optimize_lineup.py

import pyfiglet
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
sns.set(style='white')
from collections import defaultdict
import numpy as np
import click
from tabulate import tabulate
import logging
logging.basicConfig()
logging.getLogger().setLevel(logging.INFO)
from nba_matchup import get_league, simulate_h2h, CURRENT_WEEK, hill_climb, visualize_matchup, get_free_agents, simulated_annealing
league = get_league()

def winning_prob(cats, points, scores, num_samples):
    unique, nums = np.unique(points, return_counts=True)
    counts = defaultdict(int)
    counts.update(dict(zip(unique, nums)))
    return sum([counts[p] for p in range(5, 10)]) / num_samples

def ev(cats, points, scores, num_samples):
    return points.mean()

@click.command()
@click.option('--team1', type=str, default=None)
@click.option('--team2', type=str, default=None)
@click.option('--num_days', type=int, default=30)
@click.option('--num_samples', type=int, default=50000)
@click.option('--week', type=int, default=CURRENT_WEEK)
@click.option('--num_fa', type=int, default=0)
@click.option('--num_iters', type=int, default=100)
@click.option('--ignore_player', type=str, multiple=True)
@click.option('--half_life', type=float, default=14)
@click.option('--metric', type=str, default='winning_probability')
@click.option('--ignore_injured', is_flag=True)
def main(team1, team2, num_days, num_samples, week, num_fa, num_iters,
         ignore_player, half_life, metric, ignore_injured):
    league = get_league()
    decay_rate = np.log(2) / half_life
    # if week == 19: week = 18
    # if week >= 19:
        # week -= 1
    if team1 is None:
        team1 = league.current_team
    else:
        team1 = league.team_by_owner(team1)
    if team2 is None:
        team2 = league.get_matchup(team1, week=week)
    else:
        team2 = league.team_by_owner(team2)

    pyfiglet.print_figlet("%s vs. %s" % (team1.manager_name,
                                            team2.manager_name), font='banner',
                          width=160)
    if week:
        pyfiglet.print_figlet("Week %u" % week, font='big')
    if metric == 'ev':
        metric_fn = ev
    else:
        metric_fn = winning_prob
    def roster_score(roster):
        cats, points, scores, _ = simulate_h2h(roster,
                            team2.roster(week=week),
                            num_days=num_days, num_samples=num_samples,
                            week=week, decay_rate=decay_rate)
        return metric_fn(cats, points, scores, num_samples)
    def reverse_roster_score(roster):
        cats, points, scores, _ = simulate_h2h(roster,
                            team1.roster(week=week),
                            num_days=num_days, num_samples=num_samples,
                            week=week, decay_rate=decay_rate)
        return metric_fn(cats, points, scores, num_samples)
    print("%s's roster:" % team1.manager_name, roster_score(team1.roster(week=week)))
    print(tabulate([
        [position, player.name] for player, position in
        team1.roster(week=week).positions.items() if position not in {"BN", "IL"}
    ]))
    print("%s's roster:" % team2.manager_name, reverse_roster_score(team2.roster(week=week)))
    print(tabulate([
        [position, player.name] for player, position in
        team2.roster(week=week).positions.items() if position not in {"BN", "IL"}
    ]))
    print("Optimizing %s's lineup" % team1.manager_name)
    print("===========================================")
    roster = team1.roster(week=week)
    old_roster = roster
    print("Adding free agents:")
    for agent in get_free_agents(num_fa):
        print(agent.name)
        roster = roster.add(agent, "BN")
    team1.set_roster(roster)
    print("Ignoring players:", ", ".join(ignore_player))
    scores = []
    for roster, score in simulated_annealing(roster, roster_score, ignore_players={team1.roster(week=week).player_by_name(n) for n in ignore_player},
                                    num_steps=num_iters,
                                    ignore_injured=ignore_injured):
        scores.append(score)
        # print(tabulate([
            # [position, player.name] for player, position in
            # roster.positions.items() if position not in {"BN", "IL"}
        # ]))
    print("%s's optimized roster:" % team1.manager_name, score)
    print(tabulate([
        [position, player.name] for player, position in
        roster.positions.items() if position not in {"BN", "IL"}
    ]))

    def team_generator():
        for r in [old_roster, roster]:
            team1.set_roster(r)
            yield team1

    projections = visualize_matchup(team_generator(), team2,
                      num_days=num_days, num_samples=100000,
                      week=week, decay_rate=decay_rate,
                                    show_plots=False)
    with pd.option_context('display.max_rows', None, 'display.max_columns',
                           None, 'display.expand_frame_repr', False):
        for i, team in enumerate([team1, team2]):
            print("===========================================")
            print("%s's projections:" % team.manager_name)
            print(projections[1][i].round(2))
    plt.figure()
    plt.plot(scores)
    plt.show()

if __name__ == "__main__":
    main()