Medical Moneyball - The Rise Of Injury Analytics

What if?  It’s a question that many of the world’s top teams asked in the last year when faced with ill-timed injuries to key players.  What if Derrick Rose of the Chicago Bulls, Robert Griffin III of the Redskins, Derek Jeter of the Yankees or Lionel Messi of Barcelona could have avoided their season ending injuries?  

Some are just the result of unlucky, violent contact but others have their origin from a combination of fatigue and overuse.  What if athletic trainers and team physicians could find early clues and signals that an athlete was at risk of breaking down?  Now, with the use of data analytics, that crystal ball may have finally arrived.

Stan Conte, VP of medical services for the Los Angeles Dodgers, declared last year, "in a post-Moneyball world, injury risk assessment is the final frontier."  At this year’s Sloan Sports Analytics Conference, he presented some surprising data to reinforce the rising toll of injuries;  just over 50% of all starting pitchers in the MLB had some type of injury during last season, lasting an average of 65 days on the disabled list.  Across all MLB players in 2012, the salaries of injured players plus the players that replaced them cost their teams almost $600 million.

Even at the Olympics, the world’s premier athletic showcase, the impact of injuries is significant.  Big names like Paula Radcliffe, Asafa Powell, and Rafael Nadal could not complete their gold medal quest.  Lars Engebretsen, a physician and professor at the University of Oslo and chief physician of the Norwegian Olympic team, has been tracking injuries and illness at the Games for over a decade.  His latest report, released this month on the 2012 London Olympics, recorded 1,361 injuries and 758 illnesses among the 10,568 athletes, which equates to injury and illness rates of 11% and 7%, respectively.  Unfortunately, these percentages are similar with the last two Summer Olympics in Beijing and Athens, highlighting the lack of progress in reducing lost time in competition.

In this Scientific American graphic, Engebretsen’s data from the 2008 Summer Olympics and the 2010 Winter Olympics shows that overuse caused 22% of summer athletes' injuries while 54% of winter athletes were injured in training.

Like the Dodgers, teams across the globe are beginning to search for answers.  As Big Data creeps into all aspects of athlete development, injury analytics is the new secret weapon.  That is what pushed the Leicester Tigers rugby union club to dig into the details.  Leicester, 9-time English champions, faces the challenge of tight budgets that requires keeping the best players on the field.

According to Andy Shelton, Leicester’s head of sports science, strength and conditioning, any competitive edge is worth the investment.  "It gets more competitive every year and our focus must be on helping our players stay injury-free for longer," he told the BBC. "When we have our key players available against the top European sides, we can compete and we will win, so the question is how do we keep key players on the pitch?"

Metrifit Predictive Analytics

Factoring in variables like fatigue, stress, sleep and training intensity into a predictive algorithm can yield what may have been hidden trends and combinations that cause injuries. 

“Similarly we also collect data on previous injuries that they had and what they are doing in the gym, ­basically everything they do from when they walk in the door of the club in the morning and leave in the evening is collected,” Shelton added. “The aim is to be able to affect a player’s lifestyle through the week. For example, if they recorded a very good night’s sleep, then their risk of injury could go down from ‘predicted injured’ to ‘not-predicted injured.’”

Some coaches and trainers still feel that using predictive analytics to create an injury model based on volumes of underlying data seems a little over the top.  But if your job is to develop healthy, productive athletes that win, then any tool that provides an edge is worth a look.

"Traditional baseball types tell me to just give up, that this is a waste of time because injuries are mostly bad luck,” Conte commented. "Twenty-five years ago no one listened to Bill James either."

Andy Shelton agrees, "There is no point in collecting stats unless you can know what to do with it. But by predicting things before they happen is where we can make gains, and considerably enhance performance."

Follow @metrifit

Using Rate Of Perceived Exertion As A Training Metric

Coaches invest hours devising training plans that will push their athletes just to the edge, but not over.  Overtraining leads to burnout and injuries but going easy won’t get the right results.  The challenge of walking this fine line is truly understanding the intensity and workload being placed on athletes, whether its real or perceived.  Objective, wearable technology has helped in the form of heart rate and GPS monitors, but can be expensive and doesn’t capture a true sense of the player’s experience.  As an alternative, sport scientists have recently found that a self-reported rate of perceived exertion (RPE) can accurately capture the workload experience.

Two sports that are learning to rely on RPE, soccer and swimming, represent two very different training styles.  Soccer coaches spend considerable time on team drills and scrimmages while interspersing physical fitness into the sessions.  Swimmers, while part of a team, primarily focus on individual times and skills.  Despite the differences, researchers have found plenty of evidence that the athlete’s opinion of their workout difficulty is valid and reliable.

Back in the 1960s, Gunnar Borg, psychology professor at Stockholm University, created the RPE scale, now respectively called the Borg scale.  The original version asked athletes to rate their level of exertion on a range of 6 to 20, with 6 being “very light” and 20 indicating “very difficult.”  While the 6-20 range may seem an odd choice, it actually has some logic.  Borg found that if the rating is multiplied by 10, there is a high correlation to the athlete’s heart rate at that moment (i.e. a rating of 12 typically corresponds with a HR of 120).

Borg also added a 10 point scale, known as the Category(C) Ratio(R) scale or CR-10.  This produces ratings of 1-10 and is used not only in sports training but also in clinical settings to estimate levels of pain.

Last year, Spanish and Italian researchers compared the workout assessments of 28 semi-pro soccer players.  For an objective measure, they captured heart rate history and tracked their distance travelled with GPS devices.  Then, after each training session, they asked the players for their RPE using the Borg CR-10 scale.  They found a very high correlation between the HR data, the distance travelled and the players’ RPE ratings.

“Being easy to perform and inexpensive compared with HR-based methods, sRPE should be regarded as a viable way to track internal load in training setup in soccer,” concluded David Casamichana, sport scientist at the University of the Basque Country and lead author of the study published in the Journal of Strength and Conditioning Research.

But what if young athletes report a “less than truthful” RPE in an attempt to either impress or fool their coach?  In the same way, what if the coach’s interpretation of a hard workout does not match with a player’s reaction to it?

Renato Barosso, of the School of Physical Education and Sport at the University of São Paulo, Brazil, gathered together 160 swimmers of different age-groups and different competitive swimming experience, and nine of their coaches.  Looking at their training plan for the day, the coaches were asked to rate the workout using the CR-10 RPE scale, prior to the session.  Then, 30 minutes after the training, the swimmers were asked for their RPE to see how well it matched the coaches’ estimates.  Athletes were divided into three age groups, 11-12, 13-14 and 15-16, while the workouts were classified as easy (RPE less than 3), moderate (3-5) or difficult (greater than 5).

As might be expected, the agreement between coach and swimmer was higher for older swimmers and lower for younger swimmers.  While the coach’s estimate of intensity was assumed correct, the researchers found that the swimmers aged 11-14 ratings differed across all three categories, easy-moderate-difficult.  The oldest swimmers only disagreed with their coaches at the difficult level.

The study will be published in the International Journal of Sports Physiology and Performance.

So, while RPE can be trusted for an accurate estimate of training difficulty, it would benefit both athlete and coach to gather all available data in one online training system for comparison and analysis.  Being able to chart RPE over time against more objective measures like HR, repetitions or activities would enable better training plans.

Follow @metrifit

The Right Amount Of Training Time For Single Sport Kids

Most parents and coaches have heard of the growing problem of overuse injuries in youth sports but few are probably aware of the startling statistics.  In the U.S. alone, high school athletes account for an estimated 2 million sports-related injuries every year, while athletes under the age of 14 suffer 3.5 million sports injuries. 

Unfortunately, the U.S. CDC estimates that almost half of these injuries are preventable and occur because of overuse of the same muscles or bones by kids who are specializing in a single sport.  Now, an update to a long-term research study confirms the need to better monitor the type and amount of training for young athletes.

Dr. Neeru Jayanthi, associate professor and medical director at Loyola University, sees quite a few middle and high school athletes at the Loyola sports medicine clinics in Chicago.  Back in 2010, he designed a long-term research program to monitor the types of athletes and injuries that the clinic was treating.  Since then, Jayanthi and his colleagues have enrolled 1,206 athletes between the ages of 8 and 18 into the study and will be following each one for up to three years.

In an April update, he reported that there had been 859 total injuries, of which 564 were diagnosed as overuse injuries with 139 of those being considered serious that kept the athlete sidelined for one to six months or longer.

Along with their health, the team also kept track of each athletes training schedule as reported by their parents.  Combining this data with the injury report, an eye-catching statistic jumped out at them. Those young athletes who spent more hours per week training or competing than their age, in a single sport, were 70% more likely to pick up a serious overuse injury.  So, for example, if a 14 year old soccer player spent more than 14 hours per week in just soccer, they suffered an injury rate over two thirds higher than those players who practiced less than their age.

"We should be cautious about intense specialization in one sport before and during adolescence,” Jayanthi said. “Among the recommendations we can make, based on our findings, is that young athletes should not spend more hours per week in organized sports than their ages.”

Not only is the amount of hours or practice vital but also the balance between organized, structured training versus just free play like pick-up games.  The latest research showed that if this ratio of structured to unstructured time exceeded 2:1, the injury rate went up significantly.

The medical evidence has become so overwhelming that a group of leading medical authorities, including the American Academy of Orthopaedic Surgeons and the National Athletic Trainers Association, have mounted a campaign called Stop Sports Injuries to make athletes, parents and coaches aware of the dangers of early sport specialization and overtraining (see video below).

“Kids often receive pressure from their parents or coaches to be the best in one given sport, when in reality participating in free play and a multitude of sports from an early age is the best strategy to create an outstanding athlete,” said William Levine, MD, Chair of the STOP Sports Injuries Advisory Committee.

Keeping track of an athlete’s training sessions is critical to prevent crossing the threshold to overuse.  Using an online training diary system is strongly recommended to log not only the hours but also the type of activity, the body’s reaction and recovery and the progress of results over time.  Coaches and clubs can also use the diary system to monitor entire teams and player populations enrolled in long-term athletic development plans.

Follow @metrifit

Using Training Data For Long Term Player Development

Imagine if you were given the task to find the next John Terry, Andy Murray or Katie Taylor.  You know that they’re out there somewhere kicking a ball, returning a serve or winning a bout among thousands of other kids their age.  While some look like future champions at age 7, it’s unknown what they’ll be like at 17.

Finding a group with some genetic gifts and then developing them through years of physical and mental growth demands access to new tools with one secret ingredient, data.  Just ask Ben Smith and Marco Cardinale.

In a recent interview with the Big Data Insight Group, Ben Smith, Head of Development Performance Systems for Chelsea Football Club, commented, “The professionalisation of sport has been dramatic over recent years and it’s only going to continue. There’s a huge amount of money and drive within the industry today; the rewards are massive for those getting things right and they’re substantial for getting it wrong – data analytics helps us ensure we do the former and avoid the latter.”
In this talent identification and development process, breaking down the data on hundreds of prospective youth players falls into two categories, quantitative and qualitative.  The quant side measures and tracks objective data points from devices worn by the player or observed metrics like timed drills and strength workouts.  Just as important, the qualitative data tracks the subjective opinions and observations of the player and coaches related to the perceived progress of their daily training.

Collecting gigabytes of data is only step one.  Without a way to summarize and visualize the data in a format that is easy for coaches and players to understand, the effort is wasted.  “Numbers are really, really dry and people from a coaching background, even the modern coaches, are not often data driven,” according to Smith. “If you can present the numbers in a way that means they quickly understand its direct relevance to the things they’re trying to achieve then they will appreciate the significance of what it’s telling them.”

While managing athlete development data for one sport is difficult, coordinating progress across multiple sports introduces an even greater challenge.  That job falls to Dr. Marco Cardinale, Head of Sports Science and Research of the British Olympic Association.  He recently described to the MIT-Sloan Management Review some of the complexity and hurdles Team Great Britain has to overcome to keep each sport’s program moving forward.

“The biggest problem we have in sport is the difficulty in collecting data,” said Cardinale. “The real analytics we are interested in is the ability to understand what athletes do on a daily basis to be able to affect their training programs, and that’s where the difficulties occur. It’s very sport specific.”

Like the Chelsea staff, Cardinale understands the need for athletes and coaches to track the ebb and flow of daily training.  “In my view, the coach and the athlete are the main unit able to deliver success,” he commented. “I think the biggest edge will be if they understand themselves better. In too many sports, athletes train either too much or not too much, and it’s because they have to gauge what they do on a daily basis against their feelings or what the coach sees. I think if they have more data about themselves, they can have an edge because they can be smarter in the way they train.”

Improving performance with data analytics is not a quick-fix solution. Identifying patterns and trends related to training requires multiple data points over an extended period and a commitment to its long-term use. As Cardinale concludes, “It’s a long journey, which means a club or an institution really need to invest in a project for at least a good three to four years. The power of data resides in good longitudinal information rather than snapshots.”

Be sure to visit Metrifit and Follow @metrifit

To Know Where You're Going, You Have To Know Where You've Been

What happened out there? You thought you were ready. You thought your training went well last week. You thought your pre-competition routine was the same as always. Now you’re wondering why you hit the wall early and just had an off day. 

Consistently performing at a high level depends on creating the right combination and pattern of training that yields the best outcome. Even a small change to that ideal routine can result in a poor performance. Finding that wrong turn requires retracing your steps through your recent training sessions.

Unfortunately, many athletes lack a system to capture not only the quantitative data but also the qualitative information about their mood, motivation and daily activities that may have affected their results. In all of the noise of today’s high-tech monitoring devices, the simplicity of a training diary often gets overlooked.

So, what exactly is a training diary? It can range from a paper notebook with an athlete’s thoughts about the day’s practice to a sophisticated, online app. For either version, the key ingredient is consistent and accurate data. Without an athlete or coach entering data, the diary is like staring at a map with no roads.

Recently, human performance researchers at Dublin City University (DCU) studied the effectiveness of using training diaries for young Gaelic footballers as a way to assess their overall training load. Without proper management of their time and activities, young athletes can suffer burnout from overtraining.

Siobhán O’Connor, a researcher and graduate student at DCU, and Professor Noel McCaffrey gathered 162 players from U14, U18 and adult teams to measure not only the response of players to using a diary, either paper-based or online, but also to validate that what the players self-reported was an accurate reflection of their actual training.

Previous research has shown that athletes prefer easy and efficient data entry for a diary to succeed. O’Connor designed a format that, on average, took the players just under 4 minutes per day to fill out. Initially, the paper and online versions received about the same participation rate but when e-mail or text reminders were sent out for the online version, the players use of the online version increased substantially.

Filling diaries with the right information is just as important as timeliness. As they say in the computer world, “garbage input produces garbage output.” To check this, a subset of the players also wore accelerometers and/or SenseCams to objectively capture data about the training sessions. When this data was compared with what the players actually recorded in their diaries, there was a 95% agreement, confirming that the players could accurately self-report their own data.

O’Connor is encouraged by the results, “This study will benefit Gaelic Footballers throughout Ireland and beyond by enabling them to quantify their training load in a quick and easy manner.”

Although training diaries were initially designed with amateur or semi-professional sports enthusiasts in mind, online diaries combined with communication portals are being utilized more and more by professional organizations and elite athletes.

Of course, the payoff for athletes to entering this information is being able to quickly review the data and ensure consistent performance improvement. That’s where online diaries shine, especially those that can analyze the data and identify cause and effect patterns. Being able to understand how your daily habits contribute to your results makes it all worthwhile.

Be sure to visit Metrifit and Follow @metrifit