logo Football Observatory  
ENG |

Spanish clubs outrank rivals in shooting efficiency

The top ranked teams in the remaining big-5 leagues are Olympique Lyonnais (6.7 shots per goal, 5th place), Eintracht Frankfurt (6.9, 7th), Chelsea (7.0, 10th) and Palermo (7.5, 12th).

Conversely, at the bottom end of the table are Hambourg (22.9), Aston Villa (20.9), Torino (17.1), Granada (16.9) and LOSC Lille (15.2). These figures both reflect a lack of efficiency and the difficulty in taking up a favourable position to shoot. As a matter of example, Aston Villa scored only two goals out of 109 attempts in second halves!

All these statistics were calculated from data provided by our partners Opta Pro. They are to be found here.

Real Madrid heads the table for the percentage of time leading

Juventus also led for a majority of domestic league minutes (52.0%). This figure was below 40% for Paris St-Germain (39.8%), Barcelona (39.6%) and Manchester City 38.8%). The record low at big-5 league level was recorded for Sunderland (10.3%). This does not augur well for the Black Cats.

The figures for the 98 big-5 league clubs are to be found here.

First Monthly Report: club instability and its consequences

The report indicates that stability gives clubs a competitive advantage over rival teams, be it on a sporting level (better medium and long-term results) or an economic one (a greater capacity to launch careers of club-trained players and generating revenues through their transfer).

Stability indicators such as the average length of stay of players in their club or the percentage of new signings in the squad thus show their true worth when judging the pertinence of management strategies instigated by club managers.

Well-informed fans of the most unstable teams have good reason to be anxious. In order to protect football from the bad practices of certain managers, to promote training and to increase team competitiveness, it would thus be timely to consider the introduction of a limitation on the number of transfers allowed.

True to its reputation, our research group is at the disposal of football stakeholders to analyse such a scenario. This would notably be about defining more precisely the boundaries of such a limitation on transfers in order to attain the desired goals, without interfering with the free movement of players or provoking a distortion of the market with regard to the upholding of the principle of proportionality.

Issue number 94 of the Big-5 Weekly Post also deals with the subject of club stability by presenting the date of arrival of current squad members for clubs ranked in first and last position in the table of their respective league. This analysis also allows us to highlight the importance of squad stability for sustainable success.

Top transfer values: Messi outranks Ronaldo and Hazard

With the exception of Raheem Sterling, in the top 10 positions of the ranking are only players under contract with teams that are still competing in the Champions League:

  • Lionel Messi (Barcelona) : 220 million €
  • Cristiano Ronaldo (Real Madrid) : 133 million €
  • Eden Hazard (Chelsea) : 99 million €
  • Diego Costa (Chelsea) : 84 million €
  • Paul Pogba (Juventus) : 72 million €
  • Sergio Agüero (Manchester City) : 65 million €
  • Raheem Sterling (Liverpool) : 63 million €
  • Francesc Fàbregas (Chelsea) : 62 million €
  • Alexis Sánchez (Arsenal) : 61 million €
  • Gareth Bale (Real Madrid) : 60 million €

The club with the most representatives in the top 100 of the table is Spanish giant and Champions League title holder Real Madrid:

  • Cristiano Ronaldo: 2nd, 133 million €
  • Gareth Bale: 10th, 60 million €
  • James Rodríguez: 15th, 50 million €
  • Isco Alarcón: 17th, 45 million €
  • Karim Benzema: 21st, 43 million €
  • Toni Kroos: 23rd, 41 million €
  • Luka Modri?: 27th, 37 million €
  • Sergio Ramos: 40th, 33 million €
  • Daniel Carvajal: 68th, 25 million €
  • Raphaël Varane: 87th, 23 million €
  • Marcelo Vieira: 94th, 22 million €

The transfer values are calculated using an exclusive algorithm developed on the basis of over 1,500 fee paying transfers occurred since 2009. The variables included in our exclusive econometric model refer to player performances (matches, goals, dribbles, etc.), their characteristics (age, position, contract duration, etc.), as well as competition level and results achieved by their teams (clubs and national sides).

For more information, you are kindly invited to read chapter four of this paper on the CIES Football Observatory’s approach for sustainable success.

A Digital Atlas to celebrate the Football Observatory’s 10th birthday

This tool is freely available online from our website. It enables anyone to analyse the state of the game in 31 European countries and to follow its progress from year to year. The indicators offered study the characteristics of players from over 500 first division clubs from the perspective of age, height, mobility, training, origin, etc.

By consulting the Atlas, you can notably learn that there is almost 10cm of difference between the club with the tallest players, Diósgyör (186.3cm on average), and that with the shortest ones, Barcelona (177.3).

The Atlas also allows for comparisons between leagues. Regarding age, for example, it shows that the Italian Serie A regroups the oldest footballers (27.3 years on average), while the Dutch Erdivisie is made up of the youngest ones (24.2).

Issue number 92 of the Big-5 Weekly Post helps situate the five major championships in the European context from the point of view of the average age of squad members. It also presents the average age for the 98 big-5 league clubs. The Post notably shows that more than 5 years separate the youngest club, Valencia, from that which gathers the oldest players, Atalanta.

From January onwards, the Football observatory’s academic team will also offer monthly reports instead of our annual paying publications. Like our Digital Atlas, these will also be freely accessible from our website. The first issue will cover the topic of club instability. Moreover, we shall continue the publication of the Big-5 Weekly Post. We are open to sponsorship proposals.

In autumn 2015, we also plan to publish a book destined for the general reader summarising the key points garnered from our analyses over the past decade. There are indeed over 20,000 football enthusiasts like yourselves now on our mailing list. We warmly thank your for your interest in all our numerous areas of initiative.

Happy Christmas and best wishes for 2015!

The CIES Football Observatory unveils its exclusive approach for long-term success

Moreover, issue number 91 of the Big-5 Weekly Post presents the results of the analysis of the ASAR ratio for clubs from the five major European championships. The ratio is obtained by dividing the average stay of players in the first team squad of their employer club (the longest, the best) with their age at recruitment (the youngest, the best).

Indeed, the analyses carried out since 2005 show a strong correlation in all countries between this ratio and the results obtained by clubs at the end of the season. It is thus not surprising to note that the five clubs that are currently at the top of the table in their respective league are also in the top 20 positions of the ASAR ranking at European level:

  • Real Madrid (2nd both at big-5 league level and Spain)
  • Juventus (7th and 1st in Italy)
  • Bayern Munich (9th and 2nd in Germany)
  • Chelsea (11th and 1st in England)
  • Marseille (19th and 2nd in France)

From an ASAR perspective, the most over-performing teams so far are Sevilla, Genoa and Valencia. Conversely, Real Sociedad, Borussia Dortmund, Everton, Cagliari and Lille clearly under-performed.

Our academic team is at the service of professional clubs to study their squads in greater detail.

Updated transfer value calculator

Thanks to our tool, anyone can obtain an accurate, up to date transfer value for players taking part in the five principle markets of European football: the English Premier League, Spanish Primera Liga, German Bundesliga, Italian Serie A and the French Ligue 1.

Our statistical model is based on almost 1,500 fee paying transfers having taken place at big-5 league club level from 2009 onwards. The correlation between estimated and actually paid sums is greater than 80%. This shows that we have been able to reveal the logics of the transfer market. This allows us to anticipate events and highlight market bias.

Our team is at service of market actors for expert audits for both big-5 league players and professional footballers playing in other championships at worldwide level.
Do not hesitate to access the calculator here.

Best players according to area of play: shining Ronaldo

Our rankings highlight not only the best performing players overall, but also those who performed the best with regard to teammates. This approach allows us to pick out more players who are not yet part of the top clubs. This is particularly useful from a scouting perspective. To limit bias, only footballers having played at least 66% of match time since the start of the season are included in the study.

In absolute terms, the best big-5 league players at present are Cristiano Ronaldo (Real Madrid) both in shooting and chance creation, Eden Hazard (Chelsea) in take on, Xabi Alonso (Bayern Munich) in distribution, Mathieu Coutadeur (Lorient) in recovery and Javi Manquillo (Liverpool) in rigour.

At the top of the relative rankings, we find Charlie Austin (QPR) in shooting, Xabi Prieto (Real Sociedad) in chance creation, Federico Cartabia (Córdoba) in take on, Roberto Trashorras (Rayo Vallecano) in distribution, Alaixys Romao (Marseille) in recovery and Mathieu Bodmer (Nice) in rigour.

Actors from professional clubs can contact us to enquire about our range of products in the domain of performance analysis.

Key performance indicators

  • Shooting: ability to exploit goal opportunities through accurate shooting
  • Chance creation: ability to putting teammates in a favourable position to score
  • Take on: ability to create advantageous situations by successfully challenging opponents
  • Distribution: ability to keep a hold on the game through efficient passing
  • Recovery: ability to minimise opponents’ chances through proficient interception work
  • Rigour: ability to minimise opponents’ chances through robust duelling

Real-Barcelona tops the rankings for the most “expensive” matches

In the other championships, the most “expensive” matches this season were Manchester City-Chelsea (578 million €), Paris St-Germain-Marseille (503 million €), Bayern Munich-Borussia Dortmund (272 million €) and Inter-Napoli (263 million €).

This analysis shows the increasing impact of money at the highest level of European football and the difficulty encountered by the vast majority of teams to reach the performance levels of the richest clubs. This state of affairs leads to more predictable results.

Thus, the five clubs touted by our research team as favourites at the start of the season are either in the first (Bayern Munich, Chelsea) or second place (Paris St-Germain, Barcelone, Rome) of their respective championships.

New record high for expatriate presence in the big-5

The expatriate presence is particularly high in the English Premier League (59.3%. A new record high since 2009 was measured this season in Italy (54.8%) and Spain (38.9%). Expatriates represent 43.5% in Germany and 31.6% in French Ligue 1. Compared to one year ago, the percentage of football migrants went up in all big-5 leagues with the exception of the English Premier League (-1.6%).

The biggest increase since 2009 was registered in Italy: from 42.4% to 54.8% (+12.4%), while the strongest growth since November 1st 2013 was measured in the German Bundesliga: from 38.1 to 43.5% (+5.4%).

For the first time in history, the main exporting country is France: 113 players compared to 112 for Brazil (former leader) and Argentina. The strongest increase since 2009 was recorded for Spain (from 17 to 58 players in foreign big-5 leagues). During last year, the biggest growth was for Serbia (28 to 39, +11) and Belgium (31 to 41, +10).

More data is available on request at football.observatory@cies.ch

Height on the pitch: records since 2009

Since the start of the 2014/15 seasons, the teams having fielded the shortest line-ups per league are Barcelona (176.4 cm), Nice (177.0 cm), Empoli (178.2 cm), Hertha (179.4 cm) and Manchester City (179.8 cm).

Conversely, at the head of the rankings of the tallest formations during current season are Stoke City (188.0 cm), Schalke 04 (186.7 cm), Verona (186.2 cm), Monaco (185.5 cm) and Elche (183.6 cm).

CIES Football Observatory at Sportdata & Performance Forum

Key topics of the Forum also include: creating a global standard for live sport data, low cost performance tools, coach-led analytics, data in science and sport medicine, evidence-based decision making in modern coaching, monetising sports data and video content across multiple platforms, interactive panel discussions, as well as networking activities.

The presentation of the CIES Football Observatory is available on request after the event. Interested parties can contact us at football.observatory@cies.ch

View the dedicated event website here.

For the full event programme and how to be part of it, sponsorship and attendance availability email Edward Abankwa at EdwardA@pinnacle-management.com

Click here to REGISTER ONLINE. You can also ring +44(0)7956 675521 / +44 (0)208 657 4062.

Email Ian Brumpton at ian@pinnacle-management.com with partnership and marketing enquiries.

Barcelona top the charts for producing the most big-5 league players

Out of the 43 big-5 league footballers trained at Barcelona, 13 are still playing for the Catalan club. Only Olympique Lyonnais, Athletic Bilbao and Real Sociedad have a greater number of club-trained players in their current squad. To be able to rely on homegrown talent in this way has provided a key competitive advantage for Barcelona over the last decade. Munir El Haddadi and Sandro Ramírez are the latest examples of this strategy.

Barcelona is also at the top of the table of clubs having trained the most footballers playing for other big-5 league teams (30), ahead of Real Madrid (26) and Manchester United (24). This finding not only highlights the quality of training provided by these top teams, but also demonstrates the difficulty for youth academy players to breakthrough into the first team squad of the most competitive club. This is unlikely to change in the near future, irrespective of the legal framework in force.

River Plate heads the ranking for clubs not participating in the big-5 leagues. The Argentinean team has trained 17 players currently employed by teams in the five major European championships, which is two more than Le Havre, Ajax and Munich 1860.

Issue 86 of the CIES Football Observatory Big-5 Weekly Post also shows that the percentage of club-trained players went down for the fourth consecutive season and has now reached a new record low: 17.2% of squad members. This percentage varies between 24.6% in France and 9.6% in Italy. To be considered club-trained, a player must have played for at least three seasons between the ages of 15 and 21 for his employer club (UEFA criterion).

More information is available on request at football.observatory@cies.ch

Manchester United second youngest line-up in Europe

At the opposite end of the table, Manchester City fielded the third oldest line-up at big-5 league level: 29.3 years (highest figure in the English Premier League). This suggests that fresh blood is needed to maintain top level performance standards over the long term. Only Verona (31.0) and Torino (30.3) fielded more seasoned players than Manchester City.

Since 2008/09, the youngest line-up in the big-5 was fielded by Nice: 22.1 years on average against Evian on 10.05.2014. Conversely, the team having fielded the oldest line-up is Milan AC: 33.3 years against Catania on 03.05.2009.

The data on the last match day for all big-5 league clubs is available in the 85th Weekly Post. For more information, please contact us at football.observatory@cies.ch

New “ranking by minute” statistics: there is hope for Manchester United

Since the beginning of the English Premier League season, Manchester United achieved the third highest score in points achieved by minute played, ahead of Manchester City (5th). This suggests that Louis van Gaal’s team should be able to further improve its results in the near future. The gap measured with respect to Chelsea is indeed not as significant as at the end of the matches played so far (-0.42 compared to -1.14). Conversely, Arsenal is ranked only 16th.

At individual player level, in the English Premier League Marcos Rojo (Manchester United) has achieved the most points by minute played: 2.39 on average. The lowest figure among footballers who have played at least half of total match minutes was recorder for Niko Kranj?ar (QPR, 0.31).

Players with the most impressive statistics in the remaining big-5 European leagues are Pablo Piatti (Valencia), Dimitri Payet (Marseille), Arjen Robben (Bayern) and Angelo Ogbonna (Juventus). In contrast, the least successful players so far this season besides Niko Kranj?ar are Xabi Prieto (Real Sociedad), Mathieu Peybernes (Bastia), Julian Draxler (Schalke 04) and Daniele Conti (Cagliari).

The minute by minute rankings will be updated after each championship round. They will be available throughout the season on the CIES Football Observatory website.

Cesc Fàbregas makes the difference

The best performing players in the other positions are Giorgio Chiellini (centre back), Jordi Alba (full back), Xabi Alonso (central/defensive midfielder) and André-Pierre Gignac (forward). Only footballers who played at least 60% of championship minutes are included in the analysis.

Many footballers who did not yet represent a national A-team are to be found in the top 15 positions of the rankings: Robin Knoche (Wolfsburg), José Fonte (Southampton), Antonio Balzano (Cagliari), Layvin Kurzawa (Monaco), Juan Bernat (Bayern Munich), Roberto Pereyra (Juventus), James Ward-Prowse (Southampton), Manuel Nolito (Celta Vigo) and Wissam Ben Yedder (Toulouse). All these footballers deserve a chance to play for their country of origin.

Players are ranked according to their production and efficiency in six areas of the game (see below). The six key performance skills taken into account are weighted differently per position according to their impact on club results. Consequently, for example, rigour is more important for centre backs than for forwards. For more comparability, the rankings also consider the differences in the intensity and style of play between leagues.

CIES Football Observatory key performance skills

  • Shooting: ability to take advantage of goal opportunities through accurate shooting
  • Chance creation: ability to put teammates in good position to score
  • Take on: ability to create dangerous situations by successfully challenging opponents
  • Distribution: ability to hold the grip on the game through efficient passing
  • Recovery: ability to minimise goal opportunities for opponents through proficient ball recovery
  • Rigour: ability to minimise goal opportunities for opponents by efficiently blocking their actions

Records broken during 2014 summer transfer window

A new record high was also measured with regard to the percentage of transfer fees invested with 49% of the ten most active clubs now involved in this activity. In addition, the level of transfer fee compensation paid by clubs in the most active league – the English Premier League - reached a new peak of 45% of total transfer expenditure.

All of these findings show the increasing concentration of fee paying transfer activity at the very top of the football pyramid. At the same time, and as shown in issue 81 of the Big-5 Weekly Post, an increasing proportion of fees invested do also benefit other big-5 league clubs (72% during the last transfer window compared to 67% for the five previous years). These figures highlight the ongoing transformation of the transfer market into a zero-sum game involving a closed circle of top level teams and with significant amounts of money made by dominant intermediaries and/or third-parties.

The club that invested the most on the transfer market since 2008 was Real Madrid €750 million, closely followed by Manchester City (€733 million).

CIES Football Observatory online database updated

Data refers to the indicators of average age, average height, % of club-trained players, % of internationals, % of expatriates and average stay.

More detailed data relating to club level analysis is presented in our annual publications: Annual Review (big-5 leagues) and Demographic Study (31 top division leagues in Europe).

We also kindly remind you that an online tool to assess the transfer value of big-5 league players is available here.

Thank you very much for your interest!

Transfer analysis: big-5 leagues as closed system

The team who was able to generate the most incomes by transferring players to big-5 league clubs is Southampton (120.7 million €). The English team outranks Real Madrid (112 million €) and Chelsea (102.4 million €). The only non big-5 league clubs in the top 10 of the ranking is Porto (83.8 million €). However, a big part of transfer incomes of the Portuguese team was finally due to third-parties that held shares on the economic rights of players transferred.

Among the 46 teams that earned at least 15 million €, only six are not part of big-5 leagues: Benfica, Ajax, Sporting Lisbon, Feyenoord, Dinamo Kiev and Salzburg. More information is available in the 81st issue of the Big-5 Weekly Post.

CIES Football Observatory’s big-5 league predictions

This year’s predictions take into account the following areas: players’ status, squad cohesion and club ambition. The indicator of players’ status considers minutes played by squad members during the previous season, as well as overall results achieved by their employer club. Squad cohesion refers to the percentage of domestic league minutes played in 2013/14 by footballers who are still at the club, as well as the average number of seasons played by current squad members at the employer team. Finally, club ambition includes the average length of contract duration of first team players, as well as the ratio between the status of new signings and that of footballers released by the club during the last transfer window.

Analysis of the above criteria by the CIES Football Observatory academic team indicates that Chelsea, Barcelona, Paris St-Germain, Bayern Munich and Rome are in the most favourable position to win the title in their respective leagues. Our data also suggests that the runners-up will be Manchester City, Real Madrid, LOSC Lille, Borussia Dortmund and Juventus. Among promoted teams, Leicester City, Burnley, Deportivo La Coruña, Metz and Cologne are well placed to achieve good results. In contrast to this, the following clubs are forecast to struggle in achieving satisfactory results: Crystal Palace, Sunderland, Almería, Nice, Eintracht Frankfurt and Chievo Verona.

Enjoy the season and may the best teams win!

Transfer market inflation continues at big-5 league level

Our ‘right price’ has been calculated as the average between a player’s transfer value both before and after the deal. Significantly, the latter figure also considers the new contract duration of the footballer recruited. Both values have been calculated according to the exclusive CIES Football Observatory statistical model which incorporates nearly 1,500 fee paying transfers completed since 2009 (see below).

Our new data analysis has revealed that overall clubs paid on average 16% more than they invested in the five previous years for players with similar characteristics (add-ons included). This confirms the ongoing inflation trend of the transfer market at the top end of the football pyramid. The inflation trend is mainly due to the sums spent by a handful of wealthy clubs. This was notably the case for Manchester United with regard to Angel Di María (+30 million € between the right price and fee reported), for Paris St-Germain for David Luiz signing (+ 29 million €) and for Real Madrid in the case of James Rodríguez (+25 million €).

At the opposite end of the table, our analysis shows that the existence of buy-out clauses and/or non-sporting related issues allowed Barcelona to reduce their offer for Luis Suárez (-12 million € between the right price and money invested). The same holds true for Chelsea with Diego Costa’s signing (-10 million €). In absolute terms, Mario Balotelli was the most under-paid player (-16 million €).

The full picture for all players transferred for a fee of at least 10 million € (including add-ons) is presented in the 79th edition of the Big-5 Weekly Post. We also kindly remind you that our unique and simple online calculator to estimate the current transfer value of big-5 league players is freely accessible here.

The CIES Football Observatory stays at your disposal for any query or more information. Enjoy the season!

CIES Football Observatory’s statistical model

The CIES Football Observatory academic team has developed an exclusive statistical model to assess the transfer value of players and predict transfer fees. This powerful econometrical model is based on the in-depth analysis of a sample of nearly 1,500 players transferred for a fee from big-5 league teams since the summer transfer window preceding the 2009/10 season. The statistical model developed includes a multitude of variables relating to the following areas: age, length of remaining contract, position, player performance at club level for the most recent and previous seasons, results of the clubs for which players are employed, as well as players’ international experience and results of national team represented.

Groundbreaking player transfer value calculator

The new transfer value calculator uses an algorithm created on the basis of nearly 1,500 fee paying transfers occurred during the previous five years at big-5 league clubs. The exclusive algorithm will continue to be updated after every transfer window to take into account the latest market trends.

More information on the statistical model developed by the CIES Football Observatory academic team and detailed analysis of current big-5 league players is available in the latest edition of the CIES Football Observatory Annual Review. A free excerpt of this publication can be downloaded here.

The scientific analysis of player transfer values perfectly complements the two other main CIES Football Observatory research areas: squad composition and pitch performance. All of these areas of study are of high importance to clubs as they are able to assist in raising performance levels from both a sporting and economic perspective.

Moreover, the transfer value calculator will help to increase the level of transparency in the beautiful game as all stakeholders – from club officials to the general public – will be able to compare actual fees paid to those estimated as the fair transfer value by the calculator.

The CIES Football Observatory academic team is at your disposal for more detailed analysis. Its services notably include the elaboration of scenarios for future transfer values. For more information about our products and services, please contact us at football.observatory@cies.ch

Player Development Ranking for the 2014 World Cup

The ranking takes into account the number of World Cup participants trained per club, as well as the number of league matches played by footballers per team up to the age of 23. This age limit has been set according to FIFA regulations on training indemnities. Training clubs have been defined according to the UEFA definition of at least three years at the club between the ages of 15 and 21.

At the top of the table is Feyenoord Rotterdam. The Dutch team trained 9 players participating in the World Cup. Two other clubs trained as many footballers as Feyenoord: Barcelona and Deportivo Saprissa. However, World Cup participants played 882 matches at Feyenoord until the age of 23, compared to 783 at Barcelona (not including B-team matches) and 445 at Deportivo Saprissa.

At national association level, France tops the table ahead of England and Germany. This is mainly due to the high number of footballers trained in France playing for other national A-teams, notably Africans. This finding also reflects the high employment rate of U23 players in French clubs. In total, 821 clubs in 59 national associations have contributed to the development of World Cup players. The full list of clubs is available here.

We also remind you that the brand new edition of our Annual Review on the five major European leagues is available for sale from the CIES online shop. A free excerpt is available here.

The Football Observatory reveals how to select a winning World Cup squad

The Observatory academic team have selected the following indicators for assessing the potential of the 2014 World Cup participants:

  • League matches during career
  • League matches for the two years preceding the event
  • League goals during career
  • League goals for the two years preceding the event
  • National A-team matches for the two years preceding the event
  • World Cup matches during career

The results for each national team are illustrated through a diagram with green or red boxes. A green box means that the value measured for the team is superior to the minimal figure observed for the last four World Cup finalists for the 11 players with the highest values. If this is not the case, the box is red. The greater the proportion of green boxes, the higher the probability of reaching the final stages of the competition. In the case of a tie, we took into account the number of World Cup matches of national teams during the last four editions.

This approach has also allowed us to predict a hypothetical scenario for the outcome of the competition, suggesting that Spain will beat Brazil in the final, with Argentina in 3rd place and France in 4th position. The full analysis is presented in the World Cup Scenario report.

New edition of the CIES Football Observatory Annual Review

The CIES Football Observatory Annual Review is an essential tool for all forward-thinking teams to benchmark themselves against rivals and implement the most suitable strategies for sustainable success. The analysis provided is also of great interest for professionals involved in the transfer market operations, media and the general public. An excerpt of this 96-page landmark publication can be downloaded here. The paperback report is available for purchase from the CIES online shop.

As of June 2013, Lionel Messi is the big-5 league footballer with the highest market value. According to the CIES Football Observatory’s exclusive econometrical model, the estimated value of Barcelona’s player is almost twice as high as for Cristiano Ronaldo: € 216 versus € 114 million. This is mainly related to the younger age of the Argentinian prodigy. However, contrary to Ronaldo’s market value, Messi’s decreased during last year due to lower performance levels (- € 19 million).

The comparison between transfer fees paid by clubs to sign new players at the start or during the 2013/14 season and their current market value allows us to assess the best and worst recruitments from a financial perspective. The highest negative gap was recorded for Gareth Bale (€ -37 million). This finding reflects the fact that Real Madrid clearly paid over the odds to convince Tottenham to release the player. In 2013/14, the Castilian team broke the record for the most expensive line-ups fielded with an average transfer expenditure per player of € 32.3 million in their Champions League winning campaign.

Results achieved by Spanish title winners Atlético Madrid were outstanding with an average transfer fee of “only” about € 4.3 million per player on the pitch. From an economic perspective, it is also worth highlighting results obtained by some of the clubs that were among those that spent the least to sign players fielded such as Crystal Palace in England, Elche in Spain, Guingamp in France, Augsburg in Germany and Hellas Verona in Italy. This publication will allow you to understand what factors permitted these clubs to over-perform.

While not directly focused on national teams, the findings presented in this publication can also be interpreted in the light of the upcoming World Cup. For example, our analysis of the pitch performance of players suggests that it will be very difficult to score against Brazil. Indeed, Thiago Silva, Dante, Dani Alves, Maxwell, Marcelo and Maicon are among the best performing players in their respective position. Felipe Scolari could even afford not to select other very productive and efficient defenders such as Rafinha and Luis Filipe.

However, our analysis also shows that Brazilian players are less represented among the best performing midfielders and forwards. Spain appears thus to have a more complete squad with at least one player in the top 10 of the rankings for each position. Of course, the outstanding talent of the four best performing forwards, Cristiano Ronaldo (Portugal), Luis Suárez (Uruguay), Ciro Immobile (Italy) and Lionel Messi (Argentina), could prove to be a lethal weapon for opponents. Within one week, the CIES Football Observatory will also disclose its World Cup predictions using a brand new methodology.

Assessment of 2013/2014 big-5 league predictions

The average gap between the league position estimated and that finally achieved at big-5 level was 3.3 positions: from 4.2 in Germany to only 2.3 in England. The ranking gap was higher by 3 for less than two thirds of teams. More information is available here.

Amongst the teams that obtained better results than expected were Torino, Augsburg and Liverpool. In the under-performing team category, we find clubs such as Real Betis, Milan and Manchester United.

This season, the Football Observatory academic team were able to predict 12 of the 15 teams that finished in the top three positions in their respective league: 3 out of 3 in Spain and Germany, followed by 2 out of 3 in Italy (Juventus and Napoli), France (Paris St-Germain and Monaco) and England (Manchester City and Chelsea).

In relation to the actual champions, we were only able to predict 2 out of 5 (Juventus and Paris St-Germain), while Borussia Dortmund finished in second place, Champions League winners Real Madrid in third position and Manchester United ranked only 7th. Since our first predictions in 2010/11, we have correctly forecast 9 champions out of a possible 20 (45%).

Updated website launched with team of the season line-ups

Each starting 11 has been selected by taking into account players’ production and efficiency in six key areas of the game. Only footballers who played at least 60% of minutes in their respective league were taken into account. Goalkeepers have been selected with the highest % of saves.

Each team is made up of a goalkeeper, two centre backs, a left full back, a right full back, two central or defensive midfielders, two offensive midfielders and two forwards as per the 4-4-2 formation.

In the big-5 league line-up are two Juventus (Buffon and Chiellini) and Real Madrid (Di María and Ronaldo) players, as well as one player each for Barcelona (Dani Alves), Bayern Munich (Alaba), Manchester City (David Silva), Paris St-Germain (Thiago Silva), Liverpool (Suárez), Atlético Madrid (Gabi) and Roma (Pjani?). Among these players, only Gabi and Alaba will not participate in the 2014 World Cup.

All the rankings are available on the CIES Football Observatory website, see Big-5 Weekly Post (issue 76). Deeper analysis will be presented in the 9th edition of our Annual Review to be published during the first week of June.

Our metrics measure players’ pitch production and efficiency in six complementary areas of the game.

CIES Football Observatory key performance indicators

  • Shooting: ability to take advantage of goal opportunities through accurate shooting
  • Chance creation: ability to create goal opportunities through efficient passing
  • Take on: ability to create dangerous situations by successfully challenging opponents
  • Distribution: ability to hold the grip on the game through efficient passing
  • Recovery: ability to minimise goal opportunities for opponents through proficient ball recovery
  • Rigour: ability to minimise goal opportunities for opponents by efficiently blocking their actions

Data in each area is weighted according to the impact on club results per position. Consequently, for example, rigour is more important for centre backs than forwards, while the opposite holds true for shooting.

Southampton has the fourth youngest defence in Europe

In England, Fulham fielded the oldest midfielders and forwards, as well as the second most seasoned defenders after West Bromwich Albion. Conversely, Southampton defence was the fourth youngest at European level. All the information is available on the CIES Football Observatory website.

Big-5 Weekly Post on players born in the 1970’s

Crystal Palace goalkeeper Julián Speroni heads the ranking for minutes played by footballers born in the 1970’s in England. In the other leagues, at the top of the tables are Morgan De Sanctis (Roma, Italy), Olivier Sorlin (Evian, France), Esteban Suéarez (Almería, Spain), Raphael Schäfer (Nürnberg) and Zden?k Posp?ch (Mainz, Germany).

All the rankings are to be found on the CIES Football Observatory website.

April rankings of the best big-5 league players

In England, at the top of the tables are players from four clubs: Džeko (Manchester City), Cazorla (Arsenal), Schneiderlin (Southampton), Azpilicueta and Cahill (Chelsea).

In Spain, Aduriz (Athletic Bilbao) was the only best performing footballer who does not play for Real Madrid (Isco and Sergio Ramos) and Barcelona (Busquets and Dani Alves).

In France, PSG is the only club with two players at the top of the ranking: Cavani and Cabaye. Sertic (Bordeaux), Perrin (St-Etienne) and Bedimo (Lyon) head the remaining tables.

In Germany, players from four different clubs are at the top of the rankings: Oli? and de Bruyne (Wolfsburg), Schweinsteiger (Bayern), Spahi? (Leverkuesen) and Posp?ch (Mainz).

In Italy, finally, no club has more than one player at the top of the tables: Immobile (Torino), Hernanes (Inter), Inler (Napoli), Chiellini (Juventus) and Pasqual (Fiorentina).

Our metrics measure players’ pitch production and efficiency in six complementary areas of the game.

CIES Football Observatory key performance indicators

Shooting: ability to take advantage of goal opportunities through accurate shooting

Chance creation: ability to create goal opportunities through efficient passing

Take on: ability to create dangerous situations by successfully challenging opponents

Distribution: ability to hold the grip on the game through efficient passing

Recovery: ability to minimise goal opportunities for opponents through proficient ball recovery

Rigour: ability to minimise goal opportunities for opponents by efficiently blocking their actions

Data in each area are weighted according to the impact on club results per position. Consequently, for example, rigour is more important for centre backs than forwards, while the opposite holds true for shooting.

Big-5 Weekly Post on U20 players

The U20 footballer who played the most league minutes so far is Matthias Ginter (Freiburg), followed by Saúl Ñíguez (Rayo Vallecano) and Luke Shaw (Southampton). Four other U20 footballers played for more than 2,000 minutes: Aymeric Laporte (Athletic Bilbao), Issiaga Sylla (Toulouse), Morgan Sanson (Montpellier) and Hakan Çalhano?lu (Hamburg).

The most fielded U20 player in Italy is Ibrahima Mbaye (Livorno). The tenth most employed U20 player was only fielded for 270 minutes in Italy. This figure is 1,115 in Germany, 1,028 in France, 790 in England and 437 in Spain.

For more information, please go to the CIES Football Observatory website or contact us at football.observatory@cies.ch

Big-5 Weekly Post on shooting production

In England, the highest increase in shots on target per match after 1st of January was observed at Cardiff City, from 2.9 to 3.9 (+34%). Conversely, the greatest decrease was registered for Everton, from 5.8 to 4.7 (-19%). Shooting production went up at Liverpool and Chelsea, while it decreased at Manchester City and Arsenal.

In Spain, shooting production increased the most at Málaga (3.8 to 4.3, +13%). Despite the excellent results achieved, Atlético Madrid took much fewer shots on target after 1st of January than in the first part of the season (6.5 to 4.1, -37%). If this trend continues, the Colchoneros will most probably struggle to win the league title.

In France, the highest negative gap for shots on target was measured at Nice (3.6 to 2.4, -33%). The reverse holds true for Montpellier (3.3 to 4.9, +48%). Paris St-Germain shooting production remained stable: 6.0 shots on target per match (highest figure in France for both first and second semester).

In Germany, the shooting production of Hertha Berlin decreased by 26% (4.7 to 3.5). At the opposite end of the table is Hoffenheim (5.0 to 7.1, + 42%). The number of shots on target per match also increased for Bayern Munich (7.5 to 7.9, +5%).

In Italy, shooting production of the probable future league winner Juventus went heavily down after 1st of January (7.1 to 4.6, -35%). On the contrary, Genoa players took significantly more shots on target in the second part of the season than in the first one: 2.9 to 4.5 (+55%).

At big-5 league level, the ten most productive clubs in terms of shots on target in the second part of the season are Bayern Munich (7.9), Real Madrid (7.7), Hoffenheim (7.1), Liverpool (7.1), Barcelona (6.8), Borussia Dortmund (6.2), Schalke 04 (6.1), Paris St-Germain (6.0), Manchester City (5.9) and Chelsea (5.9). Six of these teams qualified for the quarter of finals of the Champions League and three will play the semi-finals.

More data are to be found on our website. For more information, please contact us at football.observatory@cies.ch.

Big-5 Weekly Post on competitive (un)balance

During current season, the three top ranked teams per league achieved 75% of possible points. This figure is 5% greater than the previous highest levels measured in 2011/12 and 2012/13. A new record high was measured in Spain (80.9%), Italy (77.8%), Germany (73.2%) and France (71.2%). In England, the current level (74.2%) is “only” the second highest since 2008/09. Our analysis clearly confirms the growing competitive unbalance in the richest European leagues.

For more information, please contact us at football.observartory@cies.ch

Best players for the month of March

In the English Premier League, at the top of the rankings are two Liverpool (Suárez, Coutinho) and three Swansea players (Britton, Chico, Rangel). This finding suggests that the Welsh club should be able to further improve its results until the end of the season.

In Spain, Barcelona beats Real Madrid 3-2: Xavi, Alves and Mascherano versus Ronaldo and Di Maria. Only three footballers in the top 3 of their respective position do not play for the three best ranked teams: Rakiti? (Sevilla), Angulo (Granada) and Galvez (Rayo Vallecano).

In France, PSG players head three rankings out of five: Ibrahimovi?, Thiago Motta and Thiago Silva. Two very promising footballers top the table for offensive midfielders (Benjamin André) and full backs (Serge Aurier). Both have the ability to play in more competitive teams or leagues.

In Germany, Bayern Munich players are ranked first in four tables out of five: Kroos, Schweinsteiger, Rafinha and J. Boateng. However, the best performing forward was Anthony Modeste, who also scored against the Bavarian side during last week-end.

In the Italian Serie A, at the top of the rankings for forwards is the best scorer of Torino Ciro Immobile. Another Torino player, Giuseppe Vives, tops the table for central and defensive midfielders. Barrientos (Catania), Jonathan (Inter) and Paletta (Parma) head the remaining tables.

Our metrics measure players’ pitch production and efficiency in six complementary areas of the game.

CIES Football Observatory key performance indicators

Shooting: ability to take advantage of goal opportunities through accurate shooting

Chance creation: ability to create goal opportunities through efficient passing

Take on: ability to create dangerous situations by successfully challenging opponents

Distribution: ability to hold the grip on the game through efficient passing

Recovery: ability to minimise goal opportunities for opponents through proficient ball recovery

Rigour: ability to minimise goal opportunities for opponents by efficiently blocking their actions

Data in each area are weighted according to the impact on club results per position. Consequently, for example, rigour is more important for centre backs than forwards, while the opposite holds true for shooting.

Big-5 Weekly Post on transfer expenditure

This fidning demonstrates the existence of a strong big-5 league club internal market. It also shows that the redistribution effect of the transfer system throughout the entire football pyramid is limited.

The club which generated the most income from the transfer of squad members to big-5 league clubs was Tottenham Hotspur (322 million €), closely followed by Spanish giant Real Madrid (304 million). Only five non big-5 league teams are among the 30 clubs that generated the most transfer incomes since 2008/09: Porto (3rd), Palermo (12th), Benfica (16th), Ajax (26th) and Portsmouth (28th). Conversely, many rich teams are ranked in the top 20 positions of the table: Liverpool (4th), Milan (5th), Inter Milan (6th), Arsenal (8th), Roma (9th), Barcelona (15th), Manchester United (17th), Chelsea (19th) and Manchester City (20th).

While the majority of the above-mentioned clubs also spent significant fees to sign new players, our findings show that the current transfer system cannot counter the growing competitive imbalance between clubs at different levels of the league structure. It also suggests that teams with a track record of training new players are not sufficiently rewarded financially as the most lucrative transfers mainly occur between the wealthiest clubs.

Moreover, the growing number of transfers highlighted in the CIES Football Observatory Demographic Study 2014 shows that the current transfer system does not efficiently prevent contract instability. On the contrary, players’ mobility is on the increase. This goes hand in hand with the expansion of third-party ownership deals, through which private investors are entitled to shares on future transfer fees. This further undermines the redistribution role of the transfer system throughout the football pyramid.

Our analysis highlights that reform of the current transfer system is needed. An efficient solution would be to redistribute any transfer income to all clubs where a footballer has played during his career according to the total number of matches played for each club. This would ensure a fairer distribution of money to the whole chain of clubs that allowed the player to develop his career before attracting interest from the richest teams. The CIES Football Observatory academic team is at your disposal to undertake simulations for different redistribution scenarios. For more information, please contact us at football.observatory@cies.ch.

Exclusive data from the 2014 World Cup Preview

The employer club whose footballers played the most minutes of World Cup qualification matches is Real Madrid, followed by six other big-5 league clubs: Barcelona, Bayern Munich, Juventus, Manchester United, Chelsea and Manchester City. The highest ranked club not participating in the big-5 leagues is Zenit St-Petersburg (8th), while the first non-European team is Esteghlal FC of Iran, in 14th place.

Big-5 league teams represent two thirds of the 30 top ranked clubs. In total, 48.9% of minutes played were by footballers under contract with top division teams in England, Spain, Italy, Germany and France. This finding reflects the high concentration of national A-team players in the most competitive European championships.

All national teams qualified fielded players employed by big-5 league clubs. However, the percentage of minutes played by big-5 league footballers was below 25% for Australia, Costa Rica, Ecuador, Honduras, Iran, South Korea and Russia. These nations will most probably struggle to qualify for the knockout stages.

We hope that you will enjoy the new statistics presented in our 2014 World Cup Preview and we stay at your disposal for more analysis at football.observatory@cies.ch.

1 | ... | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17