Yahoo NBA Fantasy Positions

fantasy_positions.py

#!/usr/bin/python

import requests
import numpy as np
import pandas as pd
from lxml import html

base_url = "http://sports.yahoo.com/nba/teams"
teams = [
    "atlanta",
    "boston",
    "brooklyn",
    "new-york",
    "philadelphia",
    "toronto",
    "chicago",
    "cleveland",
    "detroit",
    "indiana",
    "milwaukee",
    "atlanta",
    "charlotte",
    "miami",
    "orlando",
    "washington",
    "golden-state",
    "la-clippers",
    "la-lakers",
    "phoenix",
    "sacramento",
    "dallas",
    "houston",
    "memphis",
    "new-orleans",
    "san-antonio",
    "denver",
    "minnesota",
    "oklahoma-city",
    "portland",
    "utah",
]
nxpath = """//*[@id="Col1-0-Team-Proxy"]/div/div[3]/div[2]/div[2]/table/tbody/tr[*]/td[2]/div/div/a"""
pxpath = """//*[@id="Col1-0-Team-Proxy"]/div/div[3]/div[2]/div[2]/table/tbody/tr[****]/td[3]/div/span/text()"""


def main():
    rows = []
    for team in teams:
        roster_url = f"{base_url}/{team}/roster/"
        content = requests.get(roster_url).content.decode()
        tree = html.fromstring(content)
        names = [n.text for n in tree.xpath(nxpath) if n.text is not None]
        for i, name in enumerate(names):
            sname = name.split(" ")
            output_name = ", ".join(sname[1:] + [sname[0]])
            positions = [
                p for p in tree.xpath(pxpath.replace("****", str(i + 1))) if p != "/"
            ]
            print(i, output_name, positions)
            output_positions = ", ".join(positions)
            pdict = {"name": output_name, "position": output_positions}
            rows.append(pdict)

    df = pd.DataFrame(rows)
    df = df.replace(np.nan, "", regex=True)
    df.to_csv("positions.csv", index=False)


if __name__ == "__main__":
    main()

positions2016.csv

positions2017.csv

positions2019.csv

positions2020.csv

positions2022.csv