tanzhijian.org

阿森纳不败赛季传球路线图探索 03

这次用热区图的形式来探索，与平时所看的全场比赛事件分布热区图不一样的是，用的是危险传球——这次把射门前 15 秒内的传球都视为危险传球。

前面获取数据部分和上一篇大致相同

import

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

parser = Sbopen()
matches = parser.match(competition_id=2, season_id=44)
matches_events = []
for match_id in matches['match_id']:
    data = parser.event(match_id)
    matches_events.append(data[0])

获取危险传球

获取阿森纳所有比赛的射门和准确传球，且传球不是定位球，然后寻找射门前 15 秒的传球，对于时间需要考虑上下半场，伤停补时的特殊处理

danger_passes = pd.DataFrame()
for events in matches_events:
    # 上下半场
    for period in [1, 2]:
        # 只保留没有定位球的准确传球
        pass_filter = (
            events.team_name == 'Arsenal') & (
            events.type_name == "Pass") & (
            events.outcome_name.isnull()) & (
            events.period == period) & (
            events.sub_type_name.isnull())
        passes = events.loc[
            pass_filter, 
            [
                "x", "y", "end_x", "end_y", 
                "minute", "second", "player_name",
            ],
        ]
        # 只保留阿森纳的射门
        shot_filter = (
            events.team_name == 'Arsenal') & (
            events.type_name == "Shot") & (
            events.period == period)
        shots = events.loc[shot_filter, ["minute", "second"]]
        shot_times = shots['minute'] * 60 + shots['second']
        
        # 设置为射门前 15 秒
        shot_anchor = 15
        shot_start = shot_times - shot_anchor
        
        # 处理时间
        shot_start = shot_start.apply(lambda i: i if i > 0 else (period - 1) * 45)
        pass_times = passes['minute'] * 60 + passes['second']

        pass_to_shot = pass_times.apply(
            lambda x: True in ((shot_start < x) & (x < shot_times)).unique())

        # 只保留危险传球
        danger_passes_period = passes.loc[pass_to_shot]
        danger_passes = pd.concat(
            [danger_passes, danger_passes_period], 
            ignore_index=True
        )

danger_passes.head()

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

绘制热区图

这次在球场上使用 6 * 5 的格子来制作，其中宽度区域对应中肋边路，六块长度区域在瓜迪奥拉的战术板经常能看到

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(
    danger_passes.x, danger_passes.y, 
    statistic='count', bins=(6, 5), normalize=False,
)

# 绘制热区图
pcm = pitch.heatmap(
    bin_statistic, cmap='Reds', edgecolor='black', ax=ax['pitch']
)

# 绘制说明
ax_cbar = fig.add_axes((1, 0.093, 0.03, 0.786))
cbar = plt.colorbar(pcm, cax=ax_cbar)

fig.suptitle('Danger passes by Arsenal', fontsize=30)
plt.show()

绘制参与危险传球最多球员的图表

为了更加清晰的探索，将球员参与的传球次数绘制出来

这次偷个懒就不与 nickname 结合了

passes_count = danger_passes.groupby(['player_name']).x.count()

ax = passes_count.plot.barh(passes_count)
ax.set_xlabel('Number of danger passes')
ax.set_ylabel('')
plt.show()

结论

维埃拉-皮雷-亨利占据了形成射门的大多数传球，球会经过他们很多次，是整个进攻体系的核心
博格坎普受限于出场(21次首发8次替补)，他的进攻参与度重要性不亚于 1
亨利作为终结者，在进攻传球网络中也有如此重要的数据，全能
左路肋部区域承担了推进至禁区的大多任务，
右路参与的控球大多数会通过中路交到左路，少部分会在右路传到禁区