By John Considine

I often fantasise about conducting some unethical experiments in a sporting context.  My favourite involves surreptitiously swapping children in maternity hospitals.  This unnatural experiment would allow me to test if the ability to play particular sports is down to nature or nurture.  Another experiment involves surreptitiously installing recording technology in the dressing (locker) rooms of both teams in a game to establish if “what was said in the room” really makes a difference.
 
A third experiment came to mind this morning after reading an article written by Murat Mungan.  The article is in the latest edition of the Review of Law & Economics.  For this experiment I would need to measure the reaction of “sports economists” to the paper, specifically whether the person would read the article after an initial scan of its content.  My guess is that nurture would determine the reaction.  Those nurtured, and working in, the statistical approach to (sport) economics might decide against it.  That would be the vast majority of economists because the subject has become increasingly empirical since the 1970s (as explained in research by Beatrice Cherrier and Roger Backhouse).
 
Half a century ago, as the tide was turning, Alex Leijonhufvud wrote a humorous piece called “Life Among the Econ”.  He joked about the ability to econs to build “modls”.  Those who employed statistical method were called the O’Metrs (surely an Irish connection!) and were involved in strip-mining.  Murat Mungan builds a “modl” to examine real-time review standards and he uses Video Assistant Referee (VAR) to illustrate.  (Another experiment might be to ask economists “what is VAR?”.  My guess is that most would say a Vector Auto Regressive model).
 
For my part, I enjoyed Mungan’s paper.  Maybe those with an emotive reaction to VAR will read the paper.  Mungan suggests the analysis can be extended to body cameras in law enforcement and the oversight of military operatives such as snipers.  Other emotive subjects.

By David Butler

In the last few weeks Aston Villa have been working hard to retain their human capital and protect their value - John McGinn has signed a new contract to 2028, Matty Cash signed a new contract until 2029 and Morgan Rodgers new contract will keep in at the club until 2031.

No doubt these were all lucrative deals for the players and Capology estimates that McGinn, Cash and Rodgers will earn gross salaries of £6,760,000, £5,200,000 and 7,800,000 in 2025-2026 seasons respectively.

What aspects of performance command new contract pay awards? In a recent open access paper - Do sports analytics affect footballer pay? – Alex Farnell, Rob Simmons and I investigate the determinants of pay for players offred a new contract either through a renewal or a transfer.

Surprisingly, we found that few of our advanced performance metrics help to explain player salary, such as expected goals or expected assists. The evidence points toward more simple metrics such as goals and pass completion rates still being favoured by clubs. Unsurprsingly, accruing senior international status positvely affects pay, so McGinn, Cash and Rodgers all likely commanded a premium for donning Scottish, Polish and English shirts. 

​Watching the replay, I thought the handball decision — and awarding a free kick — was harsh. Anyone that understands the game (not just the rules) would probably agree. But when the referee gave this decision as a free — that is important — VAR does not have the power to overrule a freekick.

VAR can only intervene in four types of “clear and obvious errors” or “serious missed incidents”: goals, penalty decisions, direct red card incidents, and cases of mistaken identity. This was something different.

Had the ball struck the player outside the box, a free kick would have to be awarded. Liverpool were penalised because the player was inside the box and VAR thought it was not a penalty.

And to compound matters — what most people missed — is that if the referee had waved play on, Liverpool would have been awarded a corner! Because the referee awarded a free kick and because VAR was incorrectly applied — it was not a clear and obvious error, as the ball did strike the hand — Liverpool suffered three times. No free. No penalty. No corner.

Instead, the ball dropped at the feet of goalkeeper Thibaut Courtois — an uncontested hop ball.

By John Considine

​Lara Gillespie is Ireland’s latest world champion.  Her gold came in the World Track Championship in cycling.  The specific event being the elimination race.  The race is also labelled “Devil Take the Hindmost”.  Economists might think of it as a game of musical chairs because it can be linked to an analogy from The General Theory of Employment, Interest, and Money.
 
The most famous analogy in that book is between the stock market and newspaper beauty contests.  Nobel laureate, Richard Thaler, empirically examined a variation of the beauty contest with a "guess the number" competition run in the Financial Times.  Contestants were asked to guess a number between 0 and 100.  The winner would be the person who was closest to 2/3 of the average of all the guesses.  Backward induction might suggest the best guess would be zero.  That was not the winning guess.  Thaler used the finding to suggest that humans don't think like Homo Economicus.
 
Cycling’s elimination event provides data on how people play a version of musical chairs.  It can be used to examine the “greater fool” theory of investment.  Here are Keynes’s words on the issue taken in from the paragraph that immediately precedes his beauty contest analogy.

 “This battle of wits to anticipate the basis of conventional valuation a few month hence, rather than the prospective yield of an investment over a long term of years, does not even require gulls amongst the public to feed the maws of the professional; - it can be played by the professionals themselves.  For it is, so to speak, a game of Snap, of Old Maid, of Musical Chairs … when the music stops some of the players will find themselves unseated.”  Lara Gillespie successfully played a version of musical chairs at high speed.
 
One of my favourite books written in the immediate aftermath of the Great Recession is John Cassidy’s How Markets Fail: The Logic of Economic Calamities.  Cassidy says “Like John von Neumann, Keynes believed that simple parlor games have much to teach economists: they feature the sort of strategic interactions that are largely absent from orthodox economics”.  The greater use of data from sporting contests might be addressing that issue.

By Robbie Butler

Following some recent publication successes in Review of Behavioral Economics, the Journal of Sports Economics and European Sport Management Quarterly, we have now broken the 40-paper mark on the website.

All our papers can be found in the tab "Selected Publications". The cover an array of sports including football (soccer), rugby, boxing, horse racing, Gaelic games, greyhound racing, Formula One and more.

By Robbie Butler

They say good things come in threes, and perhaps there’s something to that. Following recent successes in the Review of Behavioural Economics and the Journal of Sports Economics, I’m delighted to share that a paper co-authored with regular blog contributor Daragh O’Leary has just been accepted for publication in European Sport Management Quarterly.
​
Our new paper, An Exploration of English Premier League YouTube Highlights Consumption: Unexpected Outcomes & Schadenfreude, develops a conceptual model to understand how fans engage with post-match content in the English Premier League. We explore how consumption patterns differ between live matches and highlights, and how supporters’ preferences for match outcomes, particularly when their rivals lose, shape viewing behaviour.

The central focus of the paper is Schadenfreude, that uniquely German word for taking pleasure in someone else’s misfortune, or, in modern terms, ‘hate-watching’.

Our findings reveal a significant and positive relationship between highlight views and instances of big teams losing. In short, people really do enjoy watching the mighty fall. It seems that for many viewers, football entertainment extends beyond their own team’s success; they also derive satisfaction from the setbacks of others.

This might help to explain the growing popularity of certain online influencers and fan channels. Their appeal does not necessarily come from celebrating constant victories, but from the drama, frustration and emotional rollercoaster that come with supporting a team, especially when things go wrong.
​
The full paper should be available online soon, and we look forward to sharing more about our findings once it is published. 

By John Considine

I got a pleasant surprise yesterday when someone sent me the screenshot of Figure 8 below.  The figure comes from a recent publication by the Irish Parliamentary Budget Office (available here).  The "Considine Rule" was first suggested in a paper in Economic Affairs.  The paper was the output of a group of economists from University College Cork (here).  Figure 8 might lend support to the questionable success of the "Considine Rule" as found in our paper in Public Choice (here).  Maybe the impact is more subtle than we found.

By Robbie Butler

Additional time and refereeing decisions are now a regular topic of debate in football. Recently, Liverpool played just three minutes of added time at the end of a Premier League game against Everton. The Reds were leading 2–1, and this relatively short addition was immediately questioned. One influencer even joked that the match must have been played in the 1990s!

But why do we react so strongly when this happens at the end of the game — and not at half-time? Few would question three minutes added at the end of the first half in the Merseyside Derby. But at the end of the second half, it’s suddenly a major story.

The reason is simple. The first half doesn’t decide the match. The second half does.

Our new paper, “Objective Calls under the Spotlight: Referee Consistency and Behaviour on Football’s Biggest Stage,” has just been accepted by the Journal of Sports Economics. The work is a collaboration between myself, David Butler, and Carl Singleton (University of Stirling).

We explore how referees behave differently under social and psychological pressure — especially when the game’s outcome is on the line - and hypothesise that referees face heightened social pressure near the end of matches, as outcomes become imminent. This may include subconsciously enjoy extending their time in the spotlight, prolonging those final moments.

In the closing minutes, every decision is magnified. Referees are under the greatest scrutiny, with millions watching their choices in real time. This phase offers one last opportunity to project fairness and competence — and to “manage” the optics of the match. Meanwhile, emotions among players, managers, and supporters peak. That cocktail of pressure may unconsciously shape refereeing behaviour — even if the rules are supposed to stay the same.

Some might argue that it’s natural for referees to treat each half differently. But according to the Laws of the Game, the same standards apply. Any systematic difference is a violation of consistency — and evidence of bias, however unintentional.

We examined additional time at two major tournaments - 2022 FIFA World Cup in Qatar and 2024 UEFA European Championship in Germany. Our analysis compared how much time was added in the first and second halves of matches, controlling for stoppages, goals, substitutions, and other within-match events. In theory, both halves should be treated identically. In practice, they’re not.

Even after controlling for stoppages and match events, referees consistently added more time in the second half than in the first — despite the rules being the same. We also report that when matches were tight, referees tended to add even more time in the second half — particularly at the World Cup. Interestingly, close contests there were actually shortened in the first halves.

These patterns suggest that social pressure and the stakes of the moment can influence even the most professional referees. This isn’t just about football. Our results highlight broader questions about decision-making under pressure and implicit bias in high-stakes environments.

So yes, referees do behave differently. Even in the same game.

The full paper will be available in the Journal of Sports Economics before the end of 2025.

By John Considine

Below is a screenshot of the ball position in the seconds before Paul Pogba scored to bring Manchester United level with West Ham at the London Stadium.  The ball seemed to curve out of play before coming back into play - even the words of an ex-Manchester United player in the captions suggest as much.  VAR examined the footage and did not overrule the decision of the line official.  The goal stood.  Later that evening, BBC's Match of the Day presented an alternative view were the ball was shown to be out of play.  The "error" went in favour of the team behind on the scoreboard.  Is it possible to examine if there are such compensating tendencies in the decisions of line officials? 

In a paper published in the last few weeks, Eakins et al (2025) examine the decisions of both the referees and the sideline officials and their relationship to the scoreboard (available here).  The line official has to decide if (i) the ball left the playing area and (ii) the team to last touch the ball before it left the playing area.  The task is fairly straight forward with little room for discretion where biases can intrude on the decision.  Contrast this situation with the variety of decisions that the referee has to make, e.g. many involve the evaluation of physical contact.  Partly because of this, the referee has greater discretion.  Therefore, if compensating tendencies do exist, we might expect to see a greater level of awards to the teams behind in refereeing decisions than in the decision of line officials.  Using data from an Irish stick & ball field game, Eakins et al find a significant relationship between referee awards and the state of the scoreboard.  They find no statistical significance in the relationship for the decisions of line officials.

By Daragh O'Leary

Former Republic of Ireland manager, Martin O’Neill recently garnered attention for his critique of the expected goal (xG) statistic on TalkSPORT. O’Neill called the xG statistic useless and claimed that it didn’t mean anything. He further validated his claim by saying that Brian Clough, who once famously claimed David Seaman wasn’t a good keeper because he had a ponytail, wouldn’t start players simply because they accumulated xG. 

In previous posts I have discussed my somewhat agnostic view of xG. Essentially, I understand the statistic’s usefulness in certain contexts, but I think most of the commentary surrounding xG appears to either misuse or misinterpret what the statistic actually measures.

Empirical economists have a term which they use to describe how well a variable measures a theoretical concept or phenomenon - construct validity. I think the xG measure has a high level of construct validity when it is used to measure individual chances - it’s intended purpose. This can be seen below with the images of two Bruno Fernandes’ goals. 

The image on the left where Bruno faces an open goal from a yard out with the keeper unable to stop the shot measures as an xG of 0.98 i.e. there is a 98% chance a shot from that position should result in a goal. The image on the right, which also ends in a goal, has an xG of 0.02 i.e. 2% chance a shot form that position should result in a goal.

The purpose of the xG measure is to indicate quantitatively that there is a far lower likelihood of a goal coming from the situation on the right than the situation on the left. When used to assess individual chances, I think it’s hard to argue that the xG statistic doesn’t have a decent level of construct validity. xG in that instance informs you broadly as to the goal threat associated with one chance relative to another.

The issue is that most people don’t use or interpret xG in this fashion. xG is commonly reported as an aggregate measure which adds up the total xG associated with each team’s chances throughout an entire game i.e. Team A xG of 1.45 vs Team B xG of 0.75. This doesn’t really make sense.

A team can accumulate an aggregate xG of around 0.75 by being awarded a single penalty (xG = 0.75) or by taking 10 long range shots which each have a very low likelihood of being converted into a goal – i.e. xG = 0.1 + 0.7 + 0.12 + 0.06… Meaning you can accumulate a high aggregate xG without creating a single ‘good’ chance. This means that, when aggregated, xG loses its construct validity because it may be high due to one or two great chances or 10 terrible “chances”.

Common criticisms of the xG statistic, like Martin O’Neill’s, occur because people point out that the aggregate xG stat often doesn’t predict the scoreline. While this is an understandable criticism, it is simply a case of statistics being used for a purpose they are not intended for. xG quantifies the probability that an individual chance will result in a goal. Using the statistic for any other purpose than this results in inappropriate conclusions being drawn from measures that aren’t fit-for-purpose.