By David Butler

‘Style’ has cost Sam Allardyce his job at Everton apparently. Despite being quite forward thinking throughout his career and doing a sterling job (in terms of points accumulated) since taking over at Everton, Sam has found it hard to shake his managerial identity.

It seems fans and owners of traditionally elite clubs want their team to not only win, but to win well.  Such are the demands of modern management.

Measuring style can be tricky. Goals scored from set pieces is one way that comes to mind. Another is passes completed by a team. Typically, teams that are easier on the eye can effortlessly string together passes into neat moves. Arsenal’s invincibles come to mind and so to do the Barcelona and Spain teams that adopted ‘tiki-taka’ so successfully.

The graph below shows passes per match (PPM) plotted against points for premier league clubs this season. Maybe a more granular measure would be a count of strings of forward passes.

Maybe the style of teams can be considered this way.

Manchester City were outstanding to watch this season. Manchester United on the other hand, while second, were criticised for effective but boring performances – Liverpool, Tottenham and Chelsea all ‘outpassed’ Manchester United. Arsenal’s style also stands out.  Wenger’s team had the second highest PPM in the league at 619.05 but it was accompanied by a leaky defence.

What about Sam and Everton? Their PPM is wedged between Brighton and Huddersfield. While Sam delivered in terms of points accumulation, the ‘product’ he delivered, as measured by PPM, was no more attractive than that provided by the promoted teams.

Burnley are an interesting case. The fans seem happy with Sean Dyche despite their relatively unattractive football. They have the 3rd lowest PPM 357.16 in the league. Burnley are only ahead of two relegated teams on the PPM stat - Stoke and West Brom. Expectations are important – managers not afforded big budgets or those freed from the anticipations of free-flowing football are lauded once things go right with a relatively abrasive style.

Trying to play beautiful football when you’re stuck in the trenches may not be the best idea either. Southampton have a relatively high PPM of 451.95. The Southampton fans were arguably offered a more attractive style than plenty of other club but Pellegrino still got the sack. Would they trade off the style value for a season like Burnley’s?

By Robbie Butler

The 4th sportseconomics.org workshop will be held on Monday 23rd and Tuesday 24th of July 2018 at University College Cork. The event is being hosted by the Centre for Analysis and Risk.

The theme of these year's event is sports betting and a selection of technical academic papers in the area will be presented over the two days. 

​General Information

Betting on sports is an enormously popular activity, which earns significant revenues, enhances many fans’ sense of emotional involvement with sports, and, on the downside, causes substantial welfare loss to people who gamble recklessly or addictively. For the economist interested in people’s preferences over risk, and the interaction of these preferences with their beliefs about probabilities, sports betting provides a field setting almost ideally designed for supporting rich inferences. Policy makers and regulators who have responsibilities associated with sports, leisure, and gambling can gain crucial insights into, and predictive leverage over, responses to changes in odds and incentives through observations of naturally occurring sports betting, as well as through economists’ laboratory and field experiments with punters. The CEAR workshop on Sports Betting brings together both researchers who specialize in the economics of sport, and theoretical and experimental risk specialists who use sports betting contexts as environments for establishing more general insights into risky choice.

Conference Program

Monday July 23
​
9:15 – 9:45        Launch of the UCC Centre for Sports Economics and Law
9:45 – 10:00      Workshop welcome and Background on CEAR
10:00 – 11:00    Filippos Papakonstantinou: History-dependent risk preferences: Evidence from individual choices                               and implications for the disposition effect

11:15 – 12:15    Angie Andrikogiannopoulou: Heterogeneity in risk preferences: Evidence from a real-world betting                              market
12:15 – 1:15      Glenn Harrison & Eberhard Feess: Welfare Evaluation of Sports Betting Conditioned on Elicited                                  Risk Preferences and Subjective Beliefs


2:30 – 3:30        Sylvan Herskowitz: Gambling, Saving, and Lumpy Liquidity Needs
3:30 – 4:30        Amit Gandhi: TBA

Tuesday July 24

9:30 – 10:30      Raymond Sauer: TBA
10:30 – 11:30    Rob Simmons: Betting Markets, Outcome Uncertainty and Competitive Balance in Sports: What                                    Have We Learned?

11:45 – 12:45    Brad Humphreys: A Simple Test of Semi-Strong Form Market Efficiency Using US College Sports                                Betting Data

2:00 – 3:00        Alasdair Brown: The Use of Framing Manipulations By Victorian Bookmakers
3:00 – 4:00        David Butler, Robbie Butler and Justin Doran: Turf Accountancy - A Spatial Analysis of                                              Bookmakers
4:00 - 4:05         Close 

By Robbie Butler

With the dust settling on the 2017/18 Premier League season, preparations for the league's return in August are already gathering pace. With World Cup 2018 on the horizon, club activity usually occurs in two bursts. before and after the tournament, making World Cup years feel somewhat different.

This is no more evident that the recent sacking/parting company of Premier League managers in the past number of days. Arsenal have been joined by West Ham United, Everton, and today Swansea and Stoke, who are all seeking a new managerial appointment. 

In a previous post, I illustrated the downward trend in the number of days managers stay in their job, which the League Managers Association chief Richard Beven described as "embarrassing" for the sport. His words have had little effect on club owners.

During the 2017/18, excluding the recent exists of Arsene Wenger, David Moyes, Sam Allardyce, Carlos Carvalhal and Paul Lambert, ten managers lost their jobs. This includes two changes at West Bromwich Albion. That's a total of 16 managerial changes. 

The pressure that managers come under, particularly at the start of the season, seems to be growing year on year. Within this, the league itself appears to be breaking into two groups. Teams fighting for the top 4 and teams in a relegation battle.The mid-table appears to be disappearing. In fact, this season one could argue only Burnley and possibly Watford (although their season tapered off) were not in a relegation battle or in with a realistic chance of reaching the top 4.

The graph below would support this somewhat, and illustrates how the league is squeezing out the mid-table teams.

The data used in the graphic is drawn from Premier League tables from 1995-96 season to 2017-18. Data prior to this is not useful as the league contained more than 20 teams. 

The number of points required to win the league and finish 4th (last Champions League place) is upward trending. The points needed to survive is downward sloping.

As a general rule, 40 points is used as a benchmark for safety. One would have to go back to 2002/03 to find the last time 40 points was actually required to be safe. in the past five seasons, an average of just 34.4 were required to stay in the division. The average from 1995-96 to 1999-00 was 37.4

On the other hand, the number of points needed for the top 4 is climbing. Just 65.6 points were required in the first five seasons show above. Since 2013-14 this has risen to 73.2. In the first twelve seasons (1995 to 2007) only once did the 4th place team break 70 points. In the eleven since May 2007 the 4th placed team was broken 70 points on 8 occasions. Arsenal finished 4th in 2013-14 with a whopping 79 points. 

By David Butler

For the last four seasons I’ve kept a record of the predictions for the Sky Sports pundit Paul Merson and the BBC pundit Mark Lawrenson. The ‘Follow the Pundit!’ section of the website provides updates on the success of these tipsters who systematically predict score lines for Premier League matches. 

The stats for this season are based on 367 (Merson) and 359 (Lawrenson) predictions. A similar pattern to past season exists – in any given gameweek the pundits typically call half of the outcomes correctly.
 
While Merson does slightly outperform Lawrenson again this season, both performed worse when compared to last season.  

If you were to place €1 on each of their scoreline predictions, Lawrenson would have actually turned a profit of €58.81 through predicting 12% of the scorelines correctly. Following Merson's scoreline estimates with a €1 stake per match renders a loss of -€47.77.

By John Considine

Last week, the latest edition of the Irish Sports Monitor was presented with the usual positive spin.  The first bullet point in Section 3 says "43% of the population regularly participate in sport. This is at the same level as 2015 ...".  A nice colour picture is used to illustrated the point and is reproduced below.  The website of Sport Ireland reiterates this point as one of the key findings of the latest report.

Was it really 43% in 2015?  That was not my memory of it.  But my memory is not what it used to be.  Therefore I decided to check out the 2015 Irish Sports Monitor that is available on the website of Sport Ireland.  The corresponding Figure 3.1 from the 2015 report is produced below and seems to suggest that the 2015 participation rate is 45%.  The picture says 2016 but directly above it the text talks about 2015 ("The 2015 survey finds that 45% of ...").

Maybe the method of data collection has changed?  In Section 2 of the 2017 report there is a paragraph that suggests that 2015 and 2017 are directly comparable - whatever about previous years.  A screenshot of the paragraph is reproduced below.

So it seems the figures for 2015 and 2017 are comparable.

Let us suppose it was 43% in 2015, 45% in 2016, and 43% again in 2017.  This raises further questions.  This would mean it increased by 2% in one year and decreased by 2% in the following year.  A previous Irish Sports Monitor suggested participation rarely changes by more than 1% per year.  Strange.

I will have to return to the documents and try to uncover something that I must be missing.

By John Eakins

A recent article on the42.ie caught my eye (here). It outlined how World Rugby have approved of temporary rule changes to the scoring system for a new series of invitational matches. These matches, dubbed ‘World Series Rugby’, have been established to provide games for the Australian rugby franchise team, the Western Force. For those of you who don’t know, the Western Force were dropped from the Southern Hemisphere Super Rugby competition last year as the number of teams were condensed from 18 to 15. The Force are scheduled to play matches against Fiji, Tonga, Samoa and Hong Kong along with other Super Rugby teams, the Melbourne Rebels and Crusaders, over the coming months.

While a number of temporary rule changes have been approved for the series of matches, including placing a one minute limit on scrums and allowing lineouts to be taken as quickly as possible, what has generated most interest is the introduction of a seven point try (which turns to nine points if converted). A seven point try will be awarded, if a team initiates an attack from behind their own 22 metre line and there is no break in the continuity of their possession before they score, that is, when the ball is turned over to the opposition through a penalty, scrum or lineout.

According to the42.ie article, Rugby Australia is hoping the change “will reward positive play.” I wonder if this is the main rationale to be honest. The fact that the try is to be called a ‘Power Try’ and that the goalposts will light up to indicate to fans when a ‘Power Try’ is possible or has been scored suggests to me that this is more of a gimmick to attract people's attention to the new series of matches.

Another recent example of this, is a proposal by the English Cricket Board to introduce a new domestic cricket competition which will be based on each team bowling one hundred balls, 20 balls less than the current twenty20 format (see here). The working title of the new competition is 'The Hundred'. The shorter format, they say, will appeal to families and TV audiences. In addition to the shorter nature of the game, another element being proposed is to have a 10 ball final over (as opposed to the traditional 6 ball final over in 50 over and twenty20 games). Given that most twenty20 games are decided in the last few overs, I suspect most games in this new cricket competition will be decided in this last 10 ball over. Which is probably what the organisers want, that is, to ensure the outcome is undecided until the very end. The only issue I would have with this new format is audiences may focus too much on the final over and not bother with the rest of the game.

Are these changes genuine attempts to enhance the quality of the game or are they gimmicks? I’m not so sure. Many cricket players share my feelings too (see here, here and here). And this is where I think care needs to taken in treading that fine line between a genuine rule change and something which is intended more for marketing purposes. Of course gimmicks are not necessarily a bad thing and can add to the sense of atmosphere and excitement in a game. But the danger is that, with the increasing pressure on sports executives to maintain attendances and TV audiences, games could become more about the gimmick than the game itself.

By Sean O'Connor

With the Premier League winding down after what could be considered a fairly uncompetitive title race, it’s interesting to point to some facts and figures from the season thus far. As of the time of collecting this data (02/05/2018), a total of 927 goals have been scored. Of these 927 goals there have been 245 unique goal scorers. However, how is this broken down in terms of nationality? Table 1 charts this.

​The 927 goals scored have been scored by players from 53 different countries. Interpreting Table 1 columns titled 1 to 31 indicate the number of goals scored so far in a season. Each value in the accompanying table denotes the number of nationalities which have scored that number of goals. So, one Algerian (Islam Slimani) has scored a single goal this season, while another Algerian (Riyad Mahrez) has scored 10 goals. In terms of a share of the total, Algeria has contributed to 1.2% of the total Premier League goals.
​
As is to be expected, English players contribute the largest share of goals to the Premier League (28.8% or 267). Of these 267 goals, 23 have come from 23 players scoring a single goal. Although Egypt’s Mohamed Salah is the current Premier League to scorer, Egypt, have only contributed 3.8% of total Premier League goals.  From an Irish point of you, it has been a disappointing season so far. Ireland’s share of total goals stands at 1.2% with 5 players only scoring a single goal and 3 players scoring two. Although Table 1 provides an interesting examination of a country’s share of goals in the Premier League, it would be interesting to analyse these trends on a longitudinal basis. A future blog will look at this.

By Robbie Butler

There are presently 86 racing courses operating in the United Kingdom (UK) and Ireland, or British Isles to use the geographic terminology. The breakdown by country is England (52), Ireland (24), Scotland (5), Wales (3) and Northern Ireland (2). Racing at these locations is an all year round occurrence, with only Christmas Day and Christmas Eve in the UK certain to have no fixtures each year.

The regularity of the sport ensures a constant stream of live content for sports broadcasters. Traditionally available on free-to-air television and within bookmaker betting shops, the market for live content changed in the early 2000s with the arrival of subscription channels. Much like the market for Premier League football a number of broadcasters now exist and “compete” with one another.

As Hoehn and Lancefield (2003) identify, completion occurs in a two-stage game; the first stage involves competing for rights to the race courses, if rights are non-exclusive or face close substitutes, broadcasters compete in respect of the quality of their coverage and analysis. The second is competition in the downstream market for consumers.

In the UK, there are two main racecourse owners. Arc Racing Company and Jockey Club. Between the two they control almost half of all British race courses. The remainder are owned independently.

Ireland is somewhat different. Horse Race Ireland (HRI), a government financed government body, manages the sport on an all-island basis with the two tracks in Northern Ireland directed from Co. Kildare. While ownership of the courses is largely private, direction, fixtures and the vast majority of funding comes from HRI.

Racing is available on free-to-air channels in the UK and Ireland on a weekly basis. That said,  the number of races shown is limited and coverage is normally restricted to a Saturday. Remaining races can be viewed on a subscription basis by purchasing one of two channels; Attheraces (to be rebranded Sky Sports Racing from on or before January 1st, 2019) and Racing UK.
​
As on May 2018 Attheraces has the rights to all 15 Arena Racing Company, Hereford Racecourse (run by the local council), a number of independent tracks and all 26 race course on the island of Ireland. Racing UK holds the rights all 14 Jockey Club tracks, two run by Chester Race Company and numerous independent course.

Things are set to change however as it was recently announced the Racing UK had won the rights to show all racing from Ireland from 2019 onwards. While Sky Sports Racing recently announced an agreement to take over broadcasting of live races from Chester Race Company’s course in Bangor-on-Dee and Chester, the loss of 26 courses will be seen as a major blow. The graphic above demonstrates the change in market power from 2018 to 2019 for both channels.
​
Attheraces coverage of Irish racing has been excellent over the past decade or more. It will be interesting to see how the move to Racing UK (which may now require rebranding) works out.

By Robbie Butler

It is hard to escape the Brexit debate in Ireland at the moment. The shadow of the UK's pending departure from the European Union hangs over all aspects of life on this island. Sport is no exception.

We have previously agreed the issue on various occasions in the past year or so (here, here, here, here and here). With the talks between the EU and UK ongoing, a lack of clarity as to the final outcome remains. Generally speaking, there are two broad outcomes; the UK leaves without a deal in March 2019 or a deal is agreed resulting in a transition period to the end of 2020. Given the demands on both sides it is difficult to know which outcome is more likely.

Derry City football club have a greater stake than most in the outcome of the talks n this regard. Based in County Derry, one of the six counties of Northern Ireland, the club joined the League of Ireland (in the Republic of Ireland) in 1985. Derry have since played in the league ever since from their home ground the Brandywell. 

The EU status and free-travel agreement between Ireland and the UK historical meant the process of playing away games, and welcoming visiting team, was effectively seamless. This was made even easier following the Good Friday Agreement in 1998 and, in the years that followed, an absence of controls of any sort. 

Things are about to change however. The UK's current position is to leave the EU, Single Market and Customs Union. Should the country exist all three it is extremely difficult to envisage how a customs border could be avoided on the island of Ireland, at least one that is managed on the Irish side, to prevent certain goods entering the European Union.

This will no doubt create a difficulty for Derry City each time they play an away football match. The club will be travelling from a non-EU country to the EU. The opposite is true of all clubs visiting the Brandywell. The logsictal problems this will create are still unknown, but is it likely the players, management and supporters will face challenges in the seasons ahead should a deal remain elusive.

If a "hard" Brexit is the outcome, the club itself may have to consider whether the benefits of League of Ireland membership continue to outweigh the costs. 

By Robbie Butler

I sat watching the Champions League semi-final first leg between Liverpool and Roma last week and had mixed emotions at the end of the game. As a Liverpool supporter I have been trying to reason why I was a little uneasy despite the fact the Reds had secured (another) three goal advantage going into the second leg of a Champions League knock-out game.I consoled myself with the fact that others around me had similar feelings. 

Before the game I think a three goal lead would have been taken by every Liverpool player and supporter. In fact, all would probably have taken a win, of any description, and some may have been content with a 0-0. Even a score draw would not have been an awful result. 

The root of my uneasiness can be somewhat explained by anchoring - a cognitive bias that explain the tendency for people (in this case a football supporter) to rely too heavily on an initial piece of information. The initial information was a 5-0 lead that had me thinking about the final. 5-1 was less favourable to this. 5-2, as the result turned out be, was less favourable again. 

The sequence of the score was very important. Had Roma led 2-0, a 5-2 win for Liverpool would have been heralded as one of the greatest comebacks of all time. Had the game been 2-2, and Liverpool gone onto win 5-2, the win would have been seen as a great victory. When 5-0 becomes 5-1, and then 5-2, this is not the case despite the fact the outcome is identical. 

This moves me on to prospect theory - the idea that  people make decisions based on the potential value of losses and gains rather than the final outcome. There are three fundamental pillars at the heart of the theory, created by Daniel Kahneman and Amos Tversky. One: Evaluation is relative to a neutral reference point. Two: Diminishing sensitivity to gains and losses exist. Three: Loss aversion is paramount, with losses looming larger than gains.

​The theory is best explained by the graphic to the right, which I have manipulated to show the relationship between the joy and pain from scoring and conceding a goal which I experienced last week.