英格兰超级足球联赛分析
郭迪 218025200042
2019-01-16
1 简介
英格兰超级足球联赛（Premier League）成立于1992年2月20日，属于“欧洲五大联赛”之一。它由二十支队伍组成， 采用双循环赛制。英超通常被认为是世界上最好的联赛，也是全世界最受欢迎的体育赛事之一。英超的比赛场面非常激烈，注重身体对抗，比赛节奏较快。许多传奇球星都曾在英超效力，如吉格斯、C罗以及亨利等。
本文使用的两个数据集分别记录了自2006/07赛季到2017/18赛季每场比赛的结果以及各队的技术统计。主要通过对数据的可视化以及建模来对英超的球队、比赛情况以及其他热点问题(德比等)进行分析。
2 准备工作
2.1 导入所需包
library("RColorBrewer")
library("ggthemes")
library("ggrepel")
library("gridExtra")
library("reshape2")
library("corrplot")
library("tidyverse")
library("highcharter")
library("data.table")
library("lars")2.2 导入数据
3 总体概况
3.1 赛季
首先分析各赛季的比赛结果、进球情况以及比分分布。
3.1.1 比赛结果
results %>%
group_by(result, season) %>%
count() %>%
ggplot(aes(reorder(result, n), y = n, fill = n)) +
geom_bar(stat = "identity", position = "dodge") +
xlab("Results") +
ylab("Results Count") +
ylim(0,210)+
facet_wrap( ~ season) +
scale_fill_gradientn(colours = brewer.pal(10,'Spectral')) +
mytheme+
theme(axis.text.x=element_text(angle=0) )+
geom_text(aes(label = n), vjust = -0.2, size = 10.5) +
ggtitle('Results Count through the Seasons') 
Results by Season
从赛果来看，主场球队要比客场球队更具优势。每个赛季主队的获胜的场次远高于客队，通常主队是客队的两倍，而平局场次与客队获胜场次几乎一致。
作为主队的确存在许多优势：熟悉的球场，充沛的精力以及球迷的支持等。这些因素使主场球队的实力增强：让强队更加难以战胜；让弱队有爆冷的机会。上赛季“喜鹊”（纽卡斯尔）就在最后一轮主场3比0击败切尔西。
3.1.2 进球数
tot_home_goals <- results %>%
group_by(season) %>%
summarise(total_home_goals = sum(home_goals)) %>%
ggplot(aes(x = season, y = total_home_goals, fill = total_home_goals)) +
geom_bar(stat = "identity", position = "dodge") +
xlab("") +
ylab("") +
ylim(0,700)+
mytheme+
theme(axis.text.x = element_blank())+
scale_fill_gradientn(name = '',colours = rev(brewer.pal(10,'Spectral'))) +
geom_text(aes(label = total_home_goals), vjust = -0.3, size = 10.5) +
ggtitle("Total Home Goals by Season")
tot_away_goals <- results %>%
group_by(season) %>%
summarise(total_away_goals = sum(away_goals)) %>%
ggplot(aes(x = season, y = total_away_goals, fill = total_away_goals)) +
geom_bar(stat = "identity", position = "dodge") +
xlab("") +
ylab("") +
ylim(0,700)+
mytheme+
scale_fill_gradientn(name = '',colours = rev(brewer.pal(10,'Spectral'))) +
geom_text(aes(label = total_away_goals), vjust = -0.3, size = 10.5) +
ggtitle("Total Away Goals by Season")
grid.arrange(tot_home_goals,tot_away_goals, ncol = 1, nrow = 2)
Goals by Season
从进球数来看，其反映的情况与比赛结果非常类似。在过去12个赛季，平均每个赛季主队的进球数为586球，场均1.54球；而客队仅为435球，场均1.14球。没有一个赛季客队进球数超过了主队进球数，客队很难得分。在09-10赛季主队的进球数达到了最高值，并且当年主队的获胜数也是历年来的最高，达到了惊人的193场。06-07赛季对于客队来说是个艰难的赛季，它们只有379个进球，场均不到1个进球。
3.1.3 比分分布
results%>%
mutate(home=as.numeric(.$home_goals),away=as.numeric(.$away_goals),
yep=paste(as.character(.$home_goals),":",as.character(.$away_goals),seq=""))%>%
group_by(yep,home,away)%>%
summarise(count=n())%>%
ggplot(aes(x=home,y=away,size=count),position="jitter")+
scale_fill_gradientn(colours = brewer.pal(10,'Spectral')) +
mytheme+
theme(axis.text.x = element_text(angle = 0,size=30),axis.text.y = element_text(size=30))+
geom_point(shape=21,colour="black",fill="cornsilk")+
scale_size_area(max_size = 50)+
geom_text(aes(y=away,x=home,label=count),vjust=-1.8,size=15,colour="grey60")+
scale_x_continuous(breaks=seq(0,10,1))+
scale_y_continuous(breaks=seq(0,10,1))+
xlab("Home Goals")+
ylab("Away Goals")+
ggtitle("Score Distribution")
Score Distribution
从比分的分布来看，1:1是最常见的比分，这体现了小俱乐部的生存之道。“小俱乐部”的目标通常是“保平争胜”：尽力防守，当比赛还剩10分钟左右时，再做进攻的尝试，以谋求更好的结果。另外主队一般以1:0、2:1和2:0的比分拿下比赛。
3.2 球队
以球队为单位分析，可以得到一些更加细致直观的结果。
3.2.1 参赛次数
number_fights <- stats%>%
group_by(team)%>%
count()
number_fights%>%
ggplot() +
ylab("Number of Fights")+
xlab("")+
ylim(0,12)+
geom_bar(aes(y =n, x = reorder(team,n), fill =n), stat="identity") +
mytheme+
scale_fill_gradientn(colours = brewer.pal(10,'Spectral')) +
theme(axis.text.x = element_text(angle = 0)) +
geom_text(aes(x=team,y=n,label=n),hjust=-0.5,size=10.5)+
ggtitle('Number of Fights')+
coord_flip()
Number of Fights
过去12个赛季，只有7支队伍一直留在英超，分别是英超赫赫有名的“big6”加上利物浦在默西赛德郡的宿敌埃弗顿队。其中有七支球队都只在英超停留一个赛季，可见每年保级队伍的竞争激烈程度并不亚于冠军的争夺。
3.2.2 球队获胜数
#主场获胜数
home_win <- results %>%
filter(result == "H") %>%
group_by(home_team, result) %>%
summarise(count=n()) %>%
arrange(desc(count)) %>%
ggplot(aes(x=reorder(factor(home_team), count), y=count, fill = count)) +
geom_bar(stat = "identity", position = "dodge") +
xlab("") +
ylab("") +
ylim(0,200)+
mytheme+
scale_fill_gradientn(name = '',colours = rev(brewer.pal(10,'Spectral'))) +
geom_text(aes(label = count), hjust = -0.3, size = 10.5) +
ggtitle("Teams with the Most Home Wins \nin the English Premier League from \n2006/07 to 2017/18") +
coord_flip()
#客场获胜数
away_win = results %>%
filter(result == "A") %>%
group_by(away_team, result) %>%
summarise(count=n()) %>%
arrange(desc(count)) %>%
ggplot(aes(x=reorder(factor(away_team), count), y=count, fill = count)) +
geom_bar(stat = "identity", position = "dodge") +
xlab("") +
ylab("") +
ylim(0,130)+
mytheme+
scale_fill_gradientn(name = '',colours = rev(brewer.pal(10,'Spectral'))) +
geom_text(aes(label = count), hjust = -0.3, size = 10.5) +
ggtitle("Teams with the Most away Wins \nin the English Premier League from \n2006/07 to 2017/18")+ coord_flip()
grid.arrange(home_win, away_win, ncol = 2, nrow = 1)
Team Wins
#主客场获胜总数以及主客场获胜之差
home_wins <- results %>%
filter(result == "H") %>%
group_by(home_team, result) %>%
summarise(count=n()) %>%
arrange(desc(count))
away_wins <- results %>%
filter(result == "A") %>%
group_by(away_team, result) %>%
summarise(count=n()) %>%
arrange(desc(count))
home_wins <- rename(home_wins, "Team" = "home_team")
away_wins <- rename(away_wins, "Team" = "away_team")
wins <- merge(home_wins, away_wins, by="Team")
wins$overall_wins = wins$count.x+wins$count.y #直接添加新的列
wins$difference_wins = wins$count.x-wins$count.y
##胜场和
overall_wins <- wins%>%
ggplot(aes(x=reorder(factor(Team), overall_wins), y=overall_wins, fill = overall_wins)) +
geom_bar(stat = "identity", position = "dodge") +
xlab("Team") +
ylab("Wins") +
ylim(0,320)+
mytheme+
theme(axis.title = element_blank())+
scale_fill_gradientn(colours = rev(brewer.pal(10,'Spectral'))) +
geom_text(aes(label = overall_wins), hjust = -0.3, size = 3.5) +
ggtitle("Number of Overall Wins in the English Premier League \nfrom 2006/07 to 2017/18") +
coord_flip()
#主客场胜场差
difference_wins <-wins%>%
ggplot(aes(x=reorder(factor(Team), difference_wins), y=difference_wins, fill = difference_wins)) +
geom_bar(stat = "identity", position = "dodge") +
xlab("Team") +
ylab("Wins") +
ylim(0,65)+
mytheme+
theme(axis.title = element_blank())+
scale_fill_gradientn(colours = rev(brewer.pal(10,'Spectral'))) +
geom_text(aes(label = difference_wins), hjust = -0.3, size = 3.5) +
ggtitle("Number of Difference Wins in the English Premier League \nfrom 2006/07 to 2017/18") +
coord_flip()
grid.arrange(overall_wins, difference_wins, ncol = 2, nrow = 1)
Team Wins
从各支球队的主客场战绩来看，“梦剧场”是英超最著名的魔鬼主场，曼联也是英超最成功的俱乐部。“红魔”对于在主场获胜颇有心得，过去12个赛季，曼联赢下了228场主场比赛中的171场，胜率高达75%。同时曼联也是近12年来客场胜率最高的球队，但63.6%的胜率远低于其主场胜率。即使是曼联，在客场拿下比赛也并非信手拈来。
切尔西懂得在客场取胜的艺术，他们擅长在客场“降低身段”，让出控球权，专注防守反击。最经典的“切尔西式胜利”莫过于2012年欧冠半决赛：“蓝军”在诺坎普面对“梦三”巴萨上演10人翻盘的神迹，切尔西全场顽强防守。在落后两球的情况下抓住了全场仅有的两次机会，凭借拉米雷斯的吊射以及替补出场担任左后卫的托雷斯在补时阶段的进球，送巴萨出局。切尔西的铁血与顽强是他们能够在客场比赛中取得胜利的重要原因。另外尽管阿森纳在17-18赛季的下半阶段的客场比赛中未尝一胜，他们也以103场胜利排在第三位。
埃弗顿和曼城是主客场表现差异最大的球队，他们的主场表现要远好于客场。
3.2.3 球队成绩
stat <- stats%>%
filter(rank %in% c(1,2,3,4))
stat <- stat[,c(1,42,44)]
stat <- stat%>%
spread(key=rank,value=team)
colnames(stat)[2:5] <- c("first","second","third","fourth")
stat 在过去12个赛季，曼联5次捧杯，曼城和切尔西各3次。另外表中只出现了七支球队，2015/16赛季创造奇迹的莱斯特城以及英超“big6”，可见每年的冠军的争夺也便是围绕“big6”展开。 值得注意的是，在2013/14赛季前，曼联从未跌出前二，不过之后四强便难觅“红魔”的身影。2013/14赛季也是弗格森爵士离开的第一个赛季，从此曼联走向衰落。可见2008/09赛季C罗的离开并未对曼联造成巨大影响，弗格森爵士才是曼联的灵魂。
尽管阿森纳在2003/04赛季不败夺冠后，再未染指英超冠军。但是他们的确是英超最“稳定”的球队，你总能在第三或者第四的位置上见到他们。不可否认，教授（温格）居功至伟，在每年几乎都有核心球员出走的悲哀下，在管理层赚的盆满钵满却不愿加大投入的无奈下，在金元足球的攻势下，依然奉行“美丽足球”的执念。
切尔西或许是英超“第二成功”的球队，曼城则是“金元足球”的代表，热刺也是阿森纳在北伦敦最棘手的对手。利物浦虽然也从未获得英超冠军，但其欧冠战绩是英超最优秀的。
4 Big6
“big6”的之间的比赛是英超最值得关注的比赛，六支队伍特点鲜明，没有哪支队伍的实力完全高于其他队伍。分析六支球队的各项指标，可以得到其战术风格以及不足之处等。
4.1 进攻
4.1.1 进球
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = goals, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text(size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
xlab("")+
ylab("")+
ggtitle('Boxplot of the Goals of Big6')
Goals Analysis of Big6
除热刺的进球略少而外，“Big6”每个赛季的进球数较为接近，曼联是进球最多的球队，曼城紧随其后。阿森纳的进球数最为稳定，而曼城的进球数分布则最为不均，这是曼城近些年的崛起所致。
4.1.2 射门
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = total_scoring_att, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text(size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
xlab("")+
ylab("")+
ggtitle('Boxplot of the Goals of Big6')
Scoring Attempts Analysis of Big6
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = round(ontarget_scoring_att/total_scoring_att,2), fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text(size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
xlab("")+
ylab("")+
ggtitle('Boxplot of the Scoring Attempts Accuracy of Big6')
Scoring Attempts Accuracy Analysis of Big6
利物浦每赛季射门次数最多，但射门准确率（射中门框范围内）最低，仅为33%。不过2013/14赛季其射门准确率显著地高于其他赛季，当时也是利物浦最接近冠军的一次。苏亚雷斯的神级发挥带领“红军”顽强前进，但杰拉德的回传失误让利物浦梦碎安菲尔德。阿森纳拥有最高的射门准确率，其快速的传切配合能够渗透对手防线，将球送入禁区，让前锋有直面球门的机会，但禁区内空间较为狭窄，使射门遭到封堵的几率上升。
4.1.3 头球
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = att_hd_goal, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text(size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
xlab("")+
ylab("")+
ggtitle('Boxplot of the Head Goals of Big6')
Scoring Head Goals Analysis of Big6
切尔西英超六强中最擅长头球的球队，从2006年到2012年，他们有擅长头球的“魔兽”德罗巴。后者的头球给切尔西带来了不少冠军，分量最重的一颗头球是2011/12赛季欧冠决赛对阵拜仁时打进的扳平头球。而上赛季莫拉塔加盟后，他是五大联赛取得头球第二多的球员。曼城是最不依赖头球得分但球队，其每赛季只有不到百分之十的进球来自头球。
4.1.4 任意球
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = att_freekick_goal, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text(size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
xlab("")+
ylab("")+
ggtitle('Boxplot of the Freekick Goals of Big6')
Freekick Goals Analysis of Big6
任意球是比赛中最具观赏性也是难度最高的进球方式，英超六强任意球进球非常少，每赛季2到3粒。因为英超相比于其他联赛，更加注重身体对抗，技术型球员并不受用。莫拉塔在尤文图斯、皇马的表现与在切尔西的表现大相径庭，重要原因就是他在英超的身体对抗中不占优势，难以护住球；另一方面许多英超的任意球大师都已经离开，比如C罗、亨利和贝克汉姆等。在曼联最巅峰的2007/08赛季，其有6粒任意球进球，是近十二年来的英超记录。
4.1.5 快攻进球
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = goal_fastbreak, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text(size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
xlab("")+
ylab("")+
ggtitle('Boxplot of the Fastbreak Goals of Big6')
Fastbreak Goals of Big6
虽然利物浦平均每赛季的快攻进球并不多，但在上赛季其快攻进球数已经跃居第一。特洛普崇尚全攻全守的打法，全场逼抢，然后发动快攻。另外萨拉赫和马内的速度非常快，爆点能力强，这使得快攻的成功率大大提高。
4.1.6 直塞
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = total_through_ball, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text(size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
xlab("")+
ylab("")+
ggtitle('Boxplot of the Through Balls of Big6')
Through Balls of Big6
阿森纳的进攻流畅，打法华丽，经常通过送出直塞洞穿对手防线。核心厄齐尔也是世界上最会传球的球员之一。虽然温格的时代已经结束，但“美丽足球”依旧。
4.1.7 错失机会
big6 %>%
filter(big_chance_missed!=0)%>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = big_chance_missed, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text(size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
xlab("")+
ylab("")+
ggtitle('Boxplot of the Big Chance Missed of Big6')
Big Chance Missed of Big6
错失机会需要辩证看待：球员把握机会的不强是一部分原因；对于“Big6”来说更可能导原因是球队创造机会的能力很强，因此错失的机会也就相对越多。曼城应该是近几个赛季创造机会能力最强的球队，上赛季夺冠的曼城错失了63次机会，场均1.66次。阿圭罗善于抓住机会终结对手，而斯特林则更“快乐”。
4.2 防守
4.2.1 失球
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = goals_conceded, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text(size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
xlab("")+
ylab("")+
ggtitle('Boxplot of the Goals Conceded of Big6')
Freekick Goals Conceded of Big6
曼联平均每个赛季只有32.17次失球，场均丢球0.85个。失球第二少的是切尔西，“防守反击”战术在当今攻势足球的氛围下仍不过时。热刺虽然在进攻端表现良好，但防守问题对他们而言仍是严峻的考验。阿森纳从2014年开始排名随着丢球数一起上升。另外夺冠的球队可能不是最擅长进攻的球队，但是一定是最擅长防守的球队。
4.2.2 零封对手
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = clean_sheet, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text(size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
xlab("")+
ylab("")+
ggtitle('Boxplot of the Clean Sheets of Big6')
Freekick Clean Sheets of Big6
零封对手与失球反映出的问题相同。曼联和切尔西是近些年来英超六强中防守最好的球队，曼联平均每赛季有45%的比赛零封了对手。不过曼联在2012年到2015年失去了他们在防守上的统治力，那也是他们最艰难的时期，穆里尼奥入主曼联后，着重防守，上赛季曼联是零封对手次数最多的球队，他们也时隔4年重回积分榜前三。
4.2.3 拦截传球
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = interception, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
xlab("") +
ylab("")+
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text( size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
ggtitle('Boxplot of the Interception of Big6')
Interception Analysis of Big6
拦截传球考验球员的意识以及对球路的精准判断，阿森纳平均每个赛季能够拦截615次传球，切尔西仅仅只有497.58次。不过在2016年引进坎特后，“蓝军”的拦截传球数在英超六强中已经排进前三。在欧洲团队型足球崛起的情况下，防守型后腰对于顶尖球队非常重要，它让球队的攻防更加均衡。一名优秀的防守型后腰需要频繁的跑动，出色的大局观以及精准的出球。莱斯特城2015/16赛季夺冠时，坎特在出场的37场比赛中完成了163次抢断，是当赛季的英超抢断王。皇马曾经卖掉马克莱莱后便陷入长期的低迷，巴萨的成功也离不开布斯克茨。当今曼城的弱点之一便是后腰费尔南迪尼奥老化，而无人顶替。
4.2.4 铲球
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = total_tackle, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
xlab("") +
ylab("")+
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text( size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
ggtitle('Boxplot of the Totla Tackle of Big6')
Tackle Analysis of Big6
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = last_man_tackle, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
xlab("") +
ylab("")+
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text( size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
ggtitle('Boxplot of the Last Man Tackle of Big6')
Tackle Analysis of Big6
AC米兰传奇后卫保罗·马尔蒂尼曾说：如果我必须铲球，那么我已经犯了错误。利物浦频繁使用铲球来打乱对手的节奏，他们平均每个赛季的铲球次数高达853次，场均22.45次，远高于其他球队。铲球通常作为战术策略在迫不得已时才使用，因为犯规的几率非常高。英超六强的铲球数目逐年减少，利物浦也不例外，球员倾向于用更安全的办法来解决问题。
最后一人放铲是一个非常有趣的数据，如果你必须作为最后一个人来放铲，说明己方球队的攻防转换出现了漏洞。阿森纳的打法需要边后卫频繁参与进攻。这导致当对手反击时，若边后卫回追不及时，己方球门便危在旦夕。另外“门卫”的出现让最后一人放铲的数目逐年减少，现代守门员弃门出击化险为夷的场景更加多见，因为越靠近球门，被破门的风险也就越高。这类门将的代表人物是德国门将诺伊尔，他重新定义了门将的职责。
4.2.5 封堵
big6 %>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = outfielder_block, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
xlab("") +
ylab("")+
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 40),axis.title.x = element_text(size=50),axis.text.x = element_text( size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
ggtitle('Boxplot of the Outfielder Block of Big6')
Outfielder Block Analysis of Big6
除守门员外的球员封堵射门数的产生有两个的原因：首先与被射门的次数有关，如上赛季的曼城只有53次封堵，说明对手难以威胁到“蓝月亮”的球门；防守球员合适的占位和迅速的反应也不无关系，虽然曼联在2012-2013赛季有143次封堵记录，但仍是冠军。后卫总能出现在正确的位置，化险为夷。
4.2.6 扑救
big6 %>%
filter(saves!=0)%>%
group_by(team) %>%
ggplot(aes(x = factor(team), y = saves, fill = team)) +
geom_boxplot(colours = rev(brewer.pal(10,'Set1'))) +
stat_summary(fun.y=mean, geom="point", shape=20, size=15, color="red", fill="red") +
xlab("") +
ylab("")+
theme(legend.position ='none',axis.title.y = element_text(size=50), axis.text.y = element_text(size = 25),axis.title.x = element_text(size=50),axis.text.x = element_text( size = 45), plot.title = element_text(size=50, hjust = 0.5)) +
ggtitle('Boxplot of the Saves of Big6')
Saves Analysis of Big6
守门员做出扑救说明防守出现漏洞，使对方有得分良机。阿森纳平均每个赛季的扑救次数高达110次。另外大卫·德赫亚上赛季完成了118次扑救，场均3.11次，他的存在一定程度上掩盖了曼联在防守上的不足。
4.3 球队特点
根据以上分析，结合六支队伍近期的表现，可以总结出“Big6”各自有以下特点：5 聚焦德比
德比之战往往是最激动人心的比赛，它的背后往往蕴含着深厚的历史渊源。英超最著名的德比莫过于北伦敦德比、曼市德比、双红会和默西塞德郡德比。
5.1 北伦敦德比
ars <- stats%>%
filter(team=="Arsenal")
ars <- ars[43]
ars<- as.vector(unlist(ars[1]))
tot<- stats%>%
filter(team=="Tottenham Hotspur")
tot <- tot[43]
tot <- as.vector(unlist(tot[1]))
class <- stats%>%
filter(team=="Tottenham Hotspur")
class <- class[42]
class <- as.vector( unlist(class[1]))
highchart() %>%
hc_xAxis(
list(
categories = class,
reversed = FALSE,
labels = list(step = 1)),
list(
opposite = TRUE,
categories = class,
reversed = FALSE,
linkedTo = 0,
labels = list(step = 1))
) %>%
hc_plotOptions(series = list(
borderWidth = 0))%>%
hc_yAxis(
labels = list(formatter = JS("function () { return (Math.abs(this.value) ) ;}")),
min = -100,max = 100)%>%
hc_tooltip(
formatter = JS("function () {
return '<b>' + this.series.name + this.point.category + '</b><br/>' +
'grade: ' + Highcharts.numberFormat(Math.abs(this.point.y), 0);}"))%>%
hc_title(text = "Arsenal VS. Tottenham Hotspur",align="center")%>%
hc_subtitle(text="by Grade of Each Season",align="left")%>%
hc_plotOptions(series= list(
stacking = "normal")) %>%
hc_add_series(name = "Arsenal",
data = -ars,
type = "bar") %>%
hc_add_series(name = "Tottenham Hotspur",data = tot,type = "bar") %>%
hc_add_theme(hc_theme_538())Arsenal VS.Tottenham Hotspur by Grade of Each Season
温格和他的球队在很长一段时间的表现都好于他们的北伦敦对手。不过近些年热刺展现出了足以争夺冠军的实力，而阿森纳却停滞不前。
tot_ars <- results%>%
filter(.$home_team%in%c("Tottenham Hotspur","Arsenal")&.$away_team%in%c("Tottenham Hotspur","Arsenal"))%>%
mutate(times=seq(1,24))
ars_home <- tot_ars%>%
filter(home_team == "Arsenal") %>%
rename(team = home_team) %>%
mutate(results = (if_else (result == "A", "ARS Loss", if_else(result == "D", "Draw", "ARS Win"))))
ars_home <- ars_home[,c(1,6,8)]
ars_away <- tot_ars %>%
filter(away_team == "Arsenal") %>%
rename(team = away_team) %>%
mutate(results = (if_else (result == "H", "ARS Loss", if_else(result == "D", "Draw", "ARS Win"))))
ars_away <- ars_away[,c(2,6,8)]
colnames(ars_home) <- colnames(ars_away)
ars_wdl= rbind(ars_home, ars_away)
ars_wdl%>%
ggplot()+
geom_bar(aes(x=season,fill=results))+
scale_fill_brewer(name="",palette='Set1') +
mytheme+
ylab("")+
xlab("")+
theme(legend.position = "bottom",axis.text.y=element_blank())+
labs(title="Arsenal VS. Tottenham Hotspur",subtitle="by Direct Confontation")+
theme(legend.position ='bottom',legend.text=element_text(size=25), axis.text.y = element_text(size = 25),axis.text.x = element_blank(), plot.title = element_text(size=50), plot.subtitle = element_text(size = 25),legend.key.size = unit(4,"cm"),axis.ticks = element_blank())+
coord_flip()
Arsenal VS.Tottenham Hotspur by Direct Confontation
从两队的直接交锋来看，阿森纳9胜6负8平，占据上风。但是近四年阿森纳只战胜过热刺一次。北伦敦的天空正在由红变白。但是新赛季埃梅里的工作做得很出色，枪手在主场4比2赢下了北伦敦德比，守住了一片红色。
5.2 曼市德比
mun <- stats%>%
filter(team=="Manchester United")
mun<- mun[43]
mun <- as.vector(unlist(mun[1]))
mci<- stats%>%
filter(team=="Manchester City")
mci <- mci[43]
mci <- as.vector(unlist(mci[1]))
highchart() %>%
hc_xAxis(
list(
categories = class,
reversed = FALSE,
labels = list(step = 1)),
list(
opposite = TRUE,
categories = class,
reversed = FALSE,
linkedTo = 0,
labels = list(step = 1))
) %>%
hc_plotOptions(series = list(
borderWidth = 0))%>%
hc_yAxis(
labels = list(formatter = JS("function () { return (Math.abs(this.value) ) ;}")),
min = -100,max = 100)%>%
hc_tooltip(
formatter = JS("function () {
return '<b>' + this.series.name + this.point.category + '</b><br/>' +
'grade: ' + Highcharts.numberFormat(Math.abs(this.point.y), 0);}"))%>%
hc_title(text = "Manchester City VS. Manchester United",align="center")%>%
hc_subtitle(text="by Grade of Each Season",align="left")%>%
hc_plotOptions(series= list(
stacking = "normal")) %>%
hc_add_series(name = "Manchester United",
data = -mun,
type = "bar") %>%
hc_add_series(name = "Manchester City",data = mci,type = "bar") %>%
hc_add_theme(hc_theme_538())Manchester United VS. Manchester City by Grade of Each Season
2012/13赛季后，曼联逐渐衰落，曼城成为曼彻斯特的主角。近些年来，曼城已经在转会市场净投入了6亿4000万欧元，取得了不错的效果。虽然曼联也花费了5亿6000万欧用于引援，但收效甚微。曼联现在缺的是“红魔”的灵魂，但曼联的底蕴能够让其重回正轨，再次统治英超可能只是时间问题。
mun_mci <- results%>%
filter(.$home_team %in% c("Manchester United","Manchester City")&.$away_team %in% c("Manchester United","Manchester City"))%>%
mutate(times=seq(1,24))
mun_home <- mun_mci %>%
filter(home_team == "Manchester United") %>%
rename(team = home_team) %>%
mutate(results = (if_else (result == "H", "MUN Win", if_else(result == "D", "Draw", "MUN Loss"))))
mun_home <- mun_home[,c(1,6,8)]
mun_away <- mun_mci%>%
filter(away_team == "Manchester United") %>%
rename(team = away_team) %>%
mutate(results = (if_else (result == "A", "MUN Win", if_else(result == "D", "Draw", "MUN Loss"))))
mun_away <- mun_away[,c(2,6,8)]
colnames(mun_home) <- colnames(mun_away)
mun_wdl <- rbind(mun_home,mun_away)
mun_wdl%>%
ggplot()+
geom_bar(aes(x=season,fill=results))+
scale_fill_brewer(name="",palette='Set1') +
mytheme+
ylab("")+
xlab("")+
theme(legend.position = "bottom",axis.text.y=element_blank())+
labs(title="Manchester United VS. Manchester City",subtitle="by Direct Confontation")+
theme(legend.position ='bottom',legend.text=element_text(size=25), axis.text.y = element_text(size = 25),axis.text.x = element_blank(), plot.title = element_text(size=50), plot.subtitle = element_text(size = 25),legend.key.size = unit(4,"cm"),axis.ticks = element_blank())+
coord_flip()
Manchester City VS. Manchester United by Direct Confontation
从两队的直接交锋来看，曼联取得11胜10负3平的战绩。在曼联夺冠的2007/08赛季，曼城竟然双杀了曼联。另外平局的场次非常少，双方都想拿下对手，并不想采取保平战术。 “曼市德比”从不缺乏经典镜头：英超历史上最佳进球就出自2010/11赛季鲁尼在对阵曼城时的倒钩进球，这也是公认的英超20年最佳时刻。在2011/12赛季，冠军的悬念被保留到了最后一轮，曼城主场对阵QPR，曼城在补时阶段最后三分钟连入两球，以净胜球优势力压曼联，夺得队史首个英超冠军。曼市双雄给我们留下了许多经典，我相信以后的曼市德比也会更加精彩。
5.3 双红会
liv<- stats%>%
filter(team=="Liverpool")
liv <- liv[43]
liv <- as.vector(unlist(liv[1]))
highchart() %>%
hc_xAxis(
list(
categories = class,
reversed = FALSE,
labels = list(step = 1)),
list(
opposite = TRUE,
categories = class,
reversed = FALSE,
linkedTo = 0,
labels = list(step = 1))
) %>%
hc_plotOptions(series = list(
borderWidth = 0))%>%
hc_yAxis(
labels = list(formatter = JS("function () { return (Math.abs(this.value) ) ;}")),
min = -100,max = 100)%>%
hc_tooltip(
formatter = JS("function () {
return '<b>' + this.series.name + ', age ' + this.point.category + '</b><br/>' +
'grade: ' + Highcharts.numberFormat(Math.abs(this.point.y), 0);}"))%>%
hc_title(text = "Manchester United VS. Liverpool",align="center")%>%
hc_subtitle(text="by Grade of Each Season",align="left")%>%
hc_plotOptions(series= list(
stacking = "normal")) %>%
hc_add_series(name = "Manchester United",
data = -mun,
type = "bar") %>%
hc_add_series(name = "Liverpool",data = liv,type = "bar") %>%
hc_add_theme(hc_theme_538())Manchester United VS. Liverpool by Grade of Each Season
mun_liv <- results%>%
filter(.$home_team%in%c("Manchester United","Liverpool")&.$away_team%in%c("Manchester United","Liverpool"))%>%
mutate(times=seq(1,24))
mun_home <- mun_liv %>%
filter(home_team == "Manchester United") %>%
rename(team = home_team) %>%
mutate(results = (if_else (result == "H", "MUN Win", if_else(result == "D", "Draw", "MUN Loss"))))
mun_home <- mun_home[,c(1,6,8)]
mun_away <- mun_liv%>%
filter(away_team == "Manchester United") %>%
rename(team = away_team) %>%
mutate(results = (if_else (result == "A", "MUN Win", if_else(result == "D", "Draw", "MUN Loss"))))
mun_away <- mun_away[,c(2,6,8)]
colnames(mun_home) <- colnames(mun_away)
red_wdl= rbind(mun_home,mun_away)
red_wdl%>%
ggplot()+
geom_bar(aes(x=season,fill=results))+
scale_fill_brewer(name="",palette='Set1') +
mytheme+
ylab("")+
xlab("")+
theme(legend.position = "bottom",axis.text.y=element_blank())+
labs(title="Liverpool VS. Manchester United",subtitle="by Direct Confontation")+
theme(legend.position ='bottom',legend.text=element_text(size=25), axis.text.y = element_text(size = 25),axis.text.x = element_blank(), plot.title = element_text(size=50), plot.subtitle = element_text(size = 25),legend.key.size = unit(4,"cm"),axis.ticks = element_blank())+
coord_flip()
Liverpool VS. Manchester United by Direct Confontation
不管是直接对话还是联赛战绩，曼联无疑完胜利物浦。利物浦仅在2013/14赛季和2016/17赛季的联赛排名要高于曼联，并且从未获得过英超冠军。曼联可谓是利物浦最大的“苦主”，“红魔”在两者的交战中获得了14胜6平4负的傲人战绩，利物浦更是自2014年再也没赢过曼联。不过新赛季利物浦在主场三比一战胜曼联，现在的“双红会”变得更加有趣。
6 莱斯特奇迹
在英超历史上，只有六只队伍获得过冠军，其中有两支名不见经传的球队：1994/95赛季的布莱克本和2015/16赛季的莱斯特城。莱斯特城的夺冠堪称足球史上的奇迹，因为在夺冠的前一个赛季他们有140天都排在英超倒数第一，在最后的9场比赛中获得了7场胜利，才脱离降级区。莱斯特城的成功值得想取得成绩却实力不足的“小俱乐部”学习。 首先利用Lasso回归对球队的积分进行预测，得到各变量的重要程度如下表：
coef_name <- c("goals","goals_conceded","clean_sheet","att_obox_goal","scoring_rate","outfielder_block","total_pass")
coef_value <- c(9.85170122,-5.81506382,2.15200141,0.27687161,0.27620690,-0.09051668,0.05642033)
coef_111 <- cbind(coef_name,coef_value )
coef_111 <- as.data.frame(coef_111)
colnames(coef_111)=c("变量","系数")
coef_111可以看到影响球队积分的最关键变量分别为：进球数、失球数、零封次数、禁区外进球数、射门准确率、除守门员外封堵以及传球数。
y_pre_train <- predict(model.lasso,x,s=n.Cp)
err_train <- (y-y_pre_train$fit)^2
err_train <- as.data.frame(err_train)%>%
mutate(n=seq(1,240))highchart() %>%
hc_xAxis(categories =err_train$n) %>%
hc_add_series(name = "square of dev", data =err_train$grade)%>%
hc_annotations(
list(labelOptions = list(backgroundColor = "rgba(255,255,255,0.5)",verticalAlign = "bottom", y = -5,x=20 ),
labels = list(list(point = list(x = 180, y = 106,xAxis = 0, yAxis = 0 ), text = "LEI 2015/16 dev=103.91"))))6.1 净胜球分析
champion <- stats%>%
filter(rank==1)
season_2015 <- stats%>%
filter(season=="2015-2016",rank==seq(1,10))
highchart() %>%
hc_xAxis(categories = season_2015$team)%>%
hc_add_series(name = "goals", data = season_2015$goals,type = "column") %>%
hc_add_series(name = "goals conceded", data = season_2015$goals_conceded,type = "column")%>%
hc_add_series(name = "goals difference", data = season_2015$goals-season_2015$goals_conceded,type = "column")Goals Conceded of Top10 Team in Season 2015-2016
“狐狸城”的进球数排在前十名队伍的第三位，失球数则排在第八位，净胜球个数为32个，表现良好。但是热刺端在进球与失球上的表现要优于莱斯特城，阿森纳和曼城也与其不相上下。
champion_goal <- champion%>%
arrange(desc(goals-goals_conceded))
highchart() %>%
hc_xAxis(categories = champion$team)%>%
hc_add_series(name = "goals", data = champion_goal$goals,type = "column") %>%
hc_add_series(name = "goals conceded", data = champion_goal$goals_conceded,type = "column")%>%
hc_add_series(name = "goals difference", data = champion_goal$goals-champion_goal$goals_conceded,type = "column")Goals and Goals Conceded of Each Champion
就各届冠军的情况来看，莱斯特城的进球较少，失球数也仅低于2010/11赛季和2012/13赛季的曼联的37球和42球。净胜球则排在第八位，另外在21年的所有球队的净胜球排名中，“狐狸城”仅排在第37位。可见就积分对影响最大的因素方面，莱斯特城的数据并不足以让其染指冠军。
6.2 “big6”集体迷失
avg_grade <- stats%>%
filter(team %in% season_2015$team)%>%
group_by(team)%>%
summarise(avg_grade=round(mean(grade)))
season_2015 <- season_2015%>%
left_join(avg_grade,"team")
highchart() %>%
hc_xAxis(categories = season_2015$team)%>%
hc_add_series(name="grade",data = season_2015$grade,type = "column")%>%
hc_add_series(name="average grade",data = season_2015$avg_grade,type = "spline")%>%
hc_plotOptions(column= list(marker = list(enabled =FALSE)))%>%
hc_title(text= "Grade of Top10 in Season 2015/16", align='center')%>%
hc_plotOptions(column = list(colorByPoint = TRUE, colors = c('#058DC7', '#50B432', '#ED561B', '#DDDF00','#24CBE5', '#64E572', '#FF9655', '#FFF263', '#6AF9C4')))%>%
hc_legend(enable=FALSE)Grades of Top 10 in Season 2015/16
但当赛季的莱斯特城非常幸运，英超“big6”似乎集体迷失，这一点给他们的夺冠创造了千载难逢的外部条件。虽然“Big6”的积分仍然处在前十，但除热刺外，其余五队的积分都低于其平均积分。曼联、切尔西和利物浦由于自身原因，取得了近十二年来的最差战绩：切尔西因内讧摆烂，曼联在赛季中后期将重点放在欧联杯，利物浦则还处于阵痛期，曼城的战绩也近五年来最差的。仅阿森纳和热刺发挥相对稳定，但两者都未抓住这个机会。除去这六只球队，剩下的四支球队的积分都高于其平均积分，并且他们并不是前十、甚至不是英超的常客。
highchart() %>%
hc_xAxis(categories = champion$team)%>%
hc_add_series(name = "grade", data = champion$grade,type = "column")%>%
hc_plotOptions(column= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Grade of Each Champion", align='center')%>%
hc_plotOptions(column = list(colorByPoint = TRUE, colors = c('#058DC7', '#50B432', '#ED561B', '#DDDF00','#24CBE5', '#64E572', '#FF9655', '#FFF263', '#6AF9C4')))%>%
hc_legend(enable=FALSE)Grade of Each Champion
从历年冠军的积分情况来看，冠军的积分一般在90分左右。莱斯特城的81积分只能排在倒数第二，场均得分率为2.13。可见英超强队的集体迷失在一定程度上成就了“莱斯特城奇迹”。但这并不是最重要的原因，因为莱斯特在上赛季还是一支保级队伍，并且其他发挥良好的队伍也未能从莱斯特手中夺走冠军。
6.3 高效进攻
highchart() %>%
hc_xAxis(categories =champion$team) %>%
hc_add_series(name = "scoring attempts",
data = round(champion$total_scoring_att/38,2)) %>%
hc_plotOptions(line= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Scoring Attempts of Each Champion", align='center')Scoring Attempts of Each Champion
highchart() %>%
hc_xAxis(categories =season_2015$team) %>%
hc_add_series(name = "scoring attempts",
data = round(season_2015$total_scoring_att/38,2)) %>%
hc_plotOptions(line= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Scoring Attempts of Top10 in Season 2015/16", align='center')Scoring Attempts of Top10 in Season 2015/16
莱斯特是历年冠军中射门数最少的，场均只有13.73次，在2015/16赛季也仅仅排在第七位。
highchart() %>%
hc_xAxis(categories =season_2015$ team) %>%
hc_add_series(name = "goal rate",
data = round(season_2015$goals/season_2015$total_scoring_att,2)) %>%
hc_plotOptions(line= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Goal Rate of Top10 in Season 2015/16", align='center')Goal Rate of Top10 in Season 2015/16
但其进球转化率达到了0.13，排在英超第一。马赫雷斯以及瓦尔迪展现了超强的把握机会能力以及终结能力。高效是“狐狸城”领跑英超的关键因素。
6.4 1:0主义
尽管“狐狸城”的进球转化率很高，但也仅比其他球队高0.01到0.02，差距微乎其微。但是其将1:0主义发挥到极致。
results%>%
filter(season=="2015-2016"&((home_team=="Leicester City"&result=="H")|(away_team=="Leicester City")&result=="A"))%>%
mutate(goal=if_else(home_team=="Leicester City",paste(home_goals,":",away_goals),paste(away_goals,":",home_goals)))%>%
group_by(goal)%>%
summarise(count=n())%>%
mutate(per=count/sum(count)*100)%>%
hchart("pie",hcaes(name="goal",y="per"))%>%
hc_title(text= "Result Distribution of Leicester City", align='center')%>%
hc_plotOptions(pie= list(dataLabels = list(enabled = TRUE,format='<b>{point.goal}</b> {point.per:.2f}%')))Result Distribution of Leicester City
“狐狸城”获胜的比赛中有30%的比赛是以1:0的比分击败对手,17%的比赛以2:1的比分取得胜利，三球及以上获胜的比赛比例仅为43%。可见虽然狐狸城进球数较少，但是能够让其获胜的“有效进球数”的比例非常高。
results%>%
filter(season=="2015-2016"&((home_team=="Tottenham Hotspur"&result=="H")|(away_team=="Tottenham Hotspur")&result=="A"))%>%
mutate(goal=if_else(home_team=="Tottenham Hotspur",paste(home_goals,":",away_goals),paste(away_goals,":",home_goals)))%>%
group_by(goal)%>%
summarise(count=n())%>%
mutate(per=count/sum(count)*100)%>%
hchart("pie",hcaes(name="goal",y="per"))%>%
hc_title(text= "Result Distribution of Tottenham Hotspur", align='center')%>%
hc_plotOptions(pie= list(dataLabels = list(enabled = TRUE,format='<b>{point.goal}</b> {point.per:.2f}%')))Result Distribution of Tottenham Hotspur
而反观进球和失球都优于莱斯特城的热刺队，以1:0的比分战胜对手的比赛仅占15.79%，3球以及上的比例占到了60%。可见高效加实用主义让莱斯特城能够赢得比赛。
6.5 意式防守
莱斯特城的教练拉涅利是一名传统的意大利教练。除了让莱斯特城的进攻变得高效外，也让莱斯特城领悟到“意式防守”的精髓。他们在拦截、封堵和抢断等方面做得非常出色。单赛季36粒丢球，场均丢0.95球。而莱斯特城的其他三个赛季平均丢球高达59个。
highchart() %>%
hc_xAxis(categories =season_2015$ team) %>%
hc_add_series(name = "interception", data = season_2015$interception) %>%
hc_plotOptions(line= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Interception of Top10 in Season 2015/16", align='center')Interception Analysis of Leicester City
highchart() %>%
hc_xAxis(categories =champion$ team) %>%
hc_add_series(name = "interception", data = champion$interception) %>%
hc_plotOptions(line= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Interception of Each Champion", align='center')Interception Analysis of Leicester City
“狐狸城”单赛季拦截次数高达819次，场均21.56次，冠绝英超。在近十二年的冠军中，与其最接近的是2009/10赛季的曼联，但只有705次。值得一提的是，在819次中，坎特贡献了142次。
highchart() %>%
hc_xAxis(categories =champion$ team) %>%
hc_add_series(name = "outfielder_block", data =champion$interception) %>%
hc_plotOptions(line= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Outfielder Block of Each Champion", align='center')Outfielder Block Analysis of Leicester City
highchart() %>%
hc_xAxis(categories =season_2015$ team) %>%
hc_add_series(name = "outfielder block", data =season_2015$interception) %>%
hc_plotOptions(line= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Outfielder Block of of Top10 in Season 2015/16", align='center')Outfielder Block Analysis of Leicester City
封堵情况亦是如此，虽然对手给“狐狸城”的球门造成了不少威胁，但防守队员对于球门的保护很到位。
6.6 放弃传控
由上述分析可知，莱斯特城在拦截、封堵等防守数据上表现优异。其原因除了稳健的防守外，也与莱斯特城的战术密切相关。莱斯特城并不像传统强队那样，热衷于传控。传控需要细腻的脚下技术与团队默契，这点在短时间内对于技术并非优势的莱斯特城球员来说难以做到。所以拉涅利选择让球队让出球权、简化传球，注重防守反击，这样虽然将比赛的主动权交给了对手，但是依靠防守成功获得的反击机会，他们能反客为主。从防守数据上来说，莱斯特城确实做到了这点。
highchart() %>%
hc_xAxis(categories =champion$ team) %>%
hc_add_series(name = "total pass", data =champion$total_pass) %>%
hc_plotOptions(line= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Total Pass of Each Champion", align='center')Total Pass Analysis of Leicester City
highchart() %>%
hc_xAxis(categories =season_2015$team) %>%
hc_add_series(name = "pass success rate",
data =round(season_2015$total_pass/season_2015$touches,2)) %>%
hc_plotOptions(line= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Pass Success Rate of Top10 in Season 2015/16", align='center')Total Pass Analysis of Leicester City
在夺冠的十二支球队中，莱斯特城的13370次传球远低于平均值19717.25次，场均仅351.84脚传球。在如此低的传球频率下，其传球成功率也不甚理想，0.62的传球成功率处在当赛季前十队伍的末尾。可见拉涅利对球队特点把握异常准确，让球队避开其软肋，发挥长处。值得一提的是，即使在赛季初取得不错开局的情况下，球队上下并没有骄傲自满，仍然放低身段，将自身战术贯彻到底。
highchart() %>%
hc_xAxis(categories =season_2015$team)%>%
hc_add_series(name = "dispossessed",
data =season_2015$dispossessed)%>%
hc_plotOptions(line= list(dataLabels = list(enabled =TRUE)))%>%
hc_title(text= "Dispossessed of Top10 in Season 2015/16", align='center')Dispossessed Analysis of Leicester City
让出球权可以减少因传球造成的失误，莱斯特城的丢失球权数仅有366次，整个赛季因丢失球权而被破门的次数只有一次。
“莱斯特奇迹”可谓是天时、地利、人和地完美结合。强队的集体迷失给予其创造奇迹的外部条件，拉涅利制定的适合球队特点的战术加上球员的超常发挥让其化腐朽为神奇。可见在足球世界中，奇迹并非不可能发生，小俱乐部只要战术合理加上外部因素的支持，也能取得较好的成绩。
7 总结
（1）英超的整体情况：主场作战使实力相对较弱的队伍战胜强队的几率加大，增加了比赛的不确定性；另外英超球队之间相对较接近的实力使得竞争异常激烈，在过去12年间，只有七支球队一直留在英超；从球队成绩上来看，每年的前四名也只有七支队伍出现，分别为英超”big6“以及莱斯特城。
（2）”big6“各自的特点非常鲜明：曼联拥有深厚的底蕴与一流的进攻线，但是后防稍弱；阿森纳奉行“美丽足球”，球员技术一流，但身体对抗较差；曼城资金雄厚，阵容扎实，但后腰存在因患，且底蕴稍差，面对落后局面捉襟见肘；利物浦善于逼抢与反击，但前锋射门能力差；切尔西攻守平衡，拥有铁血精神；热刺打法较单一，容易被针对。
（3）从德比上来讲：阿森纳过去在“北伦敦德比”中占据优势，但近几年逐渐被热刺超越；“曼市双雄”你追我赶，难解难分；“双红会”上曼联独占鳌头，但优势也在缩小。总的来说，几大德比正变得更加均衡，结果更加难以预料。
（4）拉涅利合理的战术安排：良好的防守与“1:0主义”。加上核心球员的爆发表现以及英超强队的集体迷失，造就了“莱斯特奇迹”。