新闻

CBA球员赛季表现排行榜（CBA球员赛季发挥排行榜）

2026-01-05

Finalizing next steps

I'm considering how to run the Python script rank_cba.py with a specific path. However, it seems like we shouldn't actually execute it just yet. Instead, it might be better to focus on proposing the next steps for the user. I want to make sure everything is clear before moving forward and that the user feels supported in whatever they need to do next. Let's go ahead and produce that final answer!

想做哪个赛季、什么口径的“赛季表现排行榜”？我先确认下需求就能给你榜单或脚本。

请确认

误率使用

赛季：如 2024-25 常规赛，或某历史赛季
比赛范围：常规赛/季后赛/合计
球员范围：是否区分外援/本土；最少出场或场均分钟门槛
指标：单项领袖（得分/篮板/助攻/抢断/盖帽/命中率/三分/TS%/正负值等）还是综合评分榜
输出：TOP10/20，是否要导出为 CSV/Excel/图片

可选方案

快速领袖榜：我按你指定的指标给出各项 TOP10（需要赛季与口径）。
综合评分榜：用标准化后的综合评分（可调权重）生成总榜，另附单项榜。
数据驱动：你给我本地 CSV/Excel，我用脚本清洗、算指标、导出榜单和图表。

如果你已有数据文件，下面这个最小脚本可直接把任意赛季的 CSV 生成多项排行榜与一个“综合评分”榜。你只要确保列名包含常见统计：Player, Team, GP, MIN, PTS, REB, AST, STL, BLK, TOV, FGA, FGM, 3PA, 3PM, FTA, FTM。

示例脚本（Python）

# scripts/cba_rank.py
import argparse, pandas as pd, numpy as np

def ts_pct(df):
    pts = df["PTS"]; fga = df["FGA"]; fta = df["FTA"]
    denom = (2*(fga + 0.44*fta)).replace(0, np.nan)
    return (pts/denom).fillna(0.0)

def per36(df, col):
    return (df[col] / df["MIN"] * 36).replace([np.inf, -np.inf], 0).fillna(0.0)

def zscore(s):
    return (s - s.mean()) / (s.std(ddof=0) if s.std(ddof=0) != 0 else 1)

def build_score(df):
    # Simple, interpretable composite; adjust weights as needed.
    df = df.copy()
    df["PTS36"] = per36(df, "PTS")
    df["REB36"] = per36(df, "REB")
    df["AST36"] = per36(df, "AST")
    df["STK36"] = per36(df, "STL") + per36(df, "BLK")
    df["TOV36"] = per36(df, "TOV")
    df["TS"] = ts_pct(df)

![件放到](/assets/images/3E9C2C1F.jpeg)

    score = (
        0.40*zscore(df["PTS36"]) +
        0.18*zscore(df["TS"]) +
        0.15*zscore(df["AST36"]) +
        0.12*zscore(df["REB36"]) +
        0.10*zscore(df["STK36"]) -
        0.15*zscore(df["TOV36"])
    )
    return score

def rank_top(df, by, n=10, desc=True):
    cols = ["Player", "Team", "GP", "MIN", by]
    return df.sort_values(by=by, ascending=not desc)[cols].head(n)

def main():
    ap = argparse.ArgumentParser()
    ap.add_argument("csv", help="input stats CSV")
    ap.add_argument("--min-gp", type=int, default=10)
    ap.add_argument("--min-min", type=float, default=10.0)
    ap.add_argument("--top", type=int, default=10)
    ap.add_argument("--out", default=None, help="optional output CSV prefix")
    args = ap.parse_args()

    df = pd.read_csv(args.csv)
    required = ["Player","Team","GP","MIN","PTS","REB","AST","STL","BLK","TOV","FGA","FGM","3PA","3PM","FTA","FTM"]
    missing = [c for c in required if c not in df.columns]
    if missing:
        raise SystemExit(f"Missing columns: {missing}")

    df = df[(df["GP"] >= args.min_gp) & (df["MIN"] >= args.min_min)].copy()
    df["TS%"] = ts_pct(df) * 100
    df["Score"] = build_score(df)

    leaderboards = {
        "Score": rank_top(df, "Score", args.top),
        "PTS":   rank_top(df, "PTS", args.top),
        "REB":   rank_top(df, "REB", args.top),
        "AST":   rank_top(df, "AST", args.top),
        "STL":   rank_top(df, "STL", args.top),
        "BLK":   rank_top(df, "BLK", args.top),
        "TS%":   rank_top(df, "TS%", args.top),
    }

    for name, tbl in leaderboards.items():
        print(f"\n=== TOP {args.top} {name} ===")
        print(tbl.to_string(index=False))
        if args.out:
            tbl.to_csv(f"{args.out}_{name}.csv", index=False, encoding="utf-8")

if __name__ == "__main__":
    main()

想法与扩展

门槛与归一化：用场均分钟或出场场次过滤，按每36分钟或每百回合标准化更公平。
指标可加：命中率、三分命中率、罚球率、正负值、进攻/防守篮板、失误率、使用率；若有回合数可算 ORtg/DRtg。
分组榜：可分别给本土球员榜与外援榜。
输出：我可加导出 Excel 工作簿、PNG 表格或条形图。

告诉我赛季与口径，或者把数据文件放到仓库里（路径+列名），我马上给你生成对应的排行榜。