Python Selenium爬取PrizePicks NBA Fantasy数据及ESPN对比求助

第一部分 PrizePicks平台数据爬取修复 #

原代码核心错误 #

• find_element_by_class_name('projections')返回单个父容器元素，不可迭代，循环直接报错，需先定位所有球员卡片节点
• 方法名拼写错误：fine_element_by_xpath应为find_element_by_xpath
• 使用绝对路径xpath，依赖页面固定层级，稳定性极差
• 无元素等待逻辑，页面未渲染完成就查找元素大概率返回空

修复后代码 #

from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
import pandas as pd

PATH = "C:\Program Files (x86)\chromedriver.exe"
driver = webdriver.Chrome(PATH)
driver.get("https://app.prizepicks.com/")
wait = WebDriverWait(driver, 15)

# 关闭弹窗
wait.until(EC.element_to_be_clickable((By.CLASS_NAME, "close"))).click()
# 选择NBA分类
wait.until(EC.element_to_be_clickable((By.XPATH, "//div[@class='name'][normalize-space()='NBA']"))).click()
# 选择Fantasy Score标签
wait.until(EC.element_to_be_clickable((By.CLASS_NAME, "segment-selector-button"))).click()
# 等待所有球员卡片加载完成
player_cards = wait.until(EC.presence_of_all_elements_located((By.CSS_SELECTOR, "#projections .projection")))

nbaPlayers = []
for card in player_cards:
    # 相对路径匹配当前卡片内的姓名和分值
    name = card.find_element(By.CSS_SELECTOR, ".player-name").text
    points = float(card.find_element(By.CSS_SELECTOR, ".score").text)
    nbaPlayers.append({
        "Name": name,
        "FantasyPoints": points
    })

df_prizepicks = pd.DataFrame(nbaPlayers)
print(df_prizepicks)
driver.quit()

第二部分 ESPN球员近10场数据爬取修复 #

原代码核心错误 #

• xpath语法错误：//*[@id='fittPageContainer'']多写了一个单引号，直接定位失败
• 只取了最近1场比赛的数据，未计算近10场平均值，不符合需求
• 未将元素文本转为数值类型，直接进行算术运算会报错
• 跳转球员页后无等待逻辑，表格未加载就查找元素会报错

修复后核心逻辑代码 #

# 接上面df_prizepicks数据，逐行遍历处理
driver = webdriver.Chrome(PATH)
wait = WebDriverWait(driver, 20)
result = []
for idx, row in df_prizepicks.iterrows():
    player_name = row["Name"]
    prizepicks_score = row["FantasyPoints"]
    
    # 打开ESPN搜索
    driver.get("https://www.espn.com/")
    # 触发搜索框
    wait.until(EC.element_to_be_clickable((By.ID, "global-search-trigger"))).click()
    search_input = wait.until(EC.element_to_be_clickable((By.XPATH, "//input[@placeholder='Search Sports, Teams or Players...']")))
    search_input.clear()
    search_input.send_keys(player_name)
    # 点击第一个搜索结果
    wait.until(EC.element_to_be_clickable((By.CSS_SELECTOR, ".search_results__details"))).click()
    # 等待比赛数据表格加载
    stats_table = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, "table.Table--align-right tbody")))
    game_rows = stats_table.find_elements(By.TAG_NAME, "tr")[:10] # 取最近10场
    
    total_pts = 0
    total_reb = 0
    total_ast = 0
    total_blk = 0
    total_stl = 0
    total_tov = 0
    valid_games = 0
    for game in game_rows:
        cols = game.find_elements(By.TAG_NAME, "td")
        if len(cols) < 13:
            continue
        try:
            pts = float(cols[12].text)
            reb = float(cols[6].text)
            ast = float(cols[7].text)
            blk = float(cols[8].text)
            stl = float(cols[9].text)
            tov = float(cols[11].text)
            total_pts += pts
            total_reb += reb
            total_ast += ast
            total_blk += blk
            total_stl += stl
            total_tov += tov
            valid_games += 1
        except:
            continue
    
    if valid_games == 0:
        continue
    # 计算近10场平均Fantasy分值
    avg_pts = total_pts / valid_games
    avg_reb = total_reb / valid_games
    avg_ast = total_ast / valid_games
    avg_blk = total_blk / valid_games
    avg_stl = total_stl / valid_games
    avg_tov = total_tov / valid_games
    espn_projected = avg_pts + avg_reb*1.2 + avg_ast*1.5 + avg_blk*3 + avg_stl*3 - avg_tov*1
    
    result.append({
        "Name": player_name,
        "FantasyPoints": prizepicks_score,
        "ESPN L10 Avg": round(espn_projected, 1),
        "Judgement": "高于预测" if espn_projected > prizepicks_score else "低于预测"
    })

df_final = pd.DataFrame(result)
print(df_final)
driver.quit()

实现逻辑说明 #

1. 先爬取PrizePicks平台所有NBA球员的Fantasy预测分值，存入DataFrame
2. 遍历DataFrame中每条球员数据，在ESPN搜索对应球员
3. 爬取球员最近10场有效比赛的基础数据，计算平均Fantasy分值
4. 对比两个平台的分值，输出高低判断结果

内容的提问来源于stack exchange，提问作者Consiglere