Paul Murray of EMB reports on research he carried out together with Thomas Trompetter of Converium that offers the most detailed analysis ever of footballer injury patterns among top players in England and Germany. Here are some of the main findings.
When Jimmy Greaves – arguably the David Beckham of his age – played for Chelsea he earned just £20 ($38) per week. In those days, of course, footballers’ wages were capped and there were no billionaires, let alone Russian ones, to pump money into their bank accounts. Even allowing for inflation, the combined reported weekly salary of John Terry and Frank Lampard could have financed the entire Chelsea team for a whole season at the start of the 1960s.
Nowadays, a player who is injured for a long period represents a sizeable financial as well as sporting setback for his employers. In the past, to protect the money they had invested in transfer fees, clubs may have insured against the risk of injury prematurely ending a player’s career. As football’s Bosman ruling has reduced the level of transfer fees (with players like Sol Campbell able to move clubs on a free transfer), it has also led to an increase in player wages (with savings in transfer fees usually heading towards the players’ wage packets). The biggest risk to clubs in terms of player injuries is increasingly that they will have to pay high wages for an extended period to players who cannot contribute on the field of play. As with any risk, if it can be measured it can be insured, and a thriving market for sporting cover has emerged in recent years. Typically, policies that cover wages for a club’s injured players start to pay out after a player has been unable to compete for more than 30 consecutive days.
But how do you model and quantify such a risk, especially in such a newly emerging and rapidly changing field? That was the question posed by EMB and reinsurer Converium, one of the biggest writers of this type of business. Two separate exercises were carried out. One looked at the German Bundesliga while the other focussed on the English Premiership.
Finding the answer to such an apparently simple question was actually quite complex. The methodology may be similar to the processes involved in creating a conventional pricing model, but where do you go for the necessary statistics? No underwriter has been in this line of business long enough to have a meaningful database of claims and, as we soon discovered, the information was scattered and certainly not presented with the needs of underwriters or actuaries in mind.
Fortunately for the English analysis, though not the German one, there is a website (physioroom.com) that carries details of all injured Premiership players. They were able to supply dates, by player, of when injuries occurred and also when injured players were able to return to full training. But there were still cavernous gaps in the information if it was to be used to assess the risk of injury. To make sense of the injury data, the incidence of injuries had to be compared with when players were exposed to the risk of injury. For example, someone who joins a club halfway through the season and is sidelined through injury for two weeks has been as costly, in proportionate terms, as a team mate who is out for a month but has been there for the entire season. Investigation was needed into how long each premiership player had been at his club (even having to adjust for all the player loans that clubs were involved in).
A key question then was what factors might influence the risk of a certain player getting injured and EMB and Converium discussed this intensively before starting the German and English injury analyses. To address the question, data was gathered on player ages, their playing positions, their clubs, where their clubs finished in the league table, and how many minutes each player competed in each Premiership match (which includes being aware of not only every substitution made but also every red card shown).
As with any statistical analysis of this kind, we then had to isolate the pure effects of the different variables (for example, isolate what effect playing position might have on the risk of injury before assessing the extra effect that player age might have). It was then important to look for constancy in any results over time (for example, were the same injury patterns shown in each season?) and identify and discount statistical aberrations that might distort the overall picture.
So what did we discover? The most headline-grabbing conclusion was the fact that top footballers in England are more likely to be suffering from long-term injuries towards the end of the season than those in Germany. The clear evidence that we unearthed shows that a mid-season break appears to be the main factor in ensuring that when the destiny of major domestic, European and international trophies is being decided, German clubs are suffering from fewer long-term injuries than at any stage of the season. At the same time, English clubs (who have no midseason break) have more serious injuries than at any other period in the calendar.
In the graph Premiership injuries of more than two weeks in length are shown by month of occurrence. August is the worst month as players return to training, not necessarily fully match fit, and are confronted with hard, dry playing fields. There is then a steady improvement until November. In the second half of the season, the injury rate soars by nearly 60%, almost reaching August’s levels by the season end.
The second graph, rather than looking at when injuries occur, looks at how many players are missing through long-term injuries at any particular time. Due to the German injury data being harder to collect, it is shown here in quarterly periods rather than by month. For both leagues, injuries lasting approximately a month or more are included to tie in with the common 30 day waiting period for such insurance policies.
Both leagues start with few long-term injuries when players report back for training after the summer. But pre-season training is probably the most dangerous time of the year for injuries and the number of players suffering from serious injury rises quickly in both countries. In years following a major international championship, more top level players will have long-term injuries when they report back to their clubs but by the start of September, the number of long-term injuries is not noticeably different from any other season.
In the first half of the season, both leagues see the rate of occurrence of injuries slow down but as many players who were injured at the start of the season are still unable to play, the number of players out through long-term injuries increases. While English and German clubs follow very similar injury patterns for the first half of the season, it is the second half of the season where the injury experience in the two countries diverges dramatically.
In the Premiership, the rate of occurrence of serious injuries starts to increase and as a result, even though many players injured at the start of the season are now able to return to training, the total number out through serious injury continues to increase steadily. But in Germany, the six week winter break intervenes. This gives currently injured players time to recover and also lessens the wear and tear on those who are still fit. When the Bundesliga season recommences, the number of players out through serious injury starts to fall. Those injured early in the season have generally had time to recover, very few extra players have picked up significant injuries in the winter break and those who have stayed fit throughout, tend to keep in reasonable shape. This means that when the Bundesliga gets going again, there isn’t the sort of peak in new injuries seen at the very start of the season.
As the season reaches its climax, the English players are once again experiencing very high levels of injury occurrence once again. But, in the Bundesliga, a combination of more rested players and a period midseason where very few players pick up major injuries means that the number of players missing through long-term injuries dramatically dips just when all the major trophies are being handed out. England manager Sven-Göran Eriksson has complained many times about the number of injuries he has had to cope with immediately before major championships. The evidence suggests he is right to keep asking for a mid-season break, as is the norm in most European countries.
Creating such a break in a crowded calendar would almost certainly mean abandoning the Carling Cup and/or reducing the number of Premiership teams to 18, as in Germany. That would require clubs to agree to financial sacrifice, as well as accepting a voluntary increase in their chances of relegation. Cynics would consider this scenario about as likely as Canvey Island winning the FA Cup.
Although this aspect of our research is the most interesting as far as the wider football public is concerned, it represents just one of many findings relevant to underwriters (see box). Individual capital providers must decide how, if at all, they affect them. Some of the main points that I have taken away from the exercise are:
• It is better to insure entire teams rather than individual players or you may find that you are selected against;
• Do not assume that minor players are any less of a risk than their potentially more expensive and better established team mates. The reverse may well be true.
Finally, do not be too surprised if, in some future World Cup, England once again thrash Germany in the qualifying stages only to see them outperform us in the tournament itself.
FOOTBALL INJURIES IN ENGLAND AND GERMANY –
KEY FINDINGS OF EMB/CONVERIUM RESEARCH
• English Premiership players face an increased likelihood of injury in the second half of the season compared to their German counterparts, who benefit from a mid-season break;
• Fringe Premiership players take longer to recover from a significant injury (as key players are rushed back into the team and perhaps have first call on their club’s medical facilities);
• Clubs with smaller squads do not appear to suffer any more long-term injuries (although they may find it harder to replace their injured players within the team);
• Goalkeepers are a third less likely to suffer long-term injury than outfield players;
• There is no significant difference in injury rates among strikers, midfield players and defenders;
• Between the ages of 18 and 26 there is a steady increase in injury rates. However, contrary to the intuitive assumption that players become more injury-prone as they get older, it seems that a 35-year-old Premiership player is no more likely to be out through injury for a month or more than a 26-year-old. It may be that the more injury-prone footballers tend to stop playing at an earlier age, protecting the injury rates at higher ages from excessive deterioration;
• There is some evidence (though this is far from conclusive) that teams finishing in the top five of the Premiership will be more injury-prone during the following season (perhaps as players at these clubs tend to play many European and international matches);
• There is also some evidence (though again not conclusive) that players from clubs who finished higher in the Premiership during the previous season take less time to recover from long-term injuries. It may be that more successful clubs are able to provide better medical treatment for injured players.
The research was conducted in conjunction with Converium at the offices of EMB in Epsom (UK) and EMB Deutschland in Cologne (Germany). Thomas Trompetter is head of medical underwriting and claims at Converium Rückversicherung (Deutschland). The German reinsurer belongs to the Converium reinsurance group of Switzerland and is responsible for the group’s worldwide life and health business, including the special sports business.
This article appeared in Insurance Day in March 2005
