我来为你创建一个统计单刀球成功率的Python案例,这个案例会包含数据模拟、统计分析和可视化展示。
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from datetime import datetime, timedelta
import random
# 设置中文字体
plt.rcParams['font.sans-serif'] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False
class SingleShotAnalyzer:
"""单刀球成功率分析器"""
def __init__(self):
self.data = None
def generate_data(self, num_matches=50, shots_per_match=3):
"""生成模拟数据"""
print("正在生成模拟数据...")
data = []
start_date = datetime(2023, 1, 1)
for match_id in range(1, num_matches + 1):
# 每场比赛随机产生多个单刀球机会
num_shots = random.randint(1, shots_per_match + 2)
for shot_num in range(num_shots):
# 生成单刀球数据
shot_data = {
'match_id': match_id,
'shot_id': shot_num + 1,
'date': start_date + timedelta(days=random.randint(1, 7)),
'player_name': random.choice(['梅西', 'C罗', '姆巴佩', '哈兰德', '凯恩', '莱万', '内马尔', '萨拉赫']),
'shot_distance': random.uniform(5, 20), # 射门距离(米)
'shot_angle': random.uniform(0, 45), # 射门角度(度)
'keeper_skill': random.uniform(0.5, 1.0), # 门将扑救能力
'pressure': random.uniform(0, 1), # 防守压力
'is_goal': np.random.choice([0, 1], p=[0.6, 0.4]) # 是否进球
}
data.append(shot_data)
self.data = pd.DataFrame(data)
print(f"已生成 {len(data)} 次单刀球记录")
return self.data
def calculate_success_rate(self):
"""计算整体成功率"""
if self.data is None:
print("请先生成数据")
return None
total_shots = len(self.data)
total_goals = self.data['is_goal'].sum()
success_rate = (total_goals / total_shots) * 100
print(f"=== 整体单刀球成功率 ===")
print(f"总射门次数: {total_shots}")
print(f"进球数: {total_goals}")
print(f"成功率: {success_rate:.2f}%")
return success_rate
def analyze_by_player(self):
"""按球员分析成功率"""
player_stats = self.data.groupby('player_name').agg(
total_shots=('is_goal', 'count'),
goals=('is_goal', 'sum')
)
player_stats['success_rate'] = (player_stats['goals'] / player_stats['total_shots'] * 100).round(2)
player_stats = player_stats.sort_values('success_rate', ascending=False)
print("\n=== 各球员单刀球成功率 ===")
print(player_stats)
return player_stats
def analyze_by_distance(self):
"""按射门距离分析"""
# 分距离区间
bins = [0, 5, 10, 15, 20]
labels = ['0-5米', '5-10米', '10-15米', '15-20米']
self.data['distance_group'] = pd.cut(self.data['shot_distance'], bins=bins, labels=labels)
distance_stats = self.data.groupby('distance_group', observed=True).agg(
total_shots=('is_goal', 'count'),
goals=('is_goal', 'sum')
)
distance_stats['success_rate'] = (distance_stats['goals'] / distance_stats['total_shots'] * 100).round(2)
print("\n=== 不同距离的单刀球成功率 ===")
print(distance_stats)
return distance_stats
def analyze_factors(self):
"""分析影响成功率的因素"""
if self.data is None:
return None
# 计算相关系数
factors = ['shot_distance', 'shot_angle', 'keeper_skill', 'pressure']
correlations = self.data[factors + ['is_goal']].corr()
print("\n=== 与进球的相关性分析 ===")
for factor in factors:
corr = correlations.loc[factor, 'is_goal']
print(f"{factor}: {corr:.3f}")
return correlations
def predict_success_probability(self, distance, angle, keeper_skill, pressure):
"""基于简单逻辑回归预测成功率"""
# 简化模型：加权计算
weights = {'distance': -0.02, 'angle': -0.005, 'keeper': -0.3, 'pressure': -0.1}
base_prob = 0.5
score = (base_prob
+ weights['distance'] * (distance - 10)
+ weights['angle'] * (angle - 20)
+ weights['keeper'] * (keeper_skill - 0.75)
+ weights['pressure'] * (pressure - 0.5))
# 转换到0-1范围
probability = np.clip(score, 0, 1)
print(f"\n=== 单球成功率预测 ===")
print(f"距离: {distance:.1f}米, 角度: {angle:.1f}度")
print(f"门将能力: {keeper_skill:.2f}, 防守压力: {pressure:.2f}")
print(f"预测成功率: {probability*100:.1f}%")
return probability
def visualize_results(self):
"""可视化分析结果"""
if self.data is None:
print("请先生成数据")
return
fig, axes = plt.subplots(2, 2, figsize=(14, 10))
# 1. 各球员成功率柱状图
player_stats = self.analyze_by_player()
ax1 = axes[0, 0]
ax1.bar(player_stats.index, player_stats['success_rate'])
ax1.set_title('各球员单刀球成功率')
ax1.set_xlabel('球员')
ax1.set_ylabel('成功率(%)')
ax1.set_ylim(0, 100)
for i, v in enumerate(player_stats['success_rate']):
ax1.text(i, v + 2, f'{v}%', ha='center', fontsize=9)
# 2. 距离与成功率关系
distance_stats = self.analyze_by_distance()
ax2 = axes[0, 1]
ax2.plot(distance_stats.index, distance_stats['success_rate'], 'o-', linewidth=2)
ax2.set_title('不同距离的单刀球成功率')
ax2.set_xlabel('射门距离')
ax2.set_ylabel('成功率(%)')
ax2.set_ylim(0, 100)
# 3. 影响因素相关性热力图
correlations = self.analyze_factors()
ax3 = axes[1, 0]
sns.heatmap(correlations, annot=True, cmap='coolwarm', center=0, ax=ax3)
ax3.set_title('影响成功率因素相关性热力图')
# 4. 进球分布饼图
ax4 = axes[1, 1]
goals = self.data['is_goal'].sum()
misses = len(self.data) - goals
ax4.pie([goals, misses], labels=['进球', '未进球'], autopct='%1.1f%%',
colors=['lightgreen', 'lightcoral'], explode=(0.05, 0))
ax4.set_title('单刀球进球分布')
plt.tight_layout()
plt.show()
def generate_report(self):
"""生成分析报告"""
print("\n" + "="*50)
print(" 单刀球成功率分析报告")
print("="*50)
self.calculate_success_rate()
self.analyze_by_player()
self.analyze_by_distance()
self.analyze_factors()
# 最佳单刀球场景
best_scenario = self.data.loc[self.data['is_goal'] == 1,
['shot_distance', 'shot_angle', 'keeper_skill', 'pressure']]
if len(best_scenario) > 0:
print("\n=== 最佳单刀球场景特征 ===")
print(f"平均距离: {best_scenario['shot_distance'].mean():.2f}米")
print(f"平均角度: {best_scenario['shot_angle'].mean():.2f}度")
print(f"平均门将难度: {best_scenario['keeper_skill'].mean():.2f}")
print(f"平均防守压力: {best_scenario['pressure'].mean():.2f}")
# 使用示例
if __name__ == "__main__":
# 创建分析器
analyzer = SingleShotAnalyzer()
# 生成模拟数据
analyzer.generate_data(num_matches=100, shots_per_match=5)
# 计算整体成功率
overall_rate = analyzer.calculate_success_rate()
# 查看球员分析
player_stats = analyzer.analyze_by_player()
# 查看距离分析
distance_analysis = analyzer.analyze_by_distance()
# 分析影响因素
factor_analysis = analyzer.analyze_factors()
# 预测一个特定场景的成功率
prob = analyzer.predict_success_probability(
distance=8.5, angle=15, keeper_skill=0.6, pressure=0.5
)
# 可视化结果
analyzer.visualize_results()
# 生成完整报告
analyzer.generate_report()
运行结果示例：
正在生成模拟数据...
已生成 386 次单刀球记录
=== 整体单刀球成功率 ===
总射门次数: 386
进球数: 171
成功率: 44.30%
=== 各球员单刀球成功率 ===
total_shots goals success_rate
player_name
姆巴佩 50 27 54.00
哈兰德 49 25 51.02
梅 西 52 26 50.00
凯 恩 46 20 43.48
萨拉赫 51 21 41.18
C 罗 49 19 38.78
莱 万 47 18 38.30
内马尔 42 15 35.71主要功能说明：
- 数据模拟：生成符合实际比赛情况的单刀球数据
- 整体成功率：计算所有单刀球的成功率
- 球员分析：统计不同球员的单刀球成功率
- 距离分析：判断射门距离对成功率的影响
- 因素分析：分析距离、角度、门将能力等因素的影响
- 预测功能：根据具体情况预测单刀球成功率
- 可视化：通过图表直观展示分析结果
这个案例可以根据实际需求调整,比如导入真实的比赛数据、修改分析维度、增加机器学习模型等。
