tanzhijian.org

阿森纳不败赛季传球路线图探索 04

这一次探索单个球员的进攻传球路线。首先还是画出热区图，准备数据的部分和前面一篇相同，就不重复写了，直接跳到 danger_passes

import pandas as pd
import matplotlib.pyplot as plt
from mplsoccer import Pitch, Sbopen, VerticalPitch

danger_passes.head()

	x	y	end_x	end_y	minute	second	player_name	player_id
0	52.6	54.3	54.6	45.0	8	4	Patrick Vieira	15515.0
1	69.9	28.5	102.0	26.9	8	8	Eduardo César Daude Gaspar	26014.0
2	102.0	26.9	104.7	37.6	8	10	Ashley Cole	12529.0
3	60.0	45.6	68.9	33.9	12	7	Robert Pirès	19312.0
4	68.5	33.9	79.1	31.1	12	9	Eduardo César Daude Gaspar	26014.0

探索单个球员

作为威胁传球的核心，首先来探索皮雷，手动在名单里获取皮雷的 id，然后过滤

bins = (6, 5)

pires = danger_passes.loc[danger_passes['player_id'] == 19312]

def plot_player_passes(player, passes):
    pitch = Pitch(line_zorder=2, line_color="grey")
    fig, ax = pitch.grid(
        grid_height=0.9,
        title_height=0.06,
        axis=False,
        endnote_height=0.04,
        title_space=0,
        endnote_space=0,
    )

    bin_statistic = pitch.bin_statistic(
        player.x,
        player.y,
        statistic="count",
        bins=bins,
        normalize=False,
    )

    # 制作热区图
    pcm = pitch.heatmap(
        bin_statistic, 
        cmap="Reds", 
        # edgecolor="grey", 
        ax=ax["pitch"],
    )

    # 绘制传球路线
    pitch.arrows(
        passes.x,
        passes.y,
        passes.end_x,
        passes.end_y,
        color="black",
        alpha=1,
        width=2,
        ax=ax["pitch"],
    )

    fig.suptitle(f"{player.iloc[0].player_name}", fontsize=30)
    plt.show()

plot_player_passes(pires, pires)

这样的路线看起来未免杂乱，这里使用 Opta 的一种传球表现图：球场分成 10 * 10 的区域，每一块区域传球次数越多颜色越深，箭头表示此区域的平均传球方向

bins = (10, 10)

def binning_groupby_mean(player):
    x_bins = [i for i in range(0, 121, 120 // bins[0])]
    y_bins = [i for i in range(0, 81, 80 // bins[1])]
    player['x_bin'] = pd.cut(player['x'], x_bins)
    player['y_bin'] = pd.cut(player['y'], y_bins)
    return player.groupby(['x_bin', 'y_bin']).mean()

mean_passes = binning_groupby_mean(pires)
plot_player_passes(pires, mean_passes)

边路进攻核心活动区域，次数最多的是左路肋部区域，在较少拿球的区域传球的距离比较长，到了不是自己的位置会将球传到熟悉的区域而不是就地组织，很少有长距离传中路线

vieira = danger_passes.loc[danger_passes['player_id'] == 15515]
mean_passes = binning_groupby_mean(vieira)
plot_player_passes(vieira, mean_passes)

覆盖了整个中场，核心传球区域与皮雷很近，传球选择路线更偏向左路

henry = danger_passes.loc[danger_passes['player_id'] == 15516]
mean_passes = binning_groupby_mean(henry)
plot_player_passes(henry, mean_passes)

与印象不同的是，在中路禁区前沿有着大量的传球，且都是向前的路线，往禁区送出了大量的威胁球，所以在进球同时会收获那么多的助攻

player = danger_passes.loc[danger_passes['player_id'] == 15042]
mean_passes = binning_groupby_mean(player)
plot_player_passes(player, mean_passes)

活动位置更像是进攻型中场，传球也大多找的禁区前的队友

player = danger_passes.loc[danger_passes['player_id'] == 15754]
mean_passes = binning_groupby_mean(player)
plot_player_passes(player, mean_passes)

与皮雷不一样的是，就地选择进攻传球更为频繁，习惯性活动范围固定在几块，可能与他场上位置更多变有关

传球路线图就到这里了，下一次探索些别的