20130.1088220.0829520.1564390.0447890.2086580.0428020.0832610.0813400.0562610.0490470.625442382
1Seahawks20130.0126590.0878020.1368410.1038930.2272790.0626140.2033640.0599770.0697530.0285190.813417231
2Titans20130.0515200.1250090.1031450.0567440.2172520.0528170.0713520.0466730.0876180.0780690.438362381
3Broncos20130.1521650.0412250.0822810.0651540.1977130.0559360.0432000.1053000.0479220.1427580.813606399
4Giants20130.1788490.0409490.0705500.0243810.1970400.0416180.0774290.0643650.1032930.0910740.438294383
\n", "
" ], "text/plain": [ " team year qb_pct rb_pct wr_pct te_pct ol_pct idl_pct \\\n", "0 Eagles 2013 0.108822 0.082952 0.156439 0.044789 0.208658 0.042802 \n", "1 Seahawks 2013 0.012659 0.087802 0.136841 0.103893 0.227279 0.062614 \n", "2 Titans 2013 0.051520 0.125009 0.103145 0.056744 0.217252 0.052817 \n", "3 Broncos 2013 0.152165 0.041225 0.082281 0.065154 0.197713 0.055936 \n", "4 Giants 2013 0.178849 0.040949 0.070550 0.024381 0.197040 0.041618 \n", "\n", " edge_pct lb_pct s_pct cb_pct win_pct points_for points_against \n", "0 0.083261 0.081340 0.056261 0.049047 0.625 442 382 \n", "1 0.203364 0.059977 0.069753 0.028519 0.813 417 231 \n", "2 0.071352 0.046673 0.087618 0.078069 0.438 362 381 \n", "3 0.043200 0.105300 0.047922 0.142758 0.813 606 399 \n", "4 0.077429 0.064365 0.103293 0.091074 0.438 294 383 " ] }, "execution_count": 36, "metadata": {}, "output_type": "execute_result" } ], "source": [ "win_stats_df = df[['team', 'year'] + positions_pct].join(details_df.set_index(['year', 'team']), on=['year', 'team'], how='inner')\n", "win_stats_df.head()" ] }, { "cell_type": "code", "execution_count": 72, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "corr = win_stats_df.drop(columns=['team', 'year']).corr()\n", "mask = np.zeros_like(corr)\n", "mask[np.triu_indices_from(mask)] = True\n", "\n", "with sns.axes_style(\"white\"):\n", " f, ax = plt.subplots(figsize=(18, 15))\n", " cmap = sns.diverging_palette(220, 10, as_cmap=True)\n", " sns.set(font_scale = 1.5)\n", " sns.heatmap(corr, mask=mask, cmap=cmap, \n", " annot=True, fmt='.2f', annot_kws={'size': 14},\n", " vmax=1, vmin=-1, center=0,\n", " square=True, linewidths=.5,\n", " cbar_kws={'shrink': 0.5})\n", " plt.title('Correlation between Positional Spending and Win Percentage.', fontsize=20)" ] }, { "cell_type": "code", "execution_count": 76, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "2020" ] }, "execution_count": 76, "metadata": {}, "output_type": "execute_result" } ], "source": [ "max_year = win_stats_df.year.max()\n", "max_year" ] }, { "cell_type": "code", "execution_count": 159, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "((224, 10), (224,), (32, 10), (32,))" ] }, "execution_count": 159, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Dividing train and test by max_year so that season doesn't creep into the data.\n", "X_train = win_stats_df.loc[win_stats_df.year != max_year, positions_pct]\n", "y_train = win_stats_df.loc[win_stats_df.year != max_year, 'win_pct']\n", "\n", "X_test = win_stats_df.loc[win_stats_df.year == max_year, positions_pct]\n", "y_test = win_stats_df.loc[win_stats_df.year == max_year, 'win_pct']\n", "\n", "X_train.shape, y_train.shape, X_test.shape, y_test.shape" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Linear Regression" ] }, { "cell_type": "code", "execution_count": 160, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "LinearRegression()" ] }, "execution_count": 160, "metadata": {}, "output_type": "execute_result" } ], "source": [ "linear_reg = LinearRegression(fit_intercept = True)\n", "linear_reg.fit(X_train,y_train)" ] }, { "cell_type": "code", "execution_count": 161, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([0.5885631 , 0.71246849, 0.45979239, 0.58286972, 0.52654104])" ] }, "execution_count": 161, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y_predict = linear_reg.predict(X_test)\n", "y_predict[:5]" ] }, { "cell_type": "code", "execution_count": 162, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Linear Regression: R2 = 0.2395\n", "Linear Regression: RMSE = 0.1857\n" ] } ], "source": [ "linear_r2 = r2_score(y_test, y_predict)\n", "linear_rmse = np.sqrt(mean_squared_error(y_test, y_predict))\n", "print(f'Linear Regression: R2 = {linear_r2:.4f}')\n", "print(f'Linear Regression: RMSE = {linear_rmse:.4f}')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Polynomial Regression\n", "Tested with degree=2 and degree=3. Both cases result in negative R2." ] }, { "cell_type": "code", "execution_count": 163, "metadata": {}, "outputs": [], "source": [ "poly_regressor = PolynomialFeatures(degree=2)" ] }, { "cell_type": "code", "execution_count": 164, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "LinearRegression()" ] }, "execution_count": 164, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X_columns = poly_regressor.fit_transform(X_train)\n", "poly_reg = LinearRegression()\n", "poly_reg.fit(X_columns, y_train)" ] }, { "cell_type": "code", "execution_count": 165, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([0.14269698, 0.77947322, 0.36924222, 0.53488198, 0.42912787])" ] }, "execution_count": 165, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y_predict = poly_reg.predict(poly_regressor.transform(X_test))\n", "y_predict[:5]" ] }, { "cell_type": "code", "execution_count": 166, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Polynomial Regression: R2 = -0.2126\n", "Polynomial Regression: RMSE = 0.2345\n" ] } ], "source": [ "poly_r2 = r2_score(y_test, y_predict)\n", "poly_rmse = np.sqrt(mean_squared_error(y_test, y_predict))\n", "print(f'Polynomial Regression: R2 = {poly_r2:.4f}')\n", "print(f'Polynomial Regression: RMSE = {poly_rmse:.4f}')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Decision Tree" ] }, { "cell_type": "code", "execution_count": 167, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "DecisionTreeRegressor(random_state=42)" ] }, "execution_count": 167, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dec_tree_regressor = DecisionTreeRegressor(random_state=42)\n", "dec_tree_regressor.fit(X_train,y_train)" ] }, { "cell_type": "code", "execution_count": 168, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([0.438, 0.813, 0.25 , 0.813, 0.375])" ] }, "execution_count": 168, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y_predict = dec_tree_regressor.predict(X_test)\n", "y_predict[:5]" ] }, { "cell_type": "code", "execution_count": 169, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Decision Tree Regression: R2 = -0.5154\n", "Decision Tree Regression: RMSE = 0.2622\n" ] } ], "source": [ "dec_tree_r2 = r2_score(y_test, y_predict)\n", "dec_tree_rmse = np.sqrt(mean_squared_error(y_test, y_predict))\n", "print(f'Decision Tree Regression: R2 = {dec_tree_r2:.4f}')\n", "print(f'Decision Tree Regression: RMSE = {dec_tree_rmse:.4f}')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Regression Notes\n", "There is some correlation between the spending, with QB, TE, Edge, and Safety spending somewhat more correlated than the other positions. \n", "However, none of the model produced any actionable predictions. With nine independent variables, we likely need more than just 288 observations to really find a pattern." ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.5" } }, "nbformat": 4, "nbformat_minor": 4 }