SUMMARY: Economists have extensively analyzed the causes, dynamics and consequences of ‘competitive balance’ in sport competition. This concept refers to the degree of equality in sport performance among all teams taking part in a competition. Competitive balance is also linked to the attractiveness for fans of a sport competition. We propose some metrics concerning the perceived competivive balance of national leagues in countries in Europe, based in betting odds about title winning probabilities. We argue that competitive balance is not the best way to approach the degree of attractiveness of a sport competition. We consider and try to justify that attractiveness is rather linked to the degree uncertainty concerning the winner of the championship, and this measure is affected by the closeness of just top 2-3 contenders.
Regular football season starts by the end of August in almost all countries in Europe. Football is the most popular sport in many countries in Europe in terms of followers. Football become again one of the hot topics in talks and chats. Football season is also a relevant event considering its commercial power. Just top 40 clubs in Europe create a combined revenue of more than 4.4 billion euro (Deloitte 2012).
One of the topics discussed at the beginning of every new football season is which are the most attractive national football league. England, Spain, Italy and Germany are considered the ‘Big Four’ as they concentrate the vast majority of football players stars, international sport success at UEFA Champions League and are the most powerful championships in economic terms. These last ten years there is a dispute among fans and experts about where global leadership lies: the ‘League of the Stars’ in English Premiership or ‘La Liga de las estrellas’ in the Spanish Liga.
We can show a first answer just by showing the economic value of the team roster of clubs inside each national league. Economic valuation of player is a very nice proxy for sportive skills and talent. The sum of all economic valuations is somehow telling us which is the league that counts with more aggregate sport talent. This should provide us the insight about the relative quality and attractiveness of the different national leagues in Europe.
We use the economic valuation estimates provided by Transfermarkt, a web site that is becoming a reference in this field. Next figure shows the results using these estimates.
The English Premiership is currently commanding the ranking, with an estimate combined value of almost 3.4 billion euro. Second most powerful league is Spanish Liga, with 2.5 billion euro. Third comes the Italian calcio (2.1 billion). Number four is German Bundesliga (1.8 billion euro). France Ligue 1 takes fifth position (1.45 billion euro).
Four other leagues count with a value bigger than 500 million euro (Russia, Portugal, Turkey and Netherlands).
Arrived to this point, many argue that economic power of a league is not a sound measure for gauging the quality and the attractiveness of a national championship. It is argued that even if total sport quality is relevant, it may even be more important to check how this total amount of quality is distributed among the different teams participating in the competition. If quality is distributed very unequally and concentrated is very few teams, then the attractiveness of the competition may fall dramatically. By contrast, it may be considered that a more modest national league in terms of football stars may be in the end more attractive to followers (and as business model) if quality is more equally distributed among teams.
All this debate is labeled in the sport economics area as ‘competitive balance’ analysis. Economists have identified different metrics for measuring the actual extent of balance of sport power among teams pertaining to a given league. A lot of scientific literature has emerged dealing with causes influencing the state and dynamics of competitive balance in sports. It has also been studied the impact of competitive balance in demand of sport, and it has derived into recommendations concerning the design of competitions for improving competitive balance.
The underlying hypothesis is that competitive balance as defined by the metrics increase the attractiveness of the competition.
As announced in the title, the main goal of this post is to gauge which are currently the most attractive national football leagues in Europe. If we follow mainstream economic analysis, we are asked to address the issue by measuring the degree of competitive balance of each championship.
Choosing a metric for pre-season competitive balance
First task is to choose a metric for giving values to the degree of competitive balance in each country. A myriad of measures of competitive balance have been developed and proposed insofar. As competitive balance refers to the degree of equality of all teams taking part in a championship, many measures are built around the difference between the status of perfect sport performance equality of all teams and current situation. Barely all measurements apply ex-post analysis based in actual results. The typical instrument used is number of points obtained by each team at the end of the season. The index of competitive balance is obtained by a composite measure of the difference between number of points if all teams finished with the same number of winning games and actual number of points (above or below).
As our intention is to provide information about the attractiveness of leagues in Europe before the season starts, we cannot adopt the traditional approach, as by definition we do not have information about current season sport performance. It would be inappropriate to use information about classification reached in the previous season, as each season supposes a rupture: new hirings and transfers, change of coaches.
We could use the information already shown concerning the economic value of the team roster in the current season. As explained, this is probably a good proxy for sportive talent of each team. Differences in economic value of teams inside a national championship could tell us a lot about expected competitive balance for the current season 2012/13.
Of course, there is a strong relationship between economic investment in a team and sport success. This is evident and it has been confirmed by all empirical studies. Money brings trophies. See for instance the close relationship between classification by points in the English Premiership and the economic value of the team roster.
We like the current economic value of the team roster as proxy for the expected competitive balance of each national league for season 2012/13. But we will not choose it because this measure is not able to capture other non quantitative aspects that do affect sport performance of a team. Consider for instance the role of the coach. He (or she) may affect dramatically the sport performance of a team, for any given value of the team roster. Also, the degree of experience and link-up of the team players may vary also strongly.
We prefer to select a rather holistic measure of expected team sport performance: title winning chances as estimated by professional betting firms. Online betting provides us a valuable and transparent information about experts’ estimations about the sport quality and skills of any sport player or team. Estimations may be wrongly determined, but we do not expect them to be biased, as betting firms are risking their money by the rewards given by the betting odds offered for any single sport event offered to bettors.
The variable we choose is the assigned probabilities for each team member of a league to win the championship by the betting firm. We use the measures provided by the leading betting firm Bwin.com, for estimated before the commencement of the season in each country. This is a measure that stresses the differences between top and bottom teams in terms of sport quality: many teams in a league have almost no chance to win the championship under reasonable assumptions, and their assigned probability to win approaches zero. This is not a methodological problem as far as there is a direct and positive relationship between ex ante winning probabilities and the final position taken in the table at the end of the season. If this is so, we can use the betting odds as an indirect measure of pre-season competitive balance.
See as example the case of the English football premiership. It shows the relation between the current value of the team roster of the clubs taking part in the English football premiership and their probabilities of becoming the new title champion, as estimated by the online betting firm Bwin. We find a clear positive relation between economic power (and team roster sport talent) and expected sport success. We find also that there is an economic threshold establishing the barrier for becoming a championship challenger: it is impossible to become a true contender for teams with a team roster value lower than 200 million euros. This is the case for 14 out of 20 participating teams.
Still taking the English case as an illustration, we have that there are three teams with substantial assigned probabilities to win the championship, while there are other three teams that count with thin but existing probabilities to win the title. Betting right now for any other of the remaining 14 teams as 2012/13 title champions is wasting your money as it is currently considered as an almost impossible to happen event.
Betting chances is the variable that we have chosen to measure the attractiveness of a competition. Does this picture correspond to an exciting or a boring season for English championship followers?
As almost always in our analysis, absolute numbers don’t carry a lot of information. Figures become informative when the are put into perspective. One option is to use dynamic analysis, comparing current situation with values in previous season. But as our goal was to talk about competitions in Europe, the other approach is to compare values in different countries for the same season. This is what we do now.
Football competition attractiveness analysis using a competitive balance approach
Remember that we are pursuing our analysis about the attractiveness of football leagues by using competitive balance methods, as in mainsteam sports economic literature it is argues that competitive balance is a necessary condition for ensuring the fans’ demand for sports. We will escape later from this approach, but right now we continue our analysis using pre-season betting odds into a competitive balance framework.
If we adapted our measure from traditional competitive balance measures, we should estimate the aggregate deviation between actual betting chances to win the championship against the one corresponding to a perfect competitive balance scenario. We will not use this measure, as we prefer to apply another one that allows a much intuitive and easier interpretation.
We calculate first which is the betting odds corresponding to a perfect competition balance status. It corresponds to the value where all teams count with the same ex ante probability to win the championship (Pcb). This is as easy as Pcb=1/Tn, being Tn the number of teams taking part in the championship. This means that this value will vary depending on the total number of teams. In Europe it ranges between 1o teams (Austria and Switzerland) and 20 (England, Spain, Italy and France).
The measure of pre-season competitive balance that we propose is to calculate the percentage of teams that are assigned by the betting firms with a probability odd to win the championship equal or bigger than Pcb, the probability under perfect competitive balance conditions.
We present the competitive balance results concerning top 23 leagues in Europe by economic power (only Ukraine is missing), for the season 2012/13. The results refer to a context of perfect competitive balance as defined above.
The minimum score (8.3%) is reached in Scotland. There is only one team, Glasgow Rangers above the Pcb value. The other most unbalanced championships are in the Spanish Liga (10%), Germany (11.1%) and Serbia (12.5%).
In the opposite, the leagues showing the highest competitive balance ratios are Russia, Poland and Sweden, with valued bigger than 30%: almost one third of participating teams can claim counting a true chance to become the champion of the tournament.
Among top 10 leagues, we find that the more balanced are Russia, Netherlands and Italy, and the ones presenting the lowest levels of competitive balance are Spain and Germany. England is in between.
We have imposed ha very stringent threshold with the reference Pcb. It would be interesting to see how the picture is modified if the competitive balance conditions are somehow relaxed. We could apply the same analysis for still relevant winning odds, but smaller than Pcb. The idea is to calculate the number of teams that reach at least the status of a fraction of the Pcb value.
Soft competitive balance values still make sense in our quest for identifying the most attractive football national leagues. We can take also into consideration for instance teams with at least 50% of the perfect competitive balance status (SPcb=Pcb/2). This new sample of ‘competitive’ teams include new teams that even if they are not really considered as realistic contenders for winning the championship (except under extraordinary conditions) they present a sufficient sport talent and quality in the squad. In normal times this will ne be enough to allow these teams to fight for the championship title till the end of the season. But this also means that this kind of teams are strong enough to punctually win any type of game, even against top ranked teams (at least when they play at home). If we include these second line teams in our competitive balance analysis, we can gather a more nuanced picture of the state of competitive balance among leagues in Europe.
We present in the following figure the results of competitive balance using a soft measure. We have chosen a quite soft ratio, SPcb=Pcb/3, which means that we include in the new set almost all teams presenting an actual competitive power to fight against almost all teams in the league. We have included as term of coparison the values reached using strict competitive balance measures (Pcb).
According to our results, Austria present by far the highest competitive balance standards: 70% of all teams are above the threshold. We find then as much as 8 different leagues with soft competitive rations ranging between 40% and 50%: Germany, Russia, Turkey, Romania, Switzerland, Poland, Croatia and Denmark.
As for the leagues presenting the lowest rates of competitive balance, we find Spain and Scotland. In both cases, only two teams count with relevant winning probabilities. There are other three championship showing Soft competitive balance rates below 30%: Portugal, Norway and Serbia.
If we restrict the analysis to top ten championships, we observe that Germany, Russia and Turkey count with a really nice number of relevant contenders in the league. They are then really open competitions. Comparing soft to strict competitive balance results we observe that Germany and Turkey change radically their status, as they presented low values in terms of Pcb. This is telling us that in bot Germany and Turkey there are few aspirants to win the title, but many other strong mid class teams. England, Italy, France and Portugal present a quite similar competitive balance structure: the competition is controlled by just 3-4 teams, but almost one third of the participating teams count with the quality to dispute and win a match against any opponent.
The Spanish league is one in a kind among top national leagues: This is a pure story of two, the archrivals FC Barcelona and Real Madrid, and there isn’t anyone else expected in this battle. This is competition completely dominated by the two teams accumulating top football stars and titles locally and internationally.
Under all competitive balance standards we should conclude that the Spanish liga is an almost completely unbalanced structure, that can be used as study case.
Then, applying competitive balance lessons and recommendations, we could conclude that the Spanish Liga is by far the most boring championship in Europe. From the very beginning we know that there are 18 teams out of 20 that have no chances at all to contest Real Madrid-Barcelona domination, and even that they have almost no chance to resist a single match against them.
This conclusion is in fact shared by many football followers in Spain, especially for those supporting a team other than Real Madrid or Barça. This perception is also shared by other international football fans.
As I don’t share this view, I will try to explain in the next section why I am convinced that the Spanish league is not the most boring league in Europe. I rather consider that it is still one of the most attractive ones, even if nobody can question the fact that it is completely unbalanced.
Searching for a more genuine measure of the attractiveness of football leagues
Increasing competitive balance means diminishing the sportive performance gap between top and bottom teams taking part in a competition. This has as a consequence an increase of the incertitude about the outcome of games between top and bottom teams. Thus, promoting more competitive balance has as expected result to increase the attractiveness of each match and of the whole competition. More competitive balance drives more spectacle. Competitive imbalance tends to create boring games and boring competitions.
Of course, I don’t pretend to question this basic result. But my point is that competitive balance, understood as it generally as increased equality among all teams in a league is not for me the main driver of attractiveness of the competition.
More important than the uncertainty of a match is the magnitude of the reward at stake in the match. Of course, the best of the worlds in terms of match and sport competition attractiveness is that you count both with a lot to win/lose in a match, and high degree of incertitude about the result of the match. But the extent of the reward is in itself the main source of spectacle and passion linked to sports.
The most precious rewards in sports competitions is becoming the champion. When analysing the attractiveness of football leagues in Europe, the right question is to estimate how hard is to win the championship in this league. It would seem that these considerations drive us back again towards traditional measures of competitive balance, as when overall competition increases in a league, it becomes more difficult to be the champion. While this is not untrue, this relationship is not the most important one.
Even if this seems paradoxical and provocative, the main factor behind a ‘hard to win championship’ is not overall competitive balance, and even not the number of top teams counting with relevant probabilities to fight for the championship. The main variable making more costly to win the championship is the difference in probabilities to win the championship between number one favourite and number two.
If top two teams in a league are tied in championship favouritism, the attractiveness of the whole competition is maintained at its hight, as the main reward of the competition, the champion, is completely undefined. Of course, if instead of having two teams tied as champion favourites a league has the chance to have three or four teams, the spectacle is even higher.
But having just two tied teams is enough to count with an exciting competition as far as the uncertainty about the future champion is maintained.
We arrive to this point to the critical problems of traditional competitive balance measures. Competitive balance would take high values in leagues having one clear favorite (having 20-30 percent points of advantage against the followers in terms of probabilities of winning the competition) plus three-four other teams that are quite strong profile but lying well behind top one favorite. Competitive balance will provide low marks for a league with just two top teams tied in winning probabilities with the other teams of the league with no real chances to get the title. The problem is that in the first case, even if you count with some four-five good teams, you count from the very beginning with a clear championship favorite: the main reward of the competition is almost assigned before the competition stars. The contenders will play nice games and eventually beat the main favorite, but they won’t be able to become true rivals in the championship fight. The league will no reach high levels of attractiveness, as there is one clear favorite to win the championship.
Consider the story in the following way: we have a national league with two-three top teams that are fighting for the championship title. As long as the distance in points between them is small, it is unpredictable to determine who will be the champion and the competition if attractive for followers: the owner of the main reward of the whole competition is undecided. The high degree of uncertainty makes that each game played by these teams are crucial: the victory is needed to maintain the race to the title championship. If a team succeeds in gaining a substantial advantage against the other one-two competitors as the season advance, then the degree of tension associated to the competition drops dramatically. Once a clear favorite emerges, there remain only second level rewards to be assigned: teams fighting for a place in next season international competitions, and teams trying to avoid relegation positions.
This drop of interest of the championship when there is a big gap between the first team and the followers may appear at the end of the season, at mid-season… or even before the competition starts.
This is our main point. We can measure at the beginning of the season how attractive a national football championship is expected to be by monitoring the assigned probabilities to win the championship to top favorite team against the chances assigned to the contenders teams. Arrived to this point, we can confirm that this measure of the expected attractiveness of a football league is not directly linked to competitive balance measures.
Our measure of expected attractiveness if national football leagues: the Boring Championship Index
The measure that we propose to check the expected degree of attractiveness of football leagues in European countries is easy, simple and direct.
The basic measure is the Boring Competition Index is simply the difference between top 1 team and top 2 team in terms of the assigned probabilities wto win the championship, as estimated by the betting firms. As explained before, we use Bwin odds as reference for our analysis.
The figure below shows the results concerning season 2012/13 for top 24 leagues.
According to our data, the leagues that are expected to be more uninteresting are Scotland, Croatia, France and Denmark. In the first three cases, the favorite team counts with more than 50 percent points of advantage over the second most powerful team of the league concerning the chances to win the title. The advantage is so massive, that there is no uncertainty at all in these leagues. Except the occurence of a major sportive disaster (that would make a lot of money for contrarian bettors), we know before the first day of the competition the name of the winner of the championship. We know that PSG will be the champion in France, Glasgow Rangers in Scotland, Dinamo Zagreb in Croatia and FC Copenhagen in Denmark.
According to us, this situation kills the main source of attractiveness of a competition, and makes it a rather boring spectacle for followers of football. The very important point is that this clear source of lack of attractiveness of a sport competition (affecting demand and revenues) is not well captured in traditional competitive balance measures, as we will proof in the next section.
Moving now to the opposite situation, we find that the most attractive lagues in terms of winner uncertainty are Romania, Czech Republic, Norway and Poland. In these leagues there are at least two teams so tied in terms of expected sport performance, that it is impossible to predict right now which one will win the championship. As long as the suspense is maintained during the season, these leagues will capture the attention of local football followers.
If we restrict the analysis to top ten leagues in Europe, we find that the less attractive leagues are France by far, followed by Russia and Italy. The leagues that count with most unpredictable champion are Spain, Turkey, Netherlands and England. The spectacle will be sustained while the gap in points is thin among top teams. As some of these big leagues also attract massive international media coverage and fans supporting, they ensure the offering of an excellent and very valuable ‘entertainment product’.
(further analysis coming)