How market values of football players coorelate to their

statistics in a certain leauge.

Number of pages: 10

1
INTRODUCION

For many years, my personal intrest was focused around football. This game has
been known worldwide since 1983 when the game as such was invented in England.
Since that time it has changed dramatically - starting from change of rules to a
creation of what so called large scale market of players. In 1895, the first transfer
took place in England when Jack Hillman was transfered from Burnley to Everton for
£200. If compared to the biggest tranfer in history, it is relativly nothing as in 2017 the
transfer value was 933 203 times higher when Neymar was transfered from
Barcelona F.C to PSG for €222 million (£186,640 million) Since then I started putting
my intrest into this topic and I was trying to find the reason for such a high values.

METHODOLOGY

In my opinion, the value of the player might depend on a certain league, goals and
assists. For my analysis I will experiment the season 2020/2021 in the top leagues in
Europe. Ligue 1 is a french league, Premier league - English, Serie A - Italian, La
Liga - Spanish, Bundesliga - German. Each of them are top football leagues in each
country and even more, they represent the top league in the whole Europe.

In my research I will use secondary resources such as football statistics websites,

which will give me in-depth information about the statistics and values of the players.
The sources which I will use are reliable as they are uptaded continously 24/7 which
suggests that the sources that are up to date and can be used effectively for my
analysis. As for my tools, I will use mathematical tools such as box-whiskers plot,
scatter diagrams and chi-square test. Box and whisker plot will help me to analyze
1-variable statistics. In scatter diagram, I will be looking for outliers and pearson
correlation coafficient. Chi-square will be used for to see if there is any dependency
between the value of the player and league.

For my analysis, I have created a table of 15 players from 5 diiferent leagues which
consit of market values, goals and assists.

2
LIMITATIONS

One of the biggest obstacles while wiriting the essay was finding the appropriate data and
statistics to which I could compare the market values of the players. Another problem
regarding the essay writing, was time as I started it relativeley late in comparison to other
essays that I have written.

Figure 1
Market values, goals and assists of players

3
4
(Transfermarkt, 2022)

5
INVESTIGATION

Copmaring market values of players

In this section I will compare the market values of all players from each league. The
aim of this process is to find any exeptions in the values meaning is any player out of
the normal range. I have analyzed these tables by using a box-whisker plot.

Figure 2
Box-whisker plot of market values

By looking at the graphs, it is visible that the minimal value is 28 million and the
maximum value 160 million. It suggests that the range of the values is broad and the
values are very high in the football market. In comparison to Portogues league, the
highest value is €38 milion (transfermarkt, 2022). By looking purely at the quantative
data, it shows that the players are from the top shelf and are being the best. Mean of
values is €65.1169 million which is not very far from the median which is equal to
€65 million. It suggests that the values are valid and there is no big difference
between the two. From box-whisker-plot I read that the lower quartile is at level of
€50 million and the upper quartile at €80 million. From those values I calculated
lower and upper outliers by using this formula: Q1 - (1.5 x IQR), for lower outlier and
Q3 + (1.5 x IQR) for upper outlier. This has helped to analyze if any of the players is
going out of the normal zone. The results for lower oulier is 5 and for upper outlier is
125. By looking at the data in Graph 1, we can see that there are two players, which
go outside the box. Erling Halland and Kylian Mbappe have market value of €150
and €160 million. On the other hand, none of the players go below the level of lower
outlier, due to the fact that I chose players which are worth more than €28 million.

6
Comparing market values of players to their goals

Figure 3
Scatter diagram of market and goals

The diagram was created using an Excel spreadsheet and I used a linear model in
order to find R. According to the scatter diagram, which I have created, the
correlation between the values and goals is very weak.

2
𝑅 is equal to 0.193 as visible in the graph and R is equal to 0.439. It suggests that
Pearsons correlation coafficient is waek and the values are not connected to the goal
statistics of the players. On the graph, it is visible that the scatter plot shows no
relationship as the points do not create any visible line. It can also be read that 4
players are far away from the regression line, which suggests the value might not be
dependent on goals statistics.

7
Comparing values of players in certain league
Purpose of this section is to see if there is any dependency on value and the league.
In order to correctly analyze the values, I created the table of players which dvides
values and leagues into sections. I took the number of players from each league
which have value lower than €65 and higher than €65. I decide to devide it in such a
way as 65 is mean of the market values

Figure 4

I assume that X1 means the laegues and X2 is equavalent to the market value of the
players. By using this graph, I utilized the chi-squared test to see if there is any
dependency. Firstly, I assumed a H0, which suggests that the values are not
dependent, X1 ≠ X2. Later I also assumed an H1 which says that values are
dependent between each other, X1 = X2. By using GDC have done chi square test
and the results are visible in the graph below
Figure 5
Chi-squared test

Figure 6
Chi-sqaured distribution table

(Stackexchange, 2022)

8
In this case deegrees of freedom = 4 and χ2 = 27.654-. When we look at the
distribution table graph, the value of χ2 is bigger than the chi-square value at
significance level of .05. Additionally to make a final decision I used p-value, p < 0.05
that is why I rejected H0 and accepted H1.

Conclusion

In my results findings, firstly I found that the market values of players is starting from
€28 million and finishing at €160 million. It suggests that the values are high and that
is why I compared it to the number of goals that players scored in season
2020/2021. From the scatter diagram, it is visible that the number of goals scored
does not always influence the market value directly. When looking at graph 1, it is
visible that Lionel Messi scored the most goals in a season , when he was still
playing in La liga league, and his market value is relatively low compared to Kylian
Mbappe, who scored only 3 goals less - a difference of €100 million. Similar situation
happens in Bundesliga, Robert Lewandowski - 41 goals, Erling Haaland - 27 and the
difference in value is the same, €100 million. The reason for such differences might
not be influenced only by goals, but also by other factors such as age, physical
performance, and others which I will discuss later.

Secondly, by using the chi-squared test, I looked to see if there is any correlation
between the market value of the players and their league. It turns out that there is
one. Not only do market values differ, but the level of a particular league is also a
factor.

Figure 7
Top Five leagues in Europe

(Transfermarkt, 2022)

9
Looking at the ranking of best leagues, it is being said that the Premier league is the
most prestigous and being concidered as the best league - looking at the quantative
information, it confirms this thesis. Other ones are sticking on the top and the reason
why the market value especailly depends on the league, is firstly the level of the
league which not always can be measured by quantative information. Secondly, the
value of clubs as such may differ from league to league. For example the top club in
the Bundesliga - Bayern Munich is worth 1.47 billion U.S. dollars in 2019 while the
top club in Ligue1 - PSG is worth $2.5bn reading from 2019 statistics (Lange D.,
2022). The difference is visible and that is why some clubs might be able to have
more expensive players than other ones, which has a general influence on lueague
as such.

10
BIBLIOGRAPHY

Flashscore, 2022, Piłka Nożna, Flashcore, retrieved January 2022 from

https://www.flashscore.pl/pilka-nozna/anglia/premier-league/

Lange D., 2022, FC Bayern Munich brand/team value from 2011 to 2021, Statista,
retrieved March 2022 from
https://www.statista.com/statistics/267282/fc-bayern-munich-brand-team-value/

Stackexchange, 2022, Chi-square Test with High Sample Size and Unbalanced
Data, stachexchange, retrieved March 2022 from
https://stats.stackexchange.com/questions/440371/chi-square-test-with-high-sample-
size-and-unbalanced-data

Transfermarkt, 2022, Wartości Rynkowe, Transfermarkt, retrieved January 2022

from
https://www.transfermarkt.pl/spieler-statistik/marktwertaenderungen/marktwertetop

Transfermarkt, 2022, WSPÓŁCZYNNIK UEFA, Transfermarkt, retrieved March

2022 from
https://www.transfermarkt.pl/statistik/5jahreswertung

11
12