English Premier League
aaron chola
Premier-League-Analysis-and-insights
The analysis of the Premier League from 1992 to 2022. This analysis will answer questions like.
1- Does the home stadium ground give any advantage? And if the answer is yes, what’s the quantity for this advantage?
2-what is the best way to collect points, defensive or attacking play?
Introduction
The packages for are analysis are tidyvese and gdata
library(tidyverse)## Warning: package 'tidyverse' was built under R version 4.2.2## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ ggplot2 3.4.0 ✔ purrr 1.0.1
## ✔ tibble 3.1.8 ✔ dplyr 1.1.0
## ✔ tidyr 1.3.0 ✔ stringr 1.5.0
## ✔ readr 2.1.3 ✔ forcats 1.0.0## Warning: package 'ggplot2' was built under R version 4.2.2## Warning: package 'tibble' was built under R version 4.2.2## Warning: package 'tidyr' was built under R version 4.2.2## Warning: package 'readr' was built under R version 4.2.2## Warning: package 'purrr' was built under R version 4.2.2## Warning: package 'dplyr' was built under R version 4.2.2## Warning: package 'stringr' was built under R version 4.2.2## Warning: package 'forcats' was built under R version 4.2.2## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()library(gdata)## gdata: Unable to locate valid perl interpreter
## gdata:
## gdata: read.xls() will be unable to read Excel XLS and XLSX files
## gdata: unless the 'perl=' argument is used to specify the location of a
## gdata: valid perl intrpreter.
## gdata:
## gdata: (To avoid display of this message in the future, please ensure
## gdata: perl is installed and available on the executable search path.)
## gdata: Unable to load perl libaries needed by read.xls()
## gdata: to support 'XLX' (Excel 97-2004) files.
##
## gdata: Unable to load perl libaries needed by read.xls()
## gdata: to support 'XLSX' (Excel 2007+) files.
##
## gdata: Run the function 'installXLSXsupport()'
## gdata: to automatically download and install the perl
## gdata: libaries needed to support Excel XLS and XLSX formats.
##
## Attaching package: 'gdata'
##
## The following objects are masked from 'package:dplyr':
##
## combine, first, last
##
## The following object is masked from 'package:purrr':
##
## keep
##
## The following object is masked from 'package:stats':
##
## nobs
##
## The following object is masked from 'package:utils':
##
## object.size
##
## The following object is masked from 'package:base':
##
## startsWithLoading epl data
library(worldfootballR)## Warning: package 'worldfootballR' was built under R version 4.2.2PL<-fb_match_results(country = "ENG", gender = "M", season_end_year = c(1993:2022), tier = "1st")names(PL)## [1] "Competition_Name" "Gender" "Country" "Season_End_Year"
## [5] "Round" "Wk" "Day" "Date"
## [9] "Time" "Home" "HomeGoals" "Away"
## [13] "AwayGoals" "Attendance" "Venue" "Referee"
## [17] "Notes" "MatchURL" "Home_xG" "Away_xG"PL_r<-PL %>%
arrange(Season_End_Year,Wk)%>%
select(Season_End_Year,Wk,Date,Home,HomeGoals,AwayGoals,Away)PL_r$Date<-as.Date(as.character(PL_r$Date))PL_r$FTR<- case_when(
PL_r$HomeGoals>PL_r$AwayGoals~"H",
PL_r$HomeGoals<PL_r$AwayGoals~"A",
PL_r$HomeGoals==PL_r$AwayGoals~"D"
)Data processing
In this stage we are going to do some descriptive statistics so as to find the summary of the data and then clean the data.
summary(PL_r)## Season_End_Year Wk Date Home
## Min. :1993 Length:11646 Min. :1992-08-15 Length:11646
## 1st Qu.:2000 Class :character 1st Qu.:1999-08-07 Class :character
## Median :2007 Mode :character Median :2007-02-07 Mode :character
## Mean :2007 Mean :2007-03-21
## 3rd Qu.:2015 3rd Qu.:2014-11-29
## Max. :2022 Max. :2022-05-22
## HomeGoals AwayGoals Away FTR
## Min. :0.000 Min. :0.00 Length:11646 Length:11646
## 1st Qu.:1.000 1st Qu.:0.00 Class :character Class :character
## Median :1.000 Median :1.00 Mode :character Mode :character
## Mean :1.521 Mean :1.14
## 3rd Qu.:2.000 3rd Qu.:2.00
## Max. :9.000 Max. :9.00table(PL_r$Season_End_Year)##
## 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
## 462 462 462 380 380 380 380 380 380 380 380 380 380 380 380 380
## 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022
## 380 380 380 380 380 380 380 380 380 380 380 380 380 380As we can see in 1993,1994 and 1995 the number of games that where played are 462, that’s because by then EPL had 22 teams, 20 teams were introduced in 1996.
Analysis Phase
Will now answer the first question.That will try to find out if the home ground favors the home team, and if yes to what extent?
## we will create a new dataframe with percentage of the game result
## (H: for home winning , A: for away team winning, and D for draw)
home_vs_away<-count(PL_r,FTR)%>%
arrange(desc(n))
home_vs_away$percentage<-(home_vs_away$n/sum((home_vs_away$n)))*100table(home_vs_away)## , , percentage = 25.845783960158
##
## n
## FTR 3010 3301 5335
## A 0 0 0
## D 1 0 0
## H 0 0 0
##
## , , percentage = 28.3444959642796
##
## n
## FTR 3010 3301 5335
## A 0 1 0
## D 0 0 0
## H 0 0 0
##
## , , percentage = 45.8097200755624
##
## n
## FTR 3010 3301 5335
## A 0 0 0
## D 0 0 0
## H 0 0 1Plotting the results
ggplot(data = home_vs_away,aes(x=reorder(FTR,-percentage),y=percentage))+
geom_col()+
ggtitle("The percentage of results at Home ground")+
xlab("who wins the Home games")
We can see that the chances of winning a home game are great than that of an away game .Now lets find out the extent at which the home ground favors the home team.
## creating a new data frame to calculate each average points Home and Away matches
home_vs_away_points<-matrix(nrow = 2,ncol = 2)
home_vs_away_points<-as.data.frame(home_vs_away_points)
names(home_vs_away_points)=c("where_to_play","Average_points")
home_vs_away_points$`where_to_play`<-c("Home","Away")
##average points at home matches
home_vs_away_points[1,2]<-((home_vs_away[1,2])*3+(home_vs_away[3,2])*1)/sum(home_vs_away$n)
##average points at away matches
home_vs_away_points[2,2]<-((home_vs_away[2,2])*3+(home_vs_away[3,2])*1)/sum(home_vs_away$n)Plotting the results
ggplot(data = home_vs_away_points,aes(x=where_to_play,y=Average_points))+
geom_col()+
ggtitle("Home Vs Away, average points")+
xlab("where to play")+ylab("Average points")
From the data above we can see that on average collecting points on home is greater than away.The advantage of getting points on a home ground is about 0.53.
Let’s track this advantage through the years.
First, we will go to see how many points the teams have collected on their home ground over the years.
## new dataframe counting final results by every year
point_year<-PL_r%>%
group_by(Season_End_Year)%>%
count(FTR)
## calculating the points that collected in a home and away ground
point_year$points<-case_when(point_year$FTR=="H"~point_year$n*3,
point_year$FTR=="A"~point_year$n*3,
point_year$FTR=="D"~point_year$n*1
)
## creating a new column"h_points" that summation points that collected in home ground either "h_points" to away points
point_year2<-point_year%>%
group_by(Season_End_Year)%>%
summarize(h_points=points[FTR=="H"]+points[FTR=="D"],a_points=(points[FTR=="A"])+points[FTR=="D"])
## tidying the data
point_year3<-point_year2%>%
pivot_longer(c(`h_points`, `a_points`), names_to = "Home_vs_Away", values_to = "Points")ggplot(data = point_year3,aes(x=Season_End_Year,y=Points,col=Home_vs_Away))+
geom_line()+
theme(legend.title=element_blank())+
theme(legend.position=c(0.9,0.9))+
scale_color_manual(labels = c("Away", "Home"),
values = c( "red", "blue"))+
ggtitle("Total points collected by the teams Home vs Away")+
xlab("Season end year")+
ylab("Total points")
As we have seen the home ground advantages the home teams.But when we check 2021 things changed that’s because crowds where not allowed in stadium because of COVID-19.
Now lets see how not attending the stadium changed the trajectory of football.
## new dataframe with only results of 2020-2021 season
season_2021<-subset(PL_r,Season_End_Year=="2021")Home vs Away winning in 2021 season
point_2021<-season_2021%>%
group_by(Season_End_Year)%>%
count(FTR)plotting the result
ggplot(data =point_2021,aes(FTR,n))+
geom_col()+
ggtitle("Home vs Away winning in 2021 season")+
ylab("Numbers of matches")+
xlab("The Results")
It’s clear that home advantage was gone without the crowd’s attendance.
Now we will make some team analysis, starting by looking for the team that won the most at its home stadium.
## new dataframe to calculate the home results for each PL team
home_point<-PL_r %>%
group_by(Home)%>%
count(FTR)
## replacing "H" by "W" for winning, "A" by "L" for losing and "D" still "D" for draw
home_point$FTR[home_point$FTR=="H"]<-"W"
home_point$FTR[home_point$FTR=="A"]<-"L"
## calculating the points collected at home ground by each team
## as we know the winning team get three points, the losing team take nothing, and 1 point for each team during draw
home_point$points<- case_when(home_point$FTR=="L"~home_point$n*0,
home_point$FTR=="W"~home_point$n*3,
home_point$FTR=="D"~home_point$n*1
)
home_point2<-home_point%>%
group_by(Home)%>%
summarize(T_point=sum(points))plotting the result
ggplot(data = home_point2,aes(x=T_point,y=reorder(Home,T_point),fill=Home))+
geom_col()+
ggtitle("Points by team in Home ground")+
ylab("Team")+
xlab("Points")+
theme(legend.position="none")
As we see, Manchester United is the team with the most points collected in their home stadium, followed by Arsenal, Liverpool, and Chelsea.
Let’s see the points collected on the home ground per match for every team.