EPL Goal Scoring Time Data Shiny App

My first Shiny app.

One of the items on my to-do list for this summer is to learn and make a Shiny app. I finally did it! Check out the app here.

Description

This is a very simple Shiny app which provides data on all goal scoring events in the English Premier League, from its first season, 1992-93, to the last completed season, 2020-21.

The data were obtained from transfermarkt. In this dataset, the most important feature is the scoring time (in minute) of each goal. As improbable as it may seem, it is impossible (at least for me) to find nicely clean and publicly available data on this specific piece of information. Back in 2019-2020 when I was working on my undergraduate honors thesis on modeling EPL goal scoring, due to the amount of time and most importantly data science skills I had back then, I couldn’t either find a dataset on scoring time, or scrape the extremely messy data from the web, so I ended up collecting the data by hand. Thus, this is my motivation for developing this app.

I certainly would like to add more information to this data, such as the dates and kick-off times of the matches. There are also several issues with this app that need to be addressed in the future. In particular, the data do not specify whether a goal was an own-goal or not. As a result, the variable goal scorer represents players that scored both for and against their team. Another problem I noticed was for the scoring events in the early seasons (1990s), most goals scored during injury time of both halves had 45 (for first half) and 90 (for second half) as the scoring minutes, which did not express how much deep into stoppage time at which the goals occurred. There are other data sources out there that I have not explored which will probably help me fix this issue.

As usual, the source code for this project can be found on GitHub.

Exploring

Let’s quickly explore the EPL goal scoring data provided by my new Shiny app.

library(tidyverse)
library(kableExtra)
theme_set(theme_bw())

goals <- read_csv("goals.csv")

Famous Goals

If you are a long time follower of the EPL, these famous goals should be familiar to you.

Beckham from the halfway line

goals %>% 
  filter(goal_scorer == "David Beckham" & season == "1996-1997" & matchweek == 1) %>% 
  kable()

season	matchweek	home_club	away_club	final_score	goal_club	goal_scorer	minute
1996-1997	1	Wimbledon FC	Man Utd	0:3	Man Utd	David Beckham	90

Rooney’s bicycle-kick

goals %>% 
  filter(goal_scorer == "Wayne Rooney" & season == "2010-2011" & matchweek == 27) %>% 
  kable()

season	matchweek	home_club	away_club	final_score	goal_club	goal_scorer	minute
2010-2011	27	Man Utd	Man City	2:1	Man Utd	Wayne Rooney	78

The Hat-trick by number 20 to clinch title number 20 for Man United

goals %>% 
  filter(goal_scorer == "Robin Van Persie" & season == "2012-2013" & matchweek == 34) %>% 
  kable()

season	matchweek	home_club	away_club	final_score	goal_club	goal_scorer	minute
2012-2013	34	Man Utd	Aston Villa	3:0	Man Utd	Robin Van Persie	2
2012-2013	34	Man Utd	Aston Villa	3:0	Man Utd	Robin Van Persie	13
2012-2013	34	Man Utd	Aston Villa	3:0	Man Utd	Robin Van Persie	33

And yes…93:20

goals %>% 
  filter(goal_scorer == "Sergio Aguero" & minute == "90+4") %>% 
  kable()

season	matchweek	home_club	away_club	final_score	goal_club	goal_scorer	minute
2011-2012	38	Man City	QPR	3:2	Man City	Sergio Aguero	90+4

EDA

Who were the Top 5 All-Time EPL Goal Scorers?

goals %>% 
  count(goal_scorer, sort = TRUE) %>% 
  slice_head(n = 5) %>% 
  kable()

goal_scorer	n
Alan Shearer	260
Wayne Rooney	209
Andy Cole	187
Sergio Aguero	184
Frank Lampard	178

Which players have scored more than 30 goals in a season?

goals %>% 
  count(goal_scorer, season, sort = TRUE) %>% 
  filter(n >= 30) %>% 
  kable()

goal_scorer	season	n
Alan Shearer	1994-1995	34
Andy Cole	1993-1994	34
Mohamed Salah	2017-2018	32
Alan Shearer	1993-1994	31
Alan Shearer	1995-1996	31
Cristiano Ronaldo	2007-2008	31
Luis Suarez	2013-2014	31
Harry Kane	2017-2018	30
Kevin Phillips	1999-2000	30
Robin Van Persie	2011-2012	30
Thierry Henry	2003-2004	30

(Note: There were 22 clubs and 42 matchweeks in the first 3 EPL seasons (92-93, 93-94, 94-95), before the number of clubs was reduced to 20 (hence 38 matchweeks) at the start of 95-96.)

What are the highest-scoring teams in the EPL history?

goals %>% 
  count(goal_club, season, sort = TRUE) %>% 
  filter(n >= 90) %>% 
  kable()

goal_club	season	n
Man City	2017-2018	106
Chelsea	2009-2010	103
Man City	2013-2014	102
Man City	2019-2020	102
Liverpool	2013-2014	101
Man Utd	1999-2000	97
Man City	2018-2019	95
Man City	2011-2012	93

What is the goal scoring trend since 2010?

goals %>% 
  filter(as.numeric(str_sub(season, end = -6)) > 2009) %>% 
  mutate(season = str_replace(season, "-", "-\n")) %>% # re-format season
  count(season) %>% 
  ggplot(aes(x = season, y = n, group = 1)) +
  geom_point(aes(size = n), show.legend = FALSE) +
  geom_line() +
  labs(title = "2014-2015 was a low-scoring season")