根据python案例，Whoscored评分对比？

wen python案例 2026-10-30 2

Whoscored 评分对比 - Python 实现方案

数据获取

Whoscored 网站有反爬机制，需使用 requests + BeautifulSoup 并模拟浏览器请求。

import requests
from bs4 import BeautifulSoup
import pandas as pd
import time
import random
def get_whoscored_data(url):
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '
                      'AppleWebKit/537.36 (KHTML, like Gecko) '
                      'Chrome/120.0.0.0 Safari/537.36',
        'Accept-Language': 'en-US,en;q=0.9',
    }
    resp = requests.get(url, headers=headers, timeout=15)
    soup = BeautifulSoup(resp.text, 'html.parser')
    return soup

⚠️ Whoscored 数据通常通过 JS 动态加载，更推荐使用 Selenium / Playwright 或第三方库 soccerdata：

# 推荐：soccerdata 直接抓取 Whoscored 数据
import soccerdata as sd
ws = sd.WhoScored(leagues="ENG-Premier League", seasons="2023/2024")
player_stats = ws.read_player_season_stats(stat_type="summary")
print(player_stats.columns.tolist())

典型对比场景

场景 1：同队球员评分对比

import matplotlib.pyplot as plt
import seaborn as sns
# 假设已有 dataframe
df = player_stats.reset_index()
# 筛选某队球员
team_df = df[df['team'] == 'Arsenal'].sort_values('rating', ascending=False)
plt.figure(figsize=(10, 6))
sns.barplot(data=team_df, x='rating', y='player', palette='viridis')'Arsenal 球员 Whoscored 评分 (2023/24)')
plt.xlabel('Whoscored Rating')
plt.tight_layout()
plt.show()

场景 2：多球员关键指标雷达图对比

import numpy as np
import matplotlib.pyplot as plt
def radar_compare(df, players, metrics):
    """多球员雷达图对比"""
    N = len(metrics)
    angles = np.linspace(0, 2 * np.pi, N, endpoint=False).tolist()
    angles += angles[:1]  # 闭合
    fig, ax = plt.subplots(figsize=(8, 8), subplot_kw=dict(polar=True))
    # 使用百分位归一化（消除量纲差异）
    for player in players:
        row = df[df['player'] == player].iloc[0]
        values = []
        for m in metrics:
            pct = (df[m] < row[m]).mean() * 100  # 百分位
            values.append(pct)
        values += values[:1]
        ax.plot(angles, values, label=player, linewidth=2)
        ax.fill(angles, values, alpha=0.15)
    ax.set_xticks(angles[:-1])
    ax.set_xticklabels(metrics)
    ax.set_title('Whoscored 多球员指标对比 (百分位)')
    ax.legend(loc='upper right', bbox_to_anchor=(1.3, 1.1))
    plt.show()
radar_compare(
    player_stats.reset_index(),
    players=['Bukayo Saka', 'Mohamed Salah', 'Son Heung-Min'],
    metrics=['goals', 'assists', 'shots_per_game', 'key_passes_per_game',
             'dribbles_per_game', 'rating']
)

场景 3：赛季评分趋势对比

# 若抓取到每轮评分数据
def plot_rating_trend(round_data, players):
    plt.figure(figsize=(12, 5))
    for p in players:
        sub = round_data[round_data['player'] == p].sort_values('round')
        plt.plot(sub['round'], sub['rating'], marker='o', label=p)
    plt.axhline(7.0, color='gray', linestyle='--', label='优秀线 7.0')
    plt.xlabel('比赛轮次')
    plt.ylabel('Whoscored 评分')
    plt.title('赛季评分趋势对比')
    plt.legend()
    plt.grid(alpha=0.3)
    plt.show()

实用对比分析函数

def compare_players(df, players, metrics=None, return_df=True):
    """返回多球员核心指标对比表"""
    if metrics is None:
        metrics = ['rating', 'goals', 'assists', 'shots_per_game',
                   'key_passes_per_game', 'pass_success',
                   'dribbles_per_game', 'tackles_per_game']
    compare = df[df['player'].isin(players)][['player'] + metrics].copy()
    compare = compare.set_index('player')
    if return_df:
        return compare.round(2)
# 使用
result = compare_players(df, ['Erling Haaland', 'Harry Kane', 'Kylian Mbappe'])
print(result.T)  # 转置更直观

注意事项与避坑

问题	解决方案
反爬封锁 403	加 `User-Agent`、`Referer`；使用 `Playwright`
数据动态加载	用 `soccerdata` 库 / Selenium 渲染后解析
换行符污染字段	`df['col'].str.replace('\n','')`
百分位口径	建议按位置（FW/MF/DF）分组计算百分位
评分无法跨赛季直接比较	Whoscored 评分是相对值，需注意样本口径
Compliance	仅用于学习研究，遵守网站 `robots.txt` 及 ToS