Manchester United had the highest average attendance in the Premier League in the 2023/24 season, attracting crowds of around 73,531 to Old Trafford. Meanwhile, city rivals and Premier League champions Manchester City had an average attendance of 53,192.

In 2022/23, the average matchday attendance in the Premier League was just over 40,000, representing a slight increase on the previous year. In the same season, Bundesliga games saw an average attendance of nearly 43,000.

The aggregate attendance of the English Premier League, commonly referred to as the Premier League, has exhibited a gradual increase from 13 million attendees in 2009/2010 to approximately 15 million attendees in 2021/22. In 2020/21, the total aggregate attendance at the games fell to its lowest value throughout this period as a result of the coronavirus (COVID-19) pandemic . The Premier League The Premier League, currently consisting of 20 professional football clubs, constitutes the highest level of professional club football within the United Kingdom (UK). Alongside the so-called “Big-Five” leagues of Europe, which include the top-tier football leagues of England, Spain, Italy, Germany, and France; the Premier League is considered to be one of the most widely followed and well-known football leagues in the world. Revenue The Premier League was established in 1992 following the decision of the then first division of the English Football league to capitalize on a lucrative television rights deal. A decision that has led to the clubs within the Premier League consistently yielding the largest combined revenue of the ‘Big-Five’ leagues of Europe with the combined revenue of all 20 Premier League clubs projected to be around 6.2 billion euros in the 2021/22 season; over two billion euros more than the expected combined revenue of their counterparts in Spain and Germany, who rank second and third respectively. The 2020/21 season The 2020/2021 season, following a delay until the 12 September due to the postponement of the previous season's conclusion as a result of the impact of the COVID-19 containment measures culminated in Manchester City FC reclaiming their title from the defending champions, Liverpool FC. Leading to the club's third championship title within the last four seasons. In the 2021/22 season the Premier League bounced back from COVID-19 as the strongest attendance wise in the Big-Five with an average of close to 40 thousand spectators per match.

Results for group 1 v group 2 balanced data set (Best Average Test Performance = 61.5% and Best Average Test Error = 11.6% with a combination of seven variables) and group 1 v group 2 model variables as means and standard deviations for player groupings.

This statistic presents the average stadium utilization of professional football matches of The Premier League in England from 2010 to 2017. In 2017, the average stadium of The Premier League was utilized at 96.5 percent.

Context

As many others I have asked myself if it is possible to use machine learning in order to create valid predictions for football (soccer) match outcomes. Hence I created a dataset consisting of historic match data for the German Bundesliga (1st and 2nd Division) as well as the English Premier League reaching back as far as 1993 up to 2016. Besides the mere information concerning goals scored and home/draw/away win the dataset also includes per site (team) data such as transfer value per team (pre-season), the squad strength, etc. Unfortunately I was only able to find sources for these advanced attributes going back to the 2005 season.
I have used this dataset with different machine learning algorithms including random forests, XGBoost as well as different recurrent neural network architectures (in order to potentially identify recurring patterns in winning streaks, etc.). I'd like to share the approaches I used as separate Kernels here as well. So far I did not manage to exceed an accuracy of 53% consistently on a validation set using 2016 season of Bundesliga 1 (no information rate = 49%).

Although I have done some visual exploration before implementing the different machine learning approaches using Tableau, I think a visual exploration kernel would be very beneficial.

Content

The data comes as an Sqlite file containing the following tables and fields:

Table: Matches

• Match_ID (int): unique ID per match
• Div (str): identifies the division the match was played in (D1 = Bundesliga, D2 = Bundesliga 2, E0 = English Premier League)
• Season (int): Season the match took place in (usually covering the period of August till May of the following year)
• Date (str): Date of the match
• HomeTeam (str): Name of the home team
• AwayTeam (str): Name of the away team
• FTHG (int) (Full Time Home Goals): Number of goals scored by the home team
• FTAG (int) (Full Time Away Goals): Number of goals scored by the away team
• FTR (str) (Full Time Result): 3-way result of the match (H = Home Win, D = Draw, A = Away Win)

Table: Teams

• Season (str): Football season for which the data is valid
• TeamName (str): Name of the team the data concerns
• KaderHome (str): Number of Players in the squad
• AvgAgeHome (str): Average age of players
• ForeignPlayersHome (str): Number of foreign players (non-German, non-English respectively) playing for the team
• OverallMarketValueHome (str): Overall market value of the team pre-season in EUR (based on data from transfermarkt.de)
• AvgMarketValueHome (str): Average market value (per player) of the team pre-season in EUR (based on data from transfermarkt.de)
• StadiumCapacity (str): Maximum stadium capacity of the team's home stadium

Table: Unique Teams

• TeamName (str): Name of a team
• Unique_Team_ID (int): Unique identifier for each team

Table: Teams_in_Matches

• Match_ID (int): Unique match ID
• Unique_Team_ID (int): Unique team ID (This table is used to easily retrieve each match a given team has played in)

Based on these tables I created a couple of views which I used as input for my machine learning models:

View: FlatView

Combination of all matches with the respective additional data from Teams table for both home and away team.

View: FlatView_Advanced

Same as Flatview but also includes Unique_Team_ID and Unique_Team in order to easily retrieve all matches played by a team in chronological order.

View: FlatView_Chrono_TeamOrder_Reduced

Similar to Flatview_Advanced, however missing the additional attributes from team in order to have a longer history including years 1993 - 2004. Especially interesting if one is only interested in analyzing winning/loosing streaks.

Acknowledgements

Thanks to football-data.co.uk and transfermarkt.de for providing the raw data used in this dataset.

Inspiration

Please feel free to use the humble dataset provided here for any purpose you want. To me it would be most interesting if others think that recurrent neural networks could in fact be of help (and even maybe outperform classical feature engineering) in identifying streaks of losses and wins. In the literature I mostly only found example of RNN application where the data were time series in a very narrow sense (e.g. temperature measurements over time) hence it would be interesting to get your input on this question.

Maybe someone also finds additional attributes per team or match which have substantial impact on match outcome. So far I have found the "Market Value" of a team to be by far the best predictor when two teams face each other, which makes sense as the market value usually tends to correlate closely with the strength of a team and it's propects at winning

In the 2022/23 season, players in the English Premier League (EPL) tended to be sent off at a higher frequency than Women's Super League (WSL) players, with a red card being shown nearly every 13 games on average. Meanwhile, in the same season, the WSL saw a red card once in nearly every 16 matches.

This statistic shows English Premier League and Championship clubs' average revenues in the 2019/20 season, by stream. During this season, clubs playing in the UEFA Champions League recorded 178 million euros in revenue from broadcasting.

Notes for Football Data from football-data.co.uk.

All data is in csv format, ready for use within standard spreadsheet applications. Please note that some abbreviations are no longer in use (in particular odds from specific bookmakers no longer used) and refer to data collected in earlier seasons. For a current list of what bookmakers are included in the dataset please visit http://www.football-data.co.uk/matches.php

Key to results data:

Div = League Division Date = Match Date (dd/mm/yy) Time = Time of match kick off HomeTeam = Home Team AwayTeam = Away Team FTHG and HG = Full Time Home Team Goals FTAG and AG = Full Time Away Team Goals FTR and Res = Full Time Result (H=Home Win, D=Draw, A=Away Win) HTHG = Half Time Home Team Goals HTAG = Half Time Away Team Goals HTR = Half Time Result (H=Home Win, D=Draw, A=Away Win)

Match Statistics (where available) Attendance = Crowd Attendance Referee = Match Referee HS = Home Team Shots AS = Away Team Shots HST = Home Team Shots on Target AST = Away Team Shots on Target HHW = Home Team Hit Woodwork AHW = Away Team Hit Woodwork HC = Home Team Corners AC = Away Team Corners HF = Home Team Fouls Committed AF = Away Team Fouls Committed HFKC = Home Team Free Kicks Conceded AFKC = Away Team Free Kicks Conceded HO = Home Team Offsides AO = Away Team Offsides HY = Home Team Yellow Cards AY = Away Team Yellow Cards HR = Home Team Red Cards AR = Away Team Red Cards HBP = Home Team Bookings Points (10 = yellow, 25 = red) ABP = Away Team Bookings Points (10 = yellow, 25 = red)

Note that Free Kicks Conceeded includes fouls, offsides and any other offense commmitted and will always be equal to or higher than the number of fouls. Fouls make up the vast majority of Free Kicks Conceded. Free Kicks Conceded are shown when specific data on Fouls are not available (France 2nd, Belgium 1st and Greece 1st divisions).

Note also that English and Scottish yellow cards do not include the initial yellow card when a second is shown to a player converting it into a red, but this is included as a yellow (plus red) for European games.

Key to 1X2 (match) betting odds data:

B365H = Bet365 home win odds B365D = Bet365 draw odds B365A = Bet365 away win odds BSH = Blue Square home win odds BSD = Blue Square draw odds BSA = Blue Square away win odds BWH = Bet&Win home win odds BWD = Bet&Win draw odds BWA = Bet&Win away win odds GBH = Gamebookers home win odds GBD = Gamebookers draw odds GBA = Gamebookers away win odds IWH = Interwetten home win odds IWD = Interwetten draw odds IWA = Interwetten away win odds LBH = Ladbrokes home win odds LBD = Ladbrokes draw odds LBA = Ladbrokes away win odds PSH and PH = Pinnacle home win odds PSD and PD = Pinnacle draw odds PSA and PA = Pinnacle away win odds SOH = Sporting Odds home win odds SOD = Sporting Odds draw odds SOA = Sporting Odds away win odds SBH = Sportingbet home win odds SBD = Sportingbet draw odds SBA = Sportingbet away win odds SJH = Stan James home win odds SJD = Stan James draw odds SJA = Stan James away win odds SYH = Stanleybet home win odds SYD = Stanleybet draw odds SYA = Stanleybet away win odds VCH = VC Bet home win odds VCD = VC Bet draw odds VCA = VC Bet away win odds WHH = William Hill home win odds WHD = William Hill draw odds WHA = William Hill away win odds

Bb1X2 = Number of BetBrain bookmakers used to calculate match odds averages and maximums BbMxH = Betbrain maximum home win odds BbAvH = Betbrain average home win odds BbMxD = Betbrain maximum draw odds BbAvD = Betbrain average draw win odds BbMxA = Betbrain maximum away win odds BbAvA = Betbrain average away win odds

MaxH = Market maximum home win odds MaxD = Market maximum draw win odds MaxA = Market maximum away win odds AvgH = Market average home win odds AvgD = Market average draw win odds AvgA = Market average away win odds

Key to total goals betting odds:

BbOU = Number of BetBrain bookmakers used to calculate over/under 2.5 goals (total goals) averages and maximums BbMx>2.5 = Betbrain maximum over 2.5 goals BbAv>2.5 = Betbrain average over 2.5 goals BbMx<2.5 = Betbrain maximum under 2.5 goals BbAv<2.5 = Betbrain average under 2.5 goals

GB>2.5 = Gamebookers over 2.5 goals GB<2.5 = Gamebookers under 2.5 goals B365>2.5 = Bet365 over 2.5 goals B365<2.5 = Bet365 under 2.5 goals P>2.5 = Pinnacle over 2.5 goals P<2.5 = Pinnacle under 2.5 goals Max>2.5 = Market maximum over 2.5 goals Max<2.5 = Market maximum under 2.5 goals Avg>2.5 = Market average over 2.5 goals Avg<2.5 = Market average under 2.5 goals

Key to Asian handicap betting odds:

BbAH = Number of BetBrain bookmakers used to Asian handicap averages and maximums BbAHh = Betbrain size of handicap (home team) AHh = Market size of handicap (home team) (since 2019/2020) BbMxAHH = Betbrain maximum Asian han...

Explore the epic showdown between Messi and Cristiano: goals, assists, averages, and more. A deep dive into the stats that define their legacy.

The statistic shows the average matchday attendance of football matches in United Kingdom (UK) from 2015 to 2020, by league. In the 2019/2020 season, matches of Premier League had average attendance of approximately 39.5 thousand people.

This statistic presents the average stadium capacity of professional football matches of The Premier League in England from 2010 to 2017. In 2017, the average stadium of The Premier League had the capacity of 37 thousand spectators.

In the 2012/13 season, the average in-play time of English Premier League matches was **** percent. By 2022/23, this had dropped to **** percent, meaning that fans may have seen a slight increase in stoppages and off-the-ball time wasting.

This statistic shows the number of television viewers of the English Premier League from the 2010/11 season to the 2016/17* season, by broadcaster. After starting this period with ***** times the TV audience of rivals BT, Sky's share of the Premier League's viewership has fallen in the past years, to end the period with *** million viewers compared to BT's *** million.

*************** received the highest share of Premier League broadcasting revenue in 2023/24, earning over *** million British pounds. Meanwhile, at the bottom of the table, **************** received ***** million British pounds in broadcasting payments.

Manchester City generated more revenue than any other Premier League club in 2023/24, with the Citizens' total revenue amounting to 719 million British pounds. This was more than five times the revenue of the league's lowest-earning club. Meanwhile, Manchester United's revenue totaled 662 million British pounds.

The English Premier League (EPL) is widely regarded as one of the most exciting and popular soccer leagues in the world. During the 2020/21 season, the Premier League average 414 thousand viewers on the NBC Sports networks, down 10 percent of the previous season's figure.

In 2022/23, the Premier League had an aggregate attendance of around 15.3 million - nearly five million more than the EFL Championship. In terms of average attendance, England's top-tier ranked behind only the Bundesliga in the Big Five leagues.

In 2023/24, the combined broadcasting revenue of all Premier League clubs amounted to nearly *** billion British pounds, representing a slight increase on the previous season. Meanwhile, revenue from matchdays totaled *** million British pounds.

This statistic presents the average number of stadium attendees of professional football matches of The Premier League in England from 2010 to 2017. In 2017, the average number of people who attended matches of The Premier League amounted to 35.8 thousand people.