EAFC 24 - Part 1: Exploratory Data Analysis
Adam Hutton
2023-10-16
EAFC 24 - Part 1: Exploratory Data Analysis
This is a two-part series analyzing the recently released player data from EAFC 24. The first part will undertake a exploratory data analysis of the female only data. The second part will focus on building a linear regression model to predict player rating.
Installing and loading the necessary libraries
To perform the EDA we need to install a range of different R packages to help manipulate and visual the data. If you have not already install the required packages, then delete the # to install.
Load the data.
The make the importing the data as seamless as possible, it is recommended to: 1. Create a new directory. 2. Create a “data” folder in the new directory. 3. Save the data in the “date” folder. 4. Follow the steps below.
The data can be found on Kaggle through this link: https://www.kaggle.com/datasets/nyagami/fc-24-players-database-and-stats-from-easports/code
# load the data and name it "eafc", na.strings turns any blank space to a NA value
eafc <- read.csv("Data/female_players.csv", na.strings = "NA")Generate data set summary
# use head() to see the first 6 rows of data.
head(eafc)## X Name Nation Club Position Age Overall Pace
## 1 0 Alexia Putellas Spain FC Barcelona CM 29 91 82
## 2 1 Aitana Bonmatí Spain FC Barcelona CM 25 90 81
## 3 2 Sam Kerr Australia Chelsea ST 30 90 85
## 4 3 Caroline Graham Hansen Norway FC Barcelona RW 28 90 89
## 5 4 Kadidiatou Diani France OL RW 28 89 89
## 6 5 Mapi León Spain FC Barcelona CB 28 89 75
## Shooting Passing Dribbling Defending Physicality Acceleration Sprint
## 1 90 91 92 72 78 81 82
## 2 84 83 91 75 73 82 80
## 3 88 74 90 42 85 86 84
## 4 86 88 93 47 75 90 88
## 5 85 83 88 56 77 88 90
## 6 74 81 68 90 81 73 77
## Positioning Finishing Shot Long Volleys Penalties Vision Crossing Free Curve
## 1 91 91 94 91 90 91 91 88 89 89
## 2 91 91 91 87 62 70 90 67 67 76
## 3 92 92 78 67 92 71 79 70 68 76
## 4 87 86 89 84 83 71 89 90 77 90
## 5 91 90 87 74 74 71 85 84 70 74
## 6 70 65 88 90 49 64 56 86 88 87
## Agility Balance Reactions Ball Composure Interceptions Heading Def Standing
## 1 90 89 92 94 92 78 74 60 81
## 2 93 86 91 91 85 88 54 70 81
## 3 90 82 89 91 91 24 93 44 39
## 4 94 75 83 90 84 35 71 45 50
## 5 89 84 87 90 84 55 81 46 59
## 6 70 86 90 82 91 90 83 91 91
## Sliding Jumping Stamina Strength Aggression Att.work.rate Def.work.rate
## 1 64 84 85 78 70 High Medium
## 2 67 75 80 75 60 High High
## 3 30 95 87 88 70 High High
## 4 46 84 84 74 66 High Medium
## 5 49 86 81 84 52 High High
## 6 90 87 75 83 84 High High
## Preferred.foot Weak.foot Skill.moves
## 1 Left 5 5
## 2 Right 5 4
## 3 Right 4 4
## 4 Right 4 5
## 5 Right 4 4
## 6 Left 3 2
## URL
## 1 https://www.ea.com/games/ea-sports-fc/ratings/player-ratings/alexia-putellas/227203
## 2 https://www.ea.com/games/ea-sports-fc/ratings/player-ratings/aitana-bonmati/241667
## 3 https://www.ea.com/games/ea-sports-fc/ratings/player-ratings/sam-kerr/227125
## 4 https://www.ea.com/games/ea-sports-fc/ratings/player-ratings/caroline-graham-hansen/227102
## 5 https://www.ea.com/games/ea-sports-fc/ratings/player-ratings/kadidiatou-diani/227361
## 6 https://www.ea.com/games/ea-sports-fc/ratings/player-ratings/mapi-leon/236479
## Gender GK
## 1 FALSE NA
## 2 FALSE NA
## 3 FALSE NA
## 4 FALSE NA
## 5 FALSE NA
## 6 FALSE NAExamine the data for missing values
# The function creates a sum for every missing value in a column.
colSums(is.na(eafc))## X Name Nation Club Position
## 0 0 0 0 0
## Age Overall Pace Shooting Passing
## 0 0 0 0 0
## Dribbling Defending Physicality Acceleration Sprint
## 0 0 0 0 0
## Positioning Finishing Shot Long Volleys
## 0 0 0 0 0
## Penalties Vision Crossing Free Curve
## 0 0 0 0 0
## Agility Balance Reactions Ball Composure
## 0 0 0 0 0
## Interceptions Heading Def Standing Sliding
## 0 0 0 0 0
## Jumping Stamina Strength Aggression Att.work.rate
## 0 0 0 0 0
## Def.work.rate Preferred.foot Weak.foot Skill.moves URL
## 0 0 0 0 0
## Gender GK
## 0 1316The missing data relates to whether the player is a GK or not. Only GK will have a rating in this column. Otherwise there is no other missing values.
# Remove the missing values from the data set, remove GK
eafc_outfield <- eafc %>%
select(-c("GK","URL")) %>%
filter(Position !="GK")Examine the data types
str(eafc_outfield)## 'data.frame': 1316 obs. of 45 variables:
## $ X : int 0 1 2 3 4 5 6 7 8 9 ...
## $ Name : chr "Alexia Putellas" "Aitana Bonmatí" "Sam Kerr" "Caroline Graham Hansen" ...
## $ Nation : chr "Spain" "Spain" "Australia" "Norway" ...
## $ Club : chr "FC Barcelona" "FC Barcelona" "Chelsea" "FC Barcelona" ...
## $ Position : chr "CM" "CM" "ST" "RW" ...
## $ Age : int 29 25 30 28 28 28 34 28 32 33 ...
## $ Overall : int 91 90 90 90 89 89 89 89 88 88 ...
## $ Pace : int 82 81 85 89 89 75 83 75 69 53 ...
## $ Shooting : int 90 84 88 86 85 74 89 89 51 67 ...
## $ Passing : int 91 83 74 88 83 81 80 77 67 69 ...
## $ Dribbling : int 92 91 90 93 88 68 89 90 57 69 ...
## $ Defending : int 72 75 42 47 56 90 47 39 89 91 ...
## $ Physicality : int 78 73 85 75 77 81 77 76 80 84 ...
## $ Acceleration : int 81 82 86 90 88 73 82 73 65 53 ...
## $ Sprint : int 82 80 84 88 90 77 84 77 72 53 ...
## $ Positioning : int 91 91 92 87 91 70 90 93 39 70 ...
## $ Finishing : int 91 91 92 86 90 65 94 91 51 65 ...
## $ Shot : int 94 91 78 89 87 88 89 83 85 80 ...
## $ Long : int 91 87 67 84 74 90 82 72 68 84 ...
## $ Volleys : int 90 62 92 83 74 49 83 90 60 55 ...
## $ Penalties : int 91 70 71 71 71 64 88 84 51 67 ...
## $ Vision : int 91 90 79 89 85 56 75 76 51 59 ...
## $ Crossing : int 88 67 70 90 84 86 71 71 59 53 ...
## $ Free : int 89 67 68 77 70 88 73 75 40 58 ...
## $ Curve : int 89 76 76 90 74 87 68 79 69 65 ...
## $ Agility : int 90 93 90 94 89 70 83 74 66 53 ...
## $ Balance : int 89 86 82 75 84 86 67 48 65 34 ...
## $ Reactions : int 92 91 89 83 87 90 82 90 91 87 ...
## $ Ball : int 94 91 91 90 90 82 92 90 79 75 ...
## $ Composure : int 92 85 91 84 84 91 91 86 75 84 ...
## $ Interceptions : int 78 88 24 35 55 90 46 23 89 89 ...
## $ Heading : int 74 54 93 71 81 83 84 92 85 94 ...
## $ Def : int 60 70 44 45 46 91 41 42 92 92 ...
## $ Standing : int 81 81 39 50 59 91 43 33 87 92 ...
## $ Sliding : int 64 67 30 46 49 90 39 30 88 81 ...
## $ Jumping : int 84 75 95 84 86 87 90 89 79 81 ...
## $ Stamina : int 85 80 87 84 81 75 87 75 72 72 ...
## $ Strength : int 78 75 88 74 84 83 81 80 82 91 ...
## $ Aggression : int 70 60 70 66 52 84 53 65 87 84 ...
## $ Att.work.rate : chr "High" "High" "High" "High" ...
## $ Def.work.rate : chr "Medium" "High" "High" "Medium" ...
## $ Preferred.foot: chr "Left" "Right" "Right" "Right" ...
## $ Weak.foot : int 5 5 4 4 4 3 5 4 3 3 ...
## $ Skill.moves : int 5 4 4 5 4 2 4 4 2 2 ...
## $ Gender : logi FALSE FALSE FALSE FALSE FALSE FALSE ...# use sapply to count the data types in "eafc"
datatype_count <- sapply(eafc_outfield, class)
print(table(datatype_count))## datatype_count
## character integer logical
## 7 37 1Categorical variable frequency
# Filter the dataframe to only have the character columns, excluding the URL column
cat_counts <- eafc_outfield %>%
select(c("Nation","Club","Position","Att.work.rate","Def.work.rate","Preferred.foot", "Weak.foot", "Skill.moves", "Gender")) %>%
lapply(table)
for (i in 1:length(cat_counts)){
cat("Counts for", names(cat_counts)[i], ":\n")
print(cat_counts[[i]])
cat("\n")
}## Counts for Nation :
##
## Algeria Andorra Argentina
## 3 1 5
## Australia Austria Belarus
## 13 18 1
## Belgium Bermuda Bosnia and Herzegovina
## 7 1 1
## Brazil Cameroon Canada
## 25 6 26
## Chile China PR Colombia
## 4 1 7
## Costa Rica Côte d'Ivoire Curaçao
## 1 4 1
## Czech Republic Denmark Ecuador
## 13 22 1
## England Finland France
## 94 6 151
## Germany Ghana Greece
## 184 1 1
## Haiti Holland Hungary
## 5 40 3
## Iceland Israel Italy
## 8 1 12
## Jamaica Japan Kenya
## 10 9 1
## Korea Republic Kosovo Mali
## 2 2 1
## Mexico Montenegro Morocco
## 7 1 4
## New Zealand Nigeria North Macedonia
## 5 14 1
## Northern Ireland Norway Panama
## 3 18 2
## Paraguay Poland Portugal
## 1 15 16
## Republic of Ireland Scotland Serbia
## 13 26 5
## Slovakia Slovenia South Africa
## 3 3 3
## Spain St. Kitts and Nevis Sweden
## 182 1 30
## Switzerland Trinidad and Tobago Turkey
## 30 1 1
## United States Venezuela Wales
## 219 6 12
## Zambia
## 2
##
## Counts for Club :
##
## 1. FC Köln 1. FC Nürnberg Ajax Angel City FC
## 19 19 18 22
## Arsenal AS Saint Étienne Aston Villa Athletic Club
## 23 19 17 22
## Atlético de Madrid Bordeaux Brighton Bristol City
## 21 12 16 15
## Chelsea Chicago Red Stars Dijon FCO En Avant Guingamp
## 25 24 12 15
## Everton FC Barcelona FC Bayern München FC Fleury 91
## 16 21 22 16
## FC Rosengård FC Zürich Frankfurt Glasgow City FC
## 16 18 22 19
## Granada CF Havre AC Houston Dash Juventus
## 3 17 24 20
## KC Current Leicester City Levante LP Levante UD
## 24 14 14 20
## Leverkusen Liverpool LOSC Lille Madrid CFF
## 21 20 15 16
## Manchester City Manchester Utd Montpellier MSV Duisburg
## 20 19 16 17
## NC Courage NJ/NY Gotham OL OL Reign
## 22 23 26 22
## Orlando Pride Paris FC Paris SG Portland Thorns
## 22 14 22 23
## Rac. Louisville RB Leipzig Real Betis Real Madrid CF
## 20 21 15 21
## Real Sociedad San Diego Wave SC Freiburg SD Eibar
## 15 21 21 3
## Sevilla FC SGS Essen SL Benfica Slavia Praha
## 15 20 20 15
## Sporting Huelva Spurs Stade de Reims SV Werder Bremen
## 12 20 18 19
## TSG Hoffenheim UDG Tenerife Valencia CF VfL Wolfsburg
## 21 21 17 23
## Villarreal CF Washington Spirit West Ham
## 16 22 17
##
## Counts for Position :
##
## CAM CB CDM CF CM LB LM LW LWB RB RM RW RWB ST
## 90 256 108 3 206 113 53 51 2 117 45 54 4 214
##
## Counts for Att.work.rate :
##
## High Low Medium
## 442 64 810
##
## Counts for Def.work.rate :
##
## High Low Medium
## 217 71 1028
##
## Counts for Preferred.foot :
##
## Left Right
## 208 1108
##
## Counts for Weak.foot :
##
## 1 2 3 4 5
## 2 248 832 201 33
##
## Counts for Skill.moves :
##
## 2 3 4 5
## 440 684 179 13
##
## Counts for Gender :
##
## FALSE
## 1316Integer-Encoding
eafc_clean <- eafc_outfield %>%
mutate(Att.work.rate_int = case_when(
Att.work.rate == "Low" ~ 1,
Att.work.rate == "Medium" ~ 2,
Att.work.rate == "High" ~ 3,
TRUE ~ NA_integer_),
Def.work.rate_int = case_when(
Att.work.rate == "Low" ~ 1,
Att.work.rate == "Medium" ~ 2,
Att.work.rate == "High" ~ 3,
TRUE ~ NA_integer_
))Top 20 Clubs
count_club <- eafc_outfield %>%
group_by(Club) %>%
summarise(count = n()) %>%
arrange(desc(count)) %>%
slice(1:20)
count_club %>%
ggplot(aes(x = fct_reorder(Club, -count),y = count)) +
geom_col(aes(fill = count)) +
scale_fill_gradient(low = "darkgreen", high = "darkblue")+
theme(axis.text.x = element_text(angle = 90))+
labs(title="Count of Players Per Club - Top 20", x = "Clubs")
Top 20 Nations
count_nation <- eafc_outfield %>%
group_by(Nation) %>%
summarise(count = n()) %>%
arrange(desc(count)) %>%
slice(1:20)
count_nation %>%
ggplot(aes(x = fct_reorder(Nation, -count),y = count)) +
geom_col(aes(fill = count)) +
scale_fill_gradient(low = "darkgreen", high = "darkblue")+
theme(axis.text.x = element_text(angle = 90))
labs(title="Count of Players Per Nation - Top 20", x = "Clubs")## $x
## [1] "Clubs"
##
## $title
## [1] "Count of Players Per Nation - Top 20"
##
## attr(,"class")
## [1] "labels"Breakdown of Position
eafc_outfield <- eafc_outfield %>%
mutate(
Position_Type = case_when(
Position %in% c("RWB","RB","CB","LB","LWB") ~ "DEF",
Position %in% c("ST","CF")~ "ATT",
TRUE ~ "MID")
)
count_pos <- eafc_outfield %>%
group_by(Position) %>%
summarise(count = n()) %>%
arrange(desc(count))
gridExtra::grid.arrange(
eafc_outfield %>%
filter(Position_Type == "ATT") %>%
ggplot(aes(x = Position, fill = after_stat(count))) +
geom_bar() +
ylim(0,300)+
scale_fill_gradient(low = "lightgreen", high = "blue")+
geom_text(aes(label = after_stat(count)), stat = "count", vjust = -0.5, size = 3) +
theme_bw() +
theme(legend.position = "none"
)+
labs(title="Count of Players Per Position - Attackers"),
eafc_outfield %>%
filter(Position_Type == "MID") %>%
ggplot(aes(x = Position, fill = after_stat(count))) +
geom_bar() +
ylim(0,300)+
scale_fill_gradient(low = "lightgreen", high = "blue")+
geom_text(aes(label = after_stat(count)), stat = "count", vjust = -0.5, size = 3)+
theme_bw() +
theme(legend.position = "none")+
labs(title="Count of Players Per Position - Midfielders")
,
eafc_outfield %>%
filter(Position_Type == "DEF") %>%
ggplot(aes(x = Position, fill = after_stat(count))) +
geom_bar() +
ylim(0,300)+
scale_fill_gradient(low = "lightgreen", high = "blue")+
geom_text(aes(label = after_stat(count)), stat = "count", vjust = -0.5, size = 3)+
theme_bw() +
theme(legend.position = "none")+
labs(title="Count of Players Per Position - Defenders"),
ncol = 1
)
Numerical Data Distribution
Using the describe() function we can generate the summary statistics, to provide insight into each variable. The skew column can be used to identify variables do not follow a normal distribution by either having a skew value of greater than 0.5 or less than -0.5. The manipulation of the skew variables will occur in Part 2.
# use describe() to examine the summary statistics of "eafc"
eafc_outfield_num <- eafc_outfield %>%
select(-c("X","Name", "Nation","Club","Position","Att.work.rate","Def.work.rate","Preferred.foot", "Weak.foot", "Skill.moves", "Gender","Position_Type"))
eafc_outfield_num %>% describe(na.rm = TRUE)## vars n mean sd median trimmed mad min max range skew
## Age 1 1316 25.09 4.35 25.0 24.88 4.45 17 40 23 0.46
## Overall 2 1316 74.16 6.10 74.0 74.28 5.93 52 91 39 -0.23
## Pace 3 1316 69.54 10.06 70.0 69.90 8.90 33 94 61 -0.43
## Shooting 4 1316 60.77 13.78 63.0 61.31 14.83 22 90 68 -0.33
## Passing 5 1316 66.62 8.28 67.0 66.80 7.41 40 91 51 -0.21
## Dribbling 6 1316 69.01 10.57 70.0 69.54 10.38 22 93 71 -0.58
## Defending 7 1316 60.46 16.18 65.0 61.45 17.79 20 91 71 -0.48
## Physicality 8 1316 67.29 8.17 67.0 67.49 7.41 36 89 53 -0.28
## Acceleration 9 1316 69.31 10.66 70.0 69.82 10.38 31 95 64 -0.57
## Sprint 10 1316 69.71 10.29 70.0 70.17 8.90 33 94 61 -0.56
## Positioning 11 1316 63.76 16.22 69.0 65.13 14.83 18 93 75 -0.69
## Finishing 12 1316 60.51 16.06 64.0 61.28 17.79 20 94 74 -0.38
## Shot 13 1316 74.61 6.58 75.0 74.81 5.93 50 94 44 -0.36
## Long 14 1316 65.83 9.75 67.0 66.22 10.38 29 93 64 -0.33
## Volleys 15 1316 52.72 15.61 54.5 52.80 18.53 17 92 75 -0.06
## Penalties 16 1316 57.95 11.21 59.0 57.91 11.86 32 91 59 0.04
## Vision 17 1316 63.45 13.25 66.0 64.38 13.34 22 92 70 -0.61
## Crossing 18 1316 60.89 14.14 64.0 61.69 14.83 22 93 71 -0.48
## Free 19 1316 54.75 12.84 56.0 54.86 14.83 20 89 69 -0.07
## Curve 20 1316 59.90 12.98 62.0 60.51 11.86 20 92 72 -0.42
## Agility 21 1316 67.77 11.26 69.0 68.23 10.38 32 94 62 -0.42
## Balance 22 1316 67.45 11.61 68.0 67.81 11.86 30 94 64 -0.35
## Reactions 23 1316 71.97 7.53 72.0 72.20 7.41 45 92 47 -0.36
## Ball 24 1316 74.24 6.94 74.0 74.30 5.93 49 94 45 -0.11
## Composure 25 1316 68.45 8.48 68.0 68.39 8.90 41 93 52 0.05
## Interceptions 26 1316 59.95 18.55 68.0 61.60 14.83 15 90 75 -0.69
## Heading 27 1316 63.75 11.94 65.0 64.16 13.34 24 94 70 -0.29
## Def 28 1316 58.80 19.19 65.0 59.87 20.76 14 92 78 -0.41
## Standing 29 1316 62.28 17.57 69.0 63.93 14.83 12 92 80 -0.75
## Sliding 30 1316 57.25 17.77 62.0 58.30 19.27 10 90 80 -0.46
## Jumping 31 1316 72.23 10.43 73.0 72.56 10.38 27 95 68 -0.36
## Stamina 32 1316 69.19 10.38 69.0 69.57 10.38 33 94 61 -0.46
## Strength 33 1316 67.81 10.66 68.5 68.22 9.64 32 92 60 -0.43
## Aggression 34 1316 62.24 12.25 63.0 62.81 11.86 22 94 72 -0.40
## kurtosis se
## Age -0.11 0.12
## Overall 0.38 0.17
## Pace 0.56 0.28
## Shooting -0.72 0.38
## Passing -0.13 0.23
## Dribbling 0.60 0.29
## Defending -0.95 0.45
## Physicality 0.22 0.23
## Acceleration 0.70 0.29
## Sprint 0.94 0.28
## Positioning -0.33 0.45
## Finishing -0.81 0.44
## Shot 0.44 0.18
## Long -0.09 0.27
## Volleys -0.88 0.43
## Penalties -0.35 0.31
## Vision -0.06 0.37
## Crossing -0.51 0.39
## Free -0.58 0.35
## Curve -0.07 0.36
## Agility 0.32 0.31
## Balance 0.16 0.32
## Reactions 0.52 0.21
## Ball 0.11 0.19
## Composure -0.16 0.23
## Interceptions -0.75 0.51
## Heading -0.40 0.33
## Def -1.12 0.53
## Standing -0.56 0.48
## Sliding -0.96 0.49
## Jumping 0.03 0.29
## Stamina 0.73 0.29
## Strength 0.46 0.29
## Aggression 0.01 0.34Divide the Numerical Data into Categories
The numerical data types can be divided into three categories. These are physical, technical and tactical.
Physical Numerical Data
Boxplots
eafc_phys <- eafc_outfield_num %>%
select(c("Age","Pace","Physicality","Acceleration","Sprint","Agility", "Balance","Reactions", "Stamina","Strength","Aggression"))
eafc_phys %>%
keep(is.numeric) %>%
gather() %>%
ggplot(aes(value))+
facet_wrap(~key, scales = "free") +
geom_boxplot(outlier.colour = "red", outlier.shape = 1)+
labs(title = "Spread of Physical Characteristics")
Histograms
eafc_phys %>%
keep(is.numeric) %>%
gather() %>%
ggplot(aes(value))+
facet_wrap(~key, scales = "free") +
geom_histogram()+
labs(title = "Distribution Physical Characteristics")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Technical Numerical Data
Boxplots
eafc_tech <- eafc_outfield_num %>%
select(c("Shooting","Passing","Dribbling","Defending", "Finishing","Shot","Long","Volleys","Penalties","Crossing","Free","Curve","Ball","Interceptions","Heading","Def","Standing","Sliding","Jumping"))
eafc_tech %>%
keep(is.numeric) %>%
gather() %>%
ggplot(aes(value))+
facet_wrap(~key, scales = "free") +
geom_boxplot(outlier.colour = "red", outlier.shape = 1)+
labs(title = "Spread of Technical Characteristics")
Histograms
eafc_tech %>%
keep(is.numeric) %>%
gather() %>%
ggplot(aes(value))+
facet_wrap(~key, scales = "free") +
geom_histogram()+
labs(title = "Distribution Technical Characteristics")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Tactical Numerical Data
Boxplots
eafc_tact <- eafc_outfield_num %>%
select(c("Positioning","Vision","Composure"))
eafc_tact %>%
keep(is.numeric) %>%
gather() %>%
ggplot(aes(value))+
facet_wrap(~key, scales = "free") +
geom_boxplot()+
labs(title = "Spread of Tactical Characteristics")
Histograms
eafc_tact %>%
keep(is.numeric) %>%
gather() %>%
ggplot(aes(value))+
facet_wrap(~key, scales = "free") +
geom_histogram()+
labs(title = "Distribution Tactical Characteristics")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Physical Ability Visualization
Strength, Sprint and Stamina for position types.
SSS_mean <- eafc_outfield %>%
select(c("Strength","Sprint","Stamina")) %>%
group_by(eafc_outfield$Position_Type) %>%
summarise(
mean_str = mean(Strength),
mean_spr = mean(Sprint),
mean_sta = mean(Stamina)
)
print(SSS_mean)## # A tibble: 3 × 4
## `eafc_outfield$Position_Type` mean_str mean_spr mean_sta
## <chr> <dbl> <dbl> <dbl>
## 1 ATT 71.0 74.1 66.0
## 2 DEF 69.3 67.3 68.3
## 3 MID 65.5 70.1 71.0Strength and Overall Value based on Position Type
gridExtra::grid.arrange(
eafc_outfield %>%
filter(Position_Type == "ATT") %>%
ggplot(aes(x = Strength, y = Overall,fill = Overall))+
geom_col() +
geom_vline(aes(xintercept = 70.95853), linetype = "dotted", color = "black", linewidth=1)+
scale_fill_viridis_c()+
theme_bw()+
labs(title="Defenders Players Overall vs Strength")
,
eafc_outfield %>%
filter(Position_Type == "MID") %>%
ggplot(aes(x = Strength, y = Overall,fill = Overall))+
geom_col() +
geom_vline(aes(xintercept = 70.95853), linetype = "dotted", color = "black", linewidth=1)+ scale_fill_viridis_c()+
theme_bw()+
labs(title="Defenders Players Overall vs Strength"),
eafc_outfield %>%
filter(Position_Type == "DEF") %>%
ggplot(aes(x = Strength, y = Overall,fill = Overall))+
geom_col() +
geom_vline(aes(xintercept = 70.95853), linetype = "dotted", color = "black", linewidth=1)+
scale_fill_viridis_c()+
theme_bw()+
labs(title="Defenders Players Overall vs Strength"),
ncol = 1
)
Sprinting Ability and Overall Value based on Position Type
gridExtra::grid.arrange(
eafc_outfield %>%
filter(Position_Type == "ATT") %>%
ggplot(aes(x = Sprint, y = Overall,fill = Overall))+
geom_col() +
geom_vline(aes(xintercept = 70.95853), linetype = "dotted", color = "black", linewidth=1)+
scale_fill_viridis_c()+
theme_bw()+
labs(title="Attacking Players Overall vs Sprint Ability")
,
eafc_outfield %>%
filter(Position_Type == "MID") %>%
ggplot(aes(x = Sprint, y = Overall,fill = Overall))+
geom_col() +
geom_vline(aes(xintercept = 70.95853), linetype = "dotted", color = "black", linewidth=1)+ scale_fill_viridis_c()+
theme_bw()+
labs(title="Midfielders Players Overall vs Sprint Ability"),
eafc_outfield %>%
filter(Position_Type == "DEF") %>%
ggplot(aes(x = Sprint, y = Overall,fill = Overall))+
geom_col() +
geom_vline(aes(xintercept = 70.95853), linetype = "dotted", color = "black", linewidth=1)+
scale_fill_viridis_c()+
theme_bw()+
labs(title="Defenders Players Overall vs Sprint Ability"),
ncol = 1
)
Stamina and Overall Value
gridExtra::grid.arrange(
eafc_outfield %>%
filter(Position_Type == "ATT") %>%
ggplot(aes(x = Stamina, y = Overall,fill = Overall))+
geom_col() +
geom_vline(aes(xintercept = 70.95853), linetype = "dotted", color = "black", linewidth=1)+
scale_fill_viridis_c()+
theme_bw()+
labs(title="Attacking Players Overall vs Stamina")
,
eafc_outfield %>%
filter(Position_Type == "MID") %>%
ggplot(aes(x = Stamina, y = Overall,fill = Overall))+
geom_col() +
geom_vline(aes(xintercept = 70.95853), linetype = "dotted", color = "black", linewidth=1)+ scale_fill_viridis_c()+
theme_bw()+
labs(title="Midfielders Players Overall vs Stamina"),
eafc_outfield %>%
filter(Position_Type == "DEF") %>%
ggplot(aes(x = Stamina, y = Overall,fill = Overall))+
geom_col() +
geom_vline(aes(xintercept = 70.95853), linetype = "dotted", color = "black", linewidth=1)+
scale_fill_viridis_c()+
theme_bw()+
labs(title="Defenders Players Overall vs Stamina"),
ncol = 1
)
Top Strength-Speed-Stamina in Clubs and Players
# create the new metric called SSS
SSS_club <- eafc_outfield %>%
group_by(Club) %>%
summarise(
SSS = mean(Strength) + mean(Sprint) + mean(Stamina)
)
SSS_player <- eafc_outfield %>%
group_by(Name) %>%
summarise(
SSS = round(mean(Strength) + mean(Sprint) + mean(Stamina),2)
)Club SSS
SSS_club %>%
arrange(desc(SSS)) %>%
slice(1:15) %>%
ggplot(aes(x = fct_reorder(Club,SSS), y = SSS, fill = SSS))+
geom_col()+
theme(axis.text.x = element_text(angle = 90))+
geom_text(aes(label = round(SSS,2)), vjust = -0.5, size = 2)+
scale_fill_viridis_c(option="viridis",direction=-1)+
labs(title="Top 10 Clubs with the highest SSS", x = "Club")
SSS for Player
SSS_player %>%
arrange(desc(SSS)) %>%
slice(1:15) %>%
ggplot(aes(x = fct_reorder(Name,SSS), y = SSS, fill = SSS))+
geom_col()+
theme(axis.text.x = element_text(angle = 90))+
geom_text(aes(label = round(SSS,2)), vjust = -0.5, size = 2)+
scale_fill_viridis_c(option="viridis",direction=-1)+
labs(title="Top 10 Players with the highest SSS", x = "Player")
SSS_player <- eafc_outfield %>%
group_by(Name) %>%
summarise(
SSS = round(mean(Strength) + mean(Sprint) + mean(Stamina),2)
)
SSS_Overall <- eafc_outfield %>%
mutate(
SSS = round((Strength) + (Sprint) + (Stamina),2)
) %>%
ggplot(aes(x=SSS, y= Overall, color = Overall))+
geom_point()+
scale_color_viridis_c(option="turbo",direction=-1)+
theme_bw()+
stat_smooth(method="lm")+
labs(title="Overall Rating vs SSS score")
ggplotly(SSS_Overall)## `geom_smooth()` using formula = 'y ~ x'## Warning: The following aesthetics were dropped during statistical transformation: colour
## ℹ This can happen when ggplot fails to infer the correct grouping structure in
## the data.
## ℹ Did you forget to specify a `group` aesthetic or to convert a numerical
## variable into a factor?Technical Ability
Set Piece Specialists
set_piece <- eafc_outfield %>%
arrange(desc(Free)) %>%
slice(1:15) %>%
ggplot(aes(x = Free, y = Penalties, color = Overall, label = Name))+
geom_point(aes(size = 5))+
scale_color_viridis_c(option="viridis",direction=-1)+
labs(title = "Set Piece Specialists",x="Free Kick")
ggplotly(set_piece)Scoring Ability
scoring <- eafc_outfield %>%
arrange(desc(Shooting)) %>%
slice(1:15) %>%
ggplot(aes(x = Finishing, y = Shooting, color = Overall, label = Name))+
geom_point(aes(size = 5))+
scale_color_viridis_c(option="viridis",direction=-1)+
labs(title = "Finishers",x="Free Kick")
ggplotly(scoring)Overall Ratings
Player Ratings
eafc_outfield %>%
arrange(desc(Overall)) %>%
slice(1:10) %>%
ggplot(aes(x = fct_reorder(Name,Overall), y = Overall, fill = Overall))+
geom_col()+
theme(axis.text.x = element_text(angle = 45, vjust = 0.5, size = 8))+
geom_text(aes(label = round(Overall,2)), vjust = -0.5, size = 3)+
scale_fill_viridis_c(option="viridis",direction=-1)+
labs(title="Top 10 Players - Overall Rating", x= "Player")
Player Ratings per Position
Attackers
eafc_outfield %>%
filter(Position_Type =="ATT") %>%
arrange(desc(Overall)) %>%
slice(1:5) %>%
ggplot(aes(x = fct_reorder(Name,Overall), y = Overall, fill = Overall))+
geom_col()+
theme(axis.text.x = element_text(angle = 45, vjust = 0.5))+
geom_text(aes(label = round(Overall,2)), vjust = -0.5, size = 3)+
scale_fill_viridis_c(option="viridis",direction=-1)+
labs(title="Top 5 Attackers - Overall Rating", x= "Player")
Midfielders
eafc_outfield %>%
filter(Position_Type =="MID") %>%
arrange(desc(Overall)) %>%
slice(1:5) %>%
ggplot(aes(x = fct_reorder(Name,Overall), y = Overall, fill = Overall))+
geom_col()+
theme(axis.text.x = element_text(angle = 45, vjust = 0.5))+
geom_text(aes(label = round(Overall,2)), vjust = -0.5, size = 3)+
scale_fill_viridis_c(option="viridis",direction=-1)+
labs(title="Top 5 Midfielders - Overall Rating", x= "Player")
Defenders
eafc_outfield %>%
filter(Position_Type =="DEF") %>%
arrange(desc(Overall)) %>%
slice(1:5) %>%
ggplot(aes(x = fct_reorder(Name,Overall), y = Overall, fill = Overall))+
geom_col()+
theme(axis.text.x = element_text(angle = 45, vjust = 0.5))+
geom_text(aes(label = round(Overall,2)), vjust = -0.5, size = 3)+
scale_fill_viridis_c(option="viridis",direction=-1)+
labs(title="Top 5 Defenders - Overall Rating", x= "Player")
Club Ratings
eafc_outfield %>%
group_by(Club) %>%
summarise(Overall = mean(Overall)) %>%
arrange(desc(Overall)) %>%
slice(1:10) %>%
ggplot(aes(x = fct_reorder(Club,Overall), y = Overall, fill = Overall))+
geom_col()+
theme(axis.text.x = element_text(angle = 45, vjust = 0.5, size = 8))+
geom_text(aes(label = round(Overall,2)), vjust = -0.5, size = 3)+
scale_fill_viridis_c(option="viridis",direction=-1)+
labs(title="Top 10 Clubs - Average Overall Rating", x= "Club")
Correlation to Overall Rating
Correlation of Physical Traits
phys_corr <- eafc_outfield_num %>%
select(c("Overall","Age","Pace","Physicality","Acceleration","Sprint","Agility", "Balance","Reactions", "Stamina","Strength","Aggression"))
ggcorr(phys_corr, nbreaks = 4, palette = "RdGy", hjust = 0.75,label = TRUE, label_size = 3, label_color = "white", size = 2)
Correlation of Technical Traits
tech_corr <- eafc_outfield_num %>%
select(c("Overall","Shooting","Passing","Dribbling","Defending", "Finishing","Shot","Long","Volleys","Penalties","Crossing","Free","Curve","Ball","Interceptions","Heading","Def","Standing","Sliding","Jumping"))
ggcorr(tech_corr, nbreaks = 4, palette = "RdGy", hjust = 0.75,label = TRUE, label_size = 3, label_color = "white", size = 2)
Correlation of Tactical Traits
tact_corr <- eafc_outfield_num %>%
select(c("Overall","Positioning","Vision","Composure"))
ggcorr(tact_corr, nbreaks = 4, palette = "RdGy", hjust = 0.75,label = TRUE, label_size = 3, label_color = "white", size = 2)