By Robbie Butler

The Premier League season normally kicks off in sunshine and blue skies and marks the end of the summer to many. As the weeks go by, the weather turns gradually darker, wetter and colder. It is hard to sometimes pin-point when the season turns from late summer to the darker days of the pending winter. One important milestone of the new season is the first international break. It is often a time to take stock and assess early form. It might even spell the end for a manager that got off to a slow start.

As we approach October, and the next round of games falls in that month, all 20 EPL managers remain in their post. Over 8 of the past 9 seasons (I had to exclude 2020/21 due to the delayed restart to football), only twice have all 20 managers gone beyond the 4th of October in their posts.

The table below lists the first sacking of each season and the date of departure.

Over the course of the past 9 seasons, more than 150 managers have left their posts. Last season, the 20 managers actually stayed in situ until almost Christmas. A pretty remarkable feat in one of the most hostile labour markets on the planet. 

As of today, a few managers might be growing concerned that October may spell the end of their reigns. Erik Ten Hag is now odds-on favourite to see the door first this season. Managers at Wolves, Southampton, West Ham and Everton are amongst the next most likely to leave, according to the bookies.

Once the first manager has been sacked - and it is enevitable - the early season feeling of summer and blue skies will be well and truly behind EPL fans. The long winter lies ahead. I wonder how many managers will survive until the end? Last season only 4 saw the door - including Julen Lopetegui - who stepped down just before the season started. 

​By David Butler

Broadcasting live sport is interesting as the seller shares in the uncertainty regarding the quality of the product with the buyer. For example, just like football fans, Sky Sports or TNT sports, do not know the outcome of a football match either. Sky did not know that Crystal Palace vs Manchester United would end 0-0 last Saturday evening – if they did, would they have chosen to televise it? Maybe. Given the international following of the Big-6 clubs, perhaps the broadcasters will still draw larger numbers independent of the result for these clubs.  That said, I wonder what damage (if any) the dreaded 0-0 does to broadcasters figures - particularly in less internationalised leagues. 

For live sports, we usually think of uncertainty from the buyer’s side/demand side. But the seller shares in this too. They often select matches to show from a menu of potential fixtures. The broadcaster (I’m assuming) doesn’t wish to pick the old maid, they would likely prefer to choose the higher scoring outcomes and more entertaining matches to broadcast from the bundle. But in the context of football, statistically, broadcasters will unintentionally sell  a 'lemon'.

There are not too many markets where two-sided asymmetric information exists - where either side does not have full knowledge on the product quality. Examples often arise when mother nature is crucial to supply.  Travel agents can provide the plane and hotel but can’t guarantee the sunny holiday conditions. Farmers may sell crops before the harvest through futures contracts but don’t know the exact quality or yield of their crops until after a growing season. Maybe other areas of the cultural industries could be included too – Art or Music for example – sellers will only learn afterwards the quality of the product they are selling. 

Back to football - it would be interesting to see the data on the effect of these scoreless draws on later broadcasts. It would also be interesting to see whether these stalemates early in a round of fixtures (or at any early match slot time) had an impact on demand for later matches.

Below is a list of results from the Premier League over the last few seasons that ended in a 0-0,  ranked by match (total) expected goals. The higher on this list, arguably, the less entertaining the match. After-all, a 0-0 may still have had lots of attacking threat, missed chances etc.  
​
Eyeballing the list, perhaps broadcasters should avoid Crystal Palace as a rule of thumb!

By Robbie Butler

Alison Becker and Rodri are the latest elite footballers to go public this week and discuss the ever-increasing demands placed on professional footballers. The new-look Champions League hasn't helped, with the inclusion of two extra group games. However, more games means more revenue - broadcasting, ticket sales, matchday operations, etc.

Before their game with Inter Milan tonight Man City's Rodri went on record and said "I think we are close to that (striking)...I think it's the general opinion of the players. And if it keeps this way, there will be a moment where we have no other option...It is not the same for all the players, not all play 60 games, but all of us think the same. It's about the time we have for a break, many things." 

If strike action were to be taken, it would not be the first time. Long before the world had heard of Jean Marc Bosman, English international George Eastam was contracted to Newcastle United. Unhappy with his conditions, Eastam went on strike.

The source of this dispute was the player's arguning "whether the house the club had supplied him was habitable, the unsatisfactory secondary job that the club had arranged and their attempts to stop him playing for the England U23 team". A secondary job! His footballer salary was not enough to make ends meet one can assume. 

It wasn't until recent years that elite footballers have earned enough money during their short careers to comfortably retire in their mid-30s. There are very few professions that can say this. Footballers that played from Victorian Britain until the early 2000s could only dream of the weekly wages on offer in places such as the English Premier League or Saudi Pro League today.

While successful players - thoses that play 50+ games per season and international games - do play regularly, I am sure footballers of the past would gladly swap their salaries in return for having to play more (and ditch their secondary job!).

My guess is most fans would do the same. I recall a banner at a football stadium, with a message to the players, once reading along the lines: "We all dreamed of playing. God chose you. Now play like we dreamed".

By Robbie Butler

I watched the Republic of Ireland on Saturday and Tuesday. The squad is not in a great place right now and it is difficult to be optimistic about the short-term future. England are an excellent team and a 2-0 loss is not unexpected. The Greece result is far more worrying. In the not to distance past, the home side would handle such opposition and a draw would be deemed a 'poor' result. How times have changed. 

The data below shows the national team's performance in qualifying (both Euro and World Cup) since 2000. The successes in 2002, 2012 and 2016 are matched by higher points total percentages. 2018 appears to be a near miss - a playoff defeat to Denmark end the Martin O'Neill era. 

The 2020 campaign under Mick McCarthy saw a falloff in performance and this has been exacerbated by the disastrous campaigns during 2022 and 2024. 

The data suggests that a win rate of 50%+ is necessary to qualify - all home games for example. This will become harder as Ireland fall down the ranking. Previously, we had been a 2nd second seed so only had to play one higher ranked team at home. In the near future we may have to play 3 higher ranked teams at home. 

Right now, Ireland are winning about 20% of qualifying games. That's about 2 games per campaign. Considering each group has the likes of a San Marino, Gibraltar or Liechtenstein, that's a pretty poor return. 

By John Considine

While sport provides experimental settings for economists to test their theories, it can be difficult to separate the physical and the mental elements.  As a result, some sport settings are better than others.  Chess is better than most.  It has a long (and checkered!) history of use by economists.  Here are some examples.
 
A September 2023 paper in Management Science examines the relationship between the quality of chess moves and the quality of the air that the participants were inhaling.  I found the use of chess and the methods more interesting that the results.  The paper uses over 30,000 moves, made by 121 players, in 609 games in official tournaments held in Germany between 2017 and 2019.  Twenty-first century journal editors and academic referees like large datasets.  The quality of these 30,000 moves were evaluated using a chess algorithm.  Each move could be measured by the change in probability of winning.  (Not surprisingly, the authors found that the quality of moves deteriorated as the quality of the air deteriorated.)
 
The Management Science paper uses the actions of chess players as data from a "field" experiment.  It is interesting to see human action being evaluated by an algorithm (that is based on previous moves by chess players).  Previously, economists like Steven Levitt and Ignacio Palacios-Huerta have also used chess as a source of data.  In their cases, chess players behaviour is used to examine the capacity of humans to undertake backward induction using other games, e.g centipede.  Research has used the ranking of chess players to test for differences in the capacity to use backward induction.
 
Rather than collecting data from chess games, Thomas Schelling famously experimented using a 16x16 chess/checkers board to test his dynamic models of segregation.  Schelling’s results were sensitive to the assigned motivation of the zinc and copper coins used to populate the board.  Schelling’s experiments were more of the trial & error type.  Schelling effectively used the board to experiment with his ideas.

Adam Smith used chess to illustrate, rather than test, his ideas.  In The Theory of Moral Sentiments (1759), Smith criticized those who believe that they “can arrange the different members of a great society with as much ease as the hand arranges the different pieces upon a chess-board” because “each single piece has a principle of motion of its own”.

Chess is a useful sport (game) for economists.

By Daragh O'Leary and Cillian Mac Sweeney

The only permanent thing in life is change and this is certainly true of sports. Today I will be taking a look at changes in the scoring patterns of All-Ireland hurling finals. For Cork the most recent All-Ireland final was again a painful one, but it does provide observations for my latest analysis – take the rough with the smooth.

I’ve looked at the score lines from the All-Ireland hurling finals from 2013-2024 and 1913-1924. Immediately one can see that there are now far more points scored in these matches than there was 100 years ago.

​On average, there is roughly 42 more points being scored in hurling All-Ireland finals now than there was a century ago. Additionally, there are also fewer goals being scored now. The change in the game from goal scoring to point scoring in the modern era is quite stark when graphed – see below.

From 2013-2024 there has been a clear trend in the scoring patterns of All-Ireland hurling finals whereby the number of points being scored is increasing, but the number of goals being scored is (slightly) decreasing. The 1913-1924 period tells a very different story.

For the majority of this period, goals are the more common type of score in finals and it is only in the last few years (early 1920s) that the number of points scored catches up to and surpasses the number of goals being scored. To put this into context with the modern game, in the last decade of All-Ireland hurling finals there are 3 instances where the winning county scored no goals at all (Limerick twice & Galway once).Overall this means that today we are getting FAR more scores, but fewer goals.
​
There may be a few reasons for this. The most obvious one would be the evolution of the game away from ground-hurling. This shift is captured well by photos from back in the earlier half of the 20th century where the difference in shape of the bosses on older hurleys is far closer to that of a hockey stick and more conducive to hitting the ball on the ground – see below. 

Another factor which may have also influenced this trend was a cultural legacy of goals being far more valuable than points. Since they were introduced to the sport in 1886, points have always had a value of 1 on the scoreboard. Whereas goals were worth 5 up until 1895 and only reduced to 3 in 1896. It is possible that in the 1910s a lot of the coaching and tactical thinking surrounding hurling was still aligned with the ideology that goals were five times more valuable than points. Subsequently, it may have been that it took time for the greater relative value of a point to be properly realised and then reflected in scoring patterns.
​
Finally, it is worth noting that modern day sliotars are far lighter than the ones which were used in the 20th century. Subsequently, it’s simply easier to hit the ball higher and further now which is conducive towards scoring points.
It’ll be interesting to see what the future holds for the scoring patterns of hurling. Will we ever see a goal again?

By Robbie Butler
]
Viewers of yesterday’s Manchester United Liverpool game may have noticed a strange addition to the new Liverpool 3rd kit, worn for the first time this season. The Nike logo – traditionally a horizontal tick – has been rotated and is now vertical double ticks. The Liverpool Echo provides an explanation for the change here.

The home (red) retains the traditional Nike logo, with the change applied to the white, third kit for 2024/25.

​Our 2021 paper “The evolution of the football jersey – an institutional perspective” in the Journal of Institutional Economics charts the proliferation of the football jersey through time. In the modern game, new jerseys are almost inevitable each season. Remarkably, more recently clubs have opted not to change their jersey season-to-season for sustainability reasons,

Options with the home kit are tight, as the colour almost never changes. Incremental changes are therefore made to aspects of the jersey such as the collar, sleaves, colour tone and fit. On some occasions, modifications happen as the sponsor manufacturer changes. Less frequently, the club may opt to redesign the crest.
​
The move by Nike is a step into a new direction (for me at least). While the jersey supplier can change, I had yet to witness a modification to the logo of the manufacturer like this. Is this a new reality football fans may have to deal with? It will be yet another way to introduce an incremental change to something that is essentially unchanged for most clubs for a century or more.

By David Butler

In a recent paper published in Sports Economics Review, colleagues and I investigate the presence of nationality salary premia in two top European football leagues (the Premier League and Serie A). 

The main finding is a substantial pay premium for South American players of between 11% and 15%. We find no evidence of a salary penalty for domestic players. This salary premium is driven mostly by Brazilian, Uruguayan and Argentine players. We explore a range of issues around this and ask if the salary effects are driven by new entrants to the league, and whether they are reflected in team attendances and team performance. Fans appear to respond to higher proportions of South American players in England, but not in Italy. One other notable findings is that African players appear to be underpaid in England given their contributions to team performances.

We offer conjecture on why certain South Americans are paid more. Perhaps their nationality is a credible (quality) signal, this may relate to recruitment and scouting networks within the countries or it could be explained by  infrastructure in these countries. Political factors may also play a role and there could be interactive mechanism here. 

The full paper can be accessed here