By Robbie Butler

The Wealth and Poverty of Nations is a masterful exploration of political economy by the late Harvard professor David Landes. Very early into this 650 page text the author raises the issue of Eurocentrism - a worldview that is inherently biased towards Western civilization, or western Europe (depending on ones' definition). 

At a recent event at the University of Oxford, UEFA President Aleksander Ceferin appeared to suggest that such Eurocentrism is alive and well in European football. The former Slovenian lawyer said the following about the recent Europa League Final stage in Baku, Azerbaijan:

"We decided a year and a half ago that we play in Baku, which has a modern stadium of 70,000. I think there is only one stadium in England that is bigger....If somebody asks me why we played in Baku, I would say: 'People live there. Homo sapiens live there...They had to watch the game at 11pm because of the time difference but nobody complained...If we have two Azerbaijani teams playing in London nobody would complain. They would come and play without any problems...We have to develop football everywhere not England, Germany only...[the] Europa League and everything else should be shared with the others who love football."

Ceferin's comment was largely motivated by the negativity surrounding the Final last month between Chelsea and Arsenal. Much of the commentary of the Final focused on the empty seats at the stadium and lack of atmosphere. Many questioned why fans had to travel so far to see their teams complete. A vivid example of Eurocentrism?

Although the 'big 5' have some of the largest populations in Europe, UEFA represents 55 full member associations . The association’s strategic plan from 2019-2024 has four pillars one of which aims to “Promote and develop football infrastructure across Europe”. Spreading the game is a mission that has been at the core of UEFA, and FIFA for that matter, for many years. This is why cities such as Baku get the chance to host major finals. 

It was impossible to predict who would play in Baku, and the distance fans had to travel was probably accentuated by the fact that Chelsea and Arsenal are under 10 miles apart. Despite what people may now believe, this was not a poorly attended final. In fact, it is the 3rd best attended Europa League Final to date.

Countries on the periphery of Europe are just as entitled to hosts these games as the England, Germany or Spain. In fact, I recall not so long ago a country on the edge of Europe hosting the Europa League final. I was there when two Portugesse teams just 55 km apart, were forced to travel a round trip in excess of 2,700 km. The location? Dublin. I don't remember the Portuguese having any problems. 

45,391 people showed up in the Aviva that night. There were 5,979 more in the Baku Olympic Stadium. 

By John Considine

At the recent Sport Economics Workshop held in University College Cork, David Forrest presented a paper titled “Spectator Demand for the Sport of Kings”.  Afterwards, in the company of David and Rob Simmons, I asked about the seemly small figures that David had presented for the average number of race meetings per track and the average number of times horses raced.  A comprehensive explanation followed.  It is always a privilege to hear David Forrest and Rob Simmons explain some of the economics surrounding horse racing.
 
Prompted by the discussion I decided to examine the Irish horse racing schedule for 2019.  What follows are some of the numbers on this schedule.
 
There are 80 days on which racing is not scheduled to take place.  On the remaining 285 days, there are 211 days on which only one race meeting will take place, 73 days on which two race meetings take place, and one day on which three race meetings are scheduled to take place (December 26th).
 
In terms of consecutive days racing at a particular venue then Galway and Listowel are unusual in that each venue has a seven days of consecutive racing.  Killarney and Punchestown have 5-day events.  However, the majority are stand alone single days racing.
 
Racing is scheduled to take place at 25 different venues.  The distribution of the schedule across these venues is listed in below.  The list is presented in descending order by number of race days.  The all-weather track in Dundalk tops the list.

When one hears the term “all-weather” then one thinks about the underfoot conditions.  However, it is the presence of artificial light that is as important in explaining Dundalk’s number of events.  There are 147 “evening” meetings.  One hundred and fourteen (114) of these are held in tracks other than Dundalk during the second and third quarters of the year.  However, for the first three months of the year, and the last three months of the year, the only evening fixtures are held in Dundalk.

by Declan Jordan

The FIFA Women's World Cup in France continues to throw up great games and surprises. The women's game is in the spotlight and, while the football has been great, there are frustrations at FIFA's organisation and commitment to the game, pay inequality, and sexism.

It is perhaps unavoidable that the competition and women's football generally is compared to the men's equivalent. It is not difficult to find patronising commentary on television and in print media whenever women's football is covered. There is almost surprise, particularly in the Irish and UK coverage with which I am most familiar, that professional athletes are technically skillful, competitive, and committed. (There are similarities in the treatment of female football analysts, referred to by Alex Scott here. That it is surprising that someone who has played 140 times for her country should be insightful about the game says a lot about our attitudes to women in football).

Perhaps these attitudes are based on an outdated notion of the women's game, though there are certainly socialised stereotypes at play as well. In a recent article in the Irish Examiner, Larry Ryan referred to a comment by England striker Toni Duggan that she was somewhat heartened to see the anger from opposition fans when she celebrated a goal she scored for Barcelona. For her it showed that the game mattered and the players were being taken seriously. The article though warns that the women's game should try to avoid going down the road of the men's game and should instead try to hold on to the values of inclusion, loyalty, and community.

This doesn't mean of course that the women's game will not get more competitive, skillful, and demanding.

The negative stereotypes will take time to break down of course, but it is interesting to review the data on previous World Cup Finals, both women's and men's, to shed some light on the game's development.The USA's 13-0 demolition of Thailand has led to some suggestions that the World Cup is losing credibility and the increase in teams from 16 to 24 will damage the tournament with more one-sided games.

The data doesn't back this up however.

The graph below shows the goals per game at each of the last seven World Cup Finals tournaments (the last men's tournament was 2018 in Russia and the last Women's tournament was in 2015 in Canada). For information, in the first round of group matches in the current tournament (12 in all) to Tuesday June 11, there have been 3.17 goals per game. Excluding the USA-Thailand game the ratio is 2.27 goals per game).

It is striking how the ratio of goals per game has declined over the seven tournaments. From having just over a goal more per game, the women's tournament has now reached a comparable level with the men's. Whether this is a positive or negative development is unclear, though even as the number of countries in the finals of the Women's World Cup has increased goals have become more difficult to come by.

Similarly with the number of low-scoring games, there has been a narrowing in the differences between the tournaments over time. The Women's World Cup Finals did not see a scoreless match until the final day of the third World Cup (in 1999) when both the third place play-off and the final ended nil all. Up to the last World Cup finals there have been nine scoreless matches. There has been one more in current tournament when Argentina parked the bus against the favoured Japanese. In comparison over the last seven men's tournaments there has been an increase in scoreless matches from close to 5% of matches to 15% in 2014. In 2018 there was only one scoreless match.

It is interesting though to include the number of games that have ended 1-0. The green bar in the graph below shows the proportion of matches that have had one goal or less (within 90 minutes). The women's World Cup has gone from about 12% of matches in the first tournament in 1991 (and only 6% in 1999) to a third of matches in 2015. (A third of the 12 games in the current tournament have had one goal or less). 

​The grey bar shows the corresponding data for the men's tournaments. Rising from 1994 to 2010 before falling in 2014 and 2018, the level has been between about a quarter and a third. The women's rate has exceeded the men's in the most recent tournament. 

It may be that the fewer number of participants in the women's finals means the best teams are playing and the matches are closer. The number of teams taking part increased however in 2015 from 16 to 24. Adding the extra teams has not diminished the closeness of the games (measured by the proportion of low scoring games). It may instead point to an increase in standards - or at least a decrease in the variability of standards - at the elite level and the women's game converging with the men's game, whether for good or ill.

The data may also point to greater levels of organisation in coaching and associated resources from national bodies for the women's national teams, and further posts may explore that idea.

By Robbie Butler

The FIFA Women's World Cup kicked off in France last week and for the first time ever all 52 games from the tournament will be shown live on free-to-air television in Ireland. This is an excellent opportunity to showcase women's football to a mass audience in Ireland. The importance of this cannot be underestimated and should increase the level of interest in Ireland for the women's game. 

There have been goals aplenty so far and not a single draw to date in the group stages. Tournament favourites and hosts France got off to an impressive start winning 4-0, while defending champions the United States will begin their quest for a record 4th World Cup crown tomorrow (Tuesday).

Speaking to friends over the weekend that had been watching games from both this tournament, and the UEFA Nations League and European Championship qualifiers, a number of people observed that technical proficiency (close control, passing accuracy, dribbling, etc), rather than physical dominance, appear to be more obvious within these games. Given the need to be technically proficient, does that then mean the relative ages effects (RAE) are less to be present at this competition?

The answer is no. ​RAE is almost always as visible in underage football for girls as it is for boys. When one considers adult teams across both gender RAE is almost never a factor. The data below presents the distribution of dates of birth for players at the World Cup in France. 

Broken down by month of birth (left-hand side) or quarter of birth (right-hand side) the distribution of birth dates is largely random.

A closer look at the Nations League semi-finals and final would probably produce similar findings. While being born earlier in the year (regardless of gender) is advantageous for selection at underage teams, deselection for elite national teams is more likely for those earlier in the calendar year.

By David Butler

A lot has been said and written of late about the best way forward for Irish football (soccer).  

It is important to make a point in relation to elite development and senior international performance in the men’s game. This concerns the role of the senior domestic league. While the League of Ireland has a role to play in Irish football, the evidence would suggest it is remiss to consider it a place to systemically develop or draw talent from for our senior international team.
  
This point is premised on the models of competing European nations. Looking at our competitors, it is clear that very few current senior internationals stay-on to play in their domestic league. Our opponents Friday night (Denmark) only have 4 players contracted to Danish clubs, 3 of whom are uncapped. There are many more examples from other nations similar to Ireland that take the same develop-and-export approach. Iceland (2), Bosnia and Herzegovina (1), Sweden (3), Austria (5), Norway (3), Slovakia (4), Switzerland (5), Portugal (7) are just some other cases where a minority of the senior squad is drawn from the domestic ranks.

Ireland, like many of smaller nations, do not have the population or the national league strength to support a develop-and-retain model. Only the Big-5 nations and a handful of others with significantly larger populations such as Russia, Ukraine and Turkey adopt this type of model.

The well-known Irish ‘problem’ is that players are drawn almost solely to British leagues. This is owing to a range of cultural and economic factors, and is partly due to the Republic sourcing English born talent. The solution to this problem is not to look inward but rather to break the tradition of exporting on mass to English clubs. My hunch is that player’s will receive greater developmental opportunities outside of British leagues. We can remedy the  current mind-set of focusing on the UK. In turn, this will aid the development of future Irish talent.

Historically, we have failed to export to other European countries. Developing policies to achieve this diversification of talent is needed. Our international competitors mentioned above manage to do this. They export talent to a diverse range of leagues. For example, the Iceland squad is represented by 14 non-domestic leagues, Bosnia and Herzegovina by 13, Denmark by 6, Sweden by 11, Slovakia by 14. This list goes on. With the exception of Sean McDermott, who is contracted to Kristiansund BK, and James Talbot (Bohemians), all other members of the current squad play in the English leagues.  

Of course, this is not a call to abandon the League of Ireland. We just need to taper our ambition for the developmental capacity of the senior game. Developing two or three clubs to the standard seen in other small European leagues would be a major achievement and should be a key goal. If developmental opportunities are on a par in two-three clubs to those in other small European leagues, the League of Ireland could be in a place to retain a minor amount of senior internationals. This is the structure for several European competitors and should be the aim for domestic football here.  

Dominance in the LOI would likely bring about balance problems, but it may be a necessary trade-off to get some of the trappings of dominance (e.g. improved player development process and entry to elite European competitions). This is seen elsewhere.  For example, 2 of the 4 Danish based players in the squad for Friday night play with F.C. Copenhagen – they have won 6 of last 10 Danish Superliga and have finished 2nd in 3 of the last 10. Similar outcomes exist in Austria. FC Red Bull Salzburg have 8 won out of last 10 Austrian Bundesliga, and were 2nd twice. 4 out of the five current Austrian national players are contracted to the club. These examples go on - Switzerland for instance is dominated by FC Basel and BSC Young Boys. The few Swiss based senior internationals are often contracted to these clubs.   

Perhaps the Croatian approach is the best example to think about. They have a similar population to Ireland. At times they suffer from a ‘tramway’ league - like Ireland they have too many clubs based in the capital. Like us, they also suffer from low domestic attendances. While we have several similarities, our international fortunes have been very different since Croatia emerged as a major player at the World Cup in 1998. This is simplistic but the differences are interesting. In the tradition of other smaller European countries their current senior squad only sources 5 players from the domestic league but is represented by players playing in 11 other countries. Like the Austria, Denmark and the Swiss examples, they have historically dominant clubs in Dinamo Zagreb and Hajduk Split that develop and export talent but are strong enough to retain a minority.

The take away message is that to improve senior international football, attention needs to be placed on developing our U13 to U19 divisions with the aim of continually exporting our best. The change however has to come in the form of players not solely being transferred to UK leagues. Developing the men’s senior league should be a secondary, long term objective to supporting a predominantly export based model. By assisting the emergence of dominants club(s) - which of course is far easier said than done – the Irish setup would develop a similar structure to other small European leagues. 

By John Considine

Prior to last week’s Europa League final between Arsenal and Chelsea, there was plenty of debate about whether or not Petr Cech should be the Arsenal goalkeeper.  It was to be Cech’s last game and there were rumours that he was to take up a non-playing position with Chelsea next season.  It begged a few questions about Cech’s commitment to the Arsenal cause.
 
Contract situations, somewhat like the one Cech found himself in, are the subject of a recent paper by Daniel Weimar and Katrin Scharfenkamp.  Published in Managerial and Decision Economics, the paper examines the effort that football players spend on behalf of their current team when they have already signed a contract to play for another team.  The authors found a statistically significant 1.23% reduction in effort by such individuals.  Crucially, this did not translate into a statistically significant reduction in team performance.
 
Weimar and Scharfenkamp measure effort using the distance the player runs during a game (hardly a measure useful for goalkeepers like Cech).  One of the more interesting elements of the paper is the discussion of why this measure was selected.  For example, the authors note that the correlation between the distance covered by a player and the market value is low (r = -0.03).  In addition, the authors go to great lengths to control for all sorts of potential other influences.
 
Establishing a measure of individual contribution to team performance is not easy.  And it can be more difficult in some sports than others.  On the surface, one would imagine that the individual contribution is easier to identify in a sport like baseball than football.  Yet, it is not straight forward in baseball as another recent paper by Heather O’Neill and Scott Deacle explains.  Published in Applied Economics, the O’Neill and Deacle paper spends a fair deal of time explaining why the authors chose on-base-plus-slugging percentage as their measure of individual effort.  This paper also presents evidence to show that there is a relationship between a player’s contract position and their effort.  Players show improved performances in the final year of their contract.

Both papers present impressive amount of data.  One of the features of sport is that sports performances tend to be public and measurable.  The accumulation of evidence makes it is hard to deny that there is some systematic variation of a player’s performance and that this is related to their contract.  It is less clear whether this variation is priced into the contract.  Or, as Joel Maxcy and others have noted in the literature, maybe the contracts are structured in such a way to deal with such situations.

By Ed Valentine

Back in December Liverpool led the Premier League, just as they did for 11 minutes on the final day of the season in May. Despite losing just one game, to eventual Champions Man City, the reds finished a point behind in 2nd place. It was an incredible title race as no team has ever hit the 95-point mark and not lifted the trophy (incl. non-Premier League era).

The table below was posted on this site in January and demonstrates the forecast I produced as part of a simulation on the outcome of this year’s title race.

The prediction was wide of the mark on a number of counts – Spurs didn’t finish 2nd and Burnley survived the drop. It did get Huddersfield’s relegation correct but a dart throwing monkey could have told you that.

It did however predict that Liverpool would finish on 99 points – 2 more than their actual final total. Had they achieved this Jordan Henderson would be waking up in his Madrid hotel suite with the Premier League Trophy. I wonder if Man City had required 100 points to win the title, would they have converted a draw to a win somewhere along the line?

Pep has boasted about winning 4 titles this season, I wonder how many he’d swap for the chance to win the Champions League in Madrid on Saturday?

A reminder of the forecasting methodology – The simulation was run using data from the 2018/19 season so far (19 games). The simulation model ran 200,000 times with the post sim averages for each attribute being captured and presented in the table.

By David Butler

The media have run with various stories concerning Arsenal and Chelsea returning unsold tickets to the Europa League final this evening in Baku. Estimates suggest it could be up to 6,000 tickets. The final has been plagued by logistical issues relating to flights and visas, in addition to safety concerns.

Since 2000-01, the average distance travelled by both finalists to the host destination is circa 4000km (one way). Arsenal and Chelsea will travel over 9000km to reach Baku. Keep in mind these finalists are 14km apart.

While this is the tenth UEFA Cup/Europa League final to feature two teams from the same country, the distance the London teams and fans are travelling is by far the greatest in recent times. Similar but less intensive travelling happened in the recent past. In 2006-2007 Sevilla and Espanyol met in Glasgow, with fans travelling nearly 5000km. Porto met Braga in Dublin for 2010-11 final, travelling over 5000km. A year later Atlético Madrid met Athletic Bilbao in Bucharest – this involved travel of over 6000km. These three finals attracted close to a full house. 

Applying a 10% capacity constraint on the venues since 2001 (as UEFA often require lower capacities) shows that only 4 out of 18 finals had an attendance of less than 95% of capacity. These slightly lower attendances may even be explained by bigger stadium capacity rather than fan preferences (i.e.  Westfalenstadion, Dortmund).
​
The Baku Olympic Stadium in Azerbaijan holds 69,870 and a UEFA limit will set tonight’s capacity at 64,000. I’ll be keeping an eye out for the attendance figure. Even if it is high, tonight's match may be more for the locals than the core supporters of the two finalists. 

By Robbie Butler

Our annual sports workshop was held last Friday. Among the very interesting papers was a presentation by Babatunde Buraimo about Premier League viewership figures over the course of the past five seasons. The data suggested that the number of those watching live games on subscription television is in decline. This piece here supports this view and claims the fall-off in demand is worrying.

A number of very interesting issues were addressed during the course of the presentation. Firstly, football is one of the few things on television that is not really conducive to "on demand" viewing. The kick-off time is the kick-off time, and that is it. This leads to a second issue regarding the residual value of sport. This is close to zero once a game ends. By this I mean, who wants to watch the full match once the game has ended? 

Of course, residual values do lie elsewhere. This is where the highlights reel comes in. Fans will gladly watch selected moments from a game after the final whistle. This is where the idea of Match of the Day must have its origins. The evolution of the show, first screened in 1964 is terribly interesting, even more so given its popularity today.

With the rise of online, mobile and tablet services, as well as streaming games, one has to wonder how a programme that goes out at around 10.30pm on a Saturday night, long after the outcome of all games is known, and highlights readily available seems to be as popular as ever. 

This certainly deserves more investigation. 

By Robbie Butler

​The 1st Reading Football Economics Workshop will take place from the 19th - 21th of September 2019 in Reading, England.

This event is hosted by the Centre for Institutions and Economic History at the University of Reading, and will include two keynote talks, a number of contributed sessions and informal social events.

Submissions are invited from scholars working in the areas of football economics or football management. This includes the applications to football, or the use of football to inform economics and other topics.

Full paper submissions are preferred, but extended abstracts will also be considered. Submissions should be send to j.j.reade@reading.ac.uk before the 21st of June 2019.

More information on this event can be found here.