During the 2013 offseason, an NFL player went down with a serious injury almost daily—or at least it seemed like it.
An ACL here.
An Achilles there.
Day after day, seasons ended before they began. When regular-season injuries followed, talk of an injury epidemic began.
Yet how founded were the epidemic claims? What role did the availability heuristic—a cognitive bias where events that are easy to recall seem to occur more frequently than they actually do—play? More importantly, was 2013 an outlier? Should the NFL do more to prevent a recurrence in 2014?
As one might guess, easy answers to any of the above questions do not exist. Some of the data, however, does.
Acting on that data? That's an entirely different story—one sometimes involving selectively interpreting "objective" statistics.
Ultimately, should injury data be the ultimate determination as to whether health and safety improvements remain at the forefront in the NFL? Probably not. After all, how much does a player with a career threatened by a blown-out knee care about statistics?
In short: Not one iota.
To try to make sense of it all—spoiler alert: we won't—let's take a look, focusing first on the preseason ACL tear for simplicity's sake.
Concerning Surface-Level Calculations
According to an outstanding and insightful article by MMQB's Jenny Vrentas, the absolute number of preseason ACL tears in 2013 indeed increased to a level never before seen—27 in total, per Vrentas' and her colleagues' data collection methods. Vrentas also notes a previous range of 12 tears in 2005 and 2006 to 25 in 2008.
In other words, 2013's preseason saw over twice as many ACL tears as 2005.
That said, while the number 27 does represent the highest sum total to date, one needs to employ a few simple calculations to determine the precise significance of the otherwise-troublesome statistic.
For example, Vrentas mentions in her column that the preseason roster size increased from 80 to 90 during the 2012 offseason. Assuming all NFL teams carried a full roster for the entire preseason (an assumption made primarily for simplification, as it is surely not entirely accurate, nor does it account for player turnover), league-wide preseason rosters therefore amounted to 2,880 total players after the change.
Now, if one takes the 27 ACL tears that occurred during the 2013 preseason, divides them by the total number of players in the league at any one time and multiplies by 100 to get a total percentage of players suffering the injury, the result comes to 0.94 percent.
In other words, just under one out of every 100 NFL players suffered an ACL tear in the 2013 preseason.
Next, compare that figure to the preseason ACL tear low of 12 in 2005—when total NFL rosters numbered 2,560—or 0.47 percent.
Said another way, despite over twice as many injuries, the NFL's best and worst absolute "ACL risk" rates in the past nine preseasons differed by less than one half of one percent. That said, a simple statistical calculation shows that compared to a 12-per-2,560 injury rate, the increased rate of 27-per-2,880 attains statistical significance.
Incomplete—or Incompletely Understood—Data
Despite the above reasoning suggesting the ACL tear rate in 2013 probably statistically significantly differed from 2005 and 2006, it's very easy to poke gaping holes in the data—and thus the conclusions.
For instance, 2005 and 2013 represent the very extremes of total preseason ACL tears—not an overall trend. Elsewhere, if the absolute rate of preseason ACL tears happened to somehow increase steadily from 12 to 27 from 2006 to 2013, it would represent an annual increase of 2.14 tears—or an absolute ACL risk rate increase of roughly 0.1 percent per year. That is a relatively low number by any respect—though not worth ignoring, either.
Elsewhere, while the number of total players throughout the NFL remains constant throughout the summer, the proportion who meaningfully participate in camps—as well as the relative amount of activity of each player—might not always be constant.
That is to say, a player taking more snaps in NFL training camps is more likely to go down with an ACL injury, and therefore, the ACL-per-player rate needs to be adjusted to reflect only those players with regular participation.
Or does it?
Are players who participate in a greater percentage of a team's snaps more likely to be in better shape and thus at lower risk of injury? Does a greater participation rate actually correlate with a lower risk? If so, why did so many big-name players go down last year?
Most importantly, as asked previously, was 2013 an outlier? Was 2008? Even more mind-boggling: Were 2005 and 2006?
Lastly, when does a trend, indeed, become a trend? Don't many ACL injuries involve a significant amount of pure bad luck?
Changing Concussion Culture: Another Variable in the Fold
Earlier this month, Dr. Jesse David—an economist for Edgeworth Economics—completed one of the most interesting studies thus far on the NFL's injury issues.
One of David's main conclusions, reported by The Associated Press (via Fox Sports): a steady upward trend in "severe injuries"—or those holding a player out for eight days or more—to the tune of about 100 per year from 2009 to 2012:
The study says there were 1,095 instances of injuries sidelining a player for eight or more days in 2009 - including practices and games in the preseason, regular season and postseason - and that climbed to 1,272 in 2010, 1,380 in 2011, and 1,496 in 2012. That's an increase of 37 percent.
Yet the numbers are not the whole story. David also mentions the increasingly cautious treatment of concussions—not at all a bad thing, of course—may play at least some role in the increase:
(The study) shows that players with concussions missed an average of 16 days last season, up from only four days in 2005, while the length of time out for other types of injuries has been steadier.
David said ''you now have more severe injuries overall'' because of the hike in lengthy absences for reported concussions.
''Are the brain injuries actually more severe now than they were five years ago? Or is that players simply being held out longer for the same injury? That we can't tell from the data,'' David said. ''My guess is it's both, but how much of each factor, I don't know.''
The data certainly suggests some impressive conclusions. That said, one must look a little deeper in order to realize just how many unanswered questions still exist.
Continued Efforts Are Needed, Regardless of Data or Trends
None of the above data matters to an individual who is writhing on the ground in pain with a knee injury. With that in mind, the NFL and the medicine community will surely continue to work together toward the common goal of a safer game—and one that still maintains its edge.
The NFL and NFLPA already somewhat routinely make changes to the game in an effort to decrease injuries.
For example, in addition to increasing roster sizes from 80 to 90, the league implemented a number of policy changes in 2012, including—according to Vrentas—"reducing the length of the offseason program, eliminating two-a-days and cutting down the number of padded practices."
Personally, I think that's taking the wrong approach. You have a gap between preparation and competition level. And I think that's where you see a lot of injuries occurring. We get a lot of breakdowns. We get a lot of situations that players just aren't as prepared as they were in previous years, in my experience anyway.
Perhaps increased time off leads to less use of team-policed training regimens—not to mention organized, personalized and positional coaching—thus leading to more injuries due to poor conditioning or preparation.
Yet the data might suggest the opposite—depending on how one looks at it. The same article suggests the number of players on injured reserve at the time it was written was the lowest annual total since 2008—well before the 2012 changes.
With more fine-tuning and research, the trend will hopefully become definitely positive—rather than questionably so—especially if Philadelphia Eagles head coach Chip Kelly is the first NFL coach to be onto something.
In Nov. 2013, CSN Philly's Geoff Mosher profiled Kelly's now famous—but secret—"sports science" approach to training. Mosher writes:
Chip Kelly’s sports science is the most common reason given for the team’s health and general well being.
It appears the personalized smoothies, stretching sessions, post-practice massages, Tuesday practices, Saturday jogthroughs and softer, gentler training camp have yielded the intended results.
Less injuries—Philadelphia's five players on injured reserve was among the lower totals last year—and fresher-feeling players.
Just as some of the above statistics can come into question as a one-time anomaly, so can Kelly's one-year success with his sports science techniques.
Nevertheless, over the coming seasons, advancements in all of the relevant fields—research, diagnostics, individualized training and exercise science to name a few—will continue to surface.
The unclear but potentially complex relationship between injuries and rule or policy updates will also gather a larger foundation of data for continued analysis.
As that data then continues to pour in, the injury picture will probably initially become cloudier before it clears. Eventually, though, the true validity of the injury spikes and upward trends the NFL may be seeing—or lack thereof—will make itself known in one way or another.
All told, if decreasing injuries leads to more star players on the field rather than on the sidelines or in crutches, both the NFL and NFLPA—as well as fans—win, and the right combination of pre- and regular-season rules, policies and safety equipment to offer a true injury nadir may still exist—somewhere.
Someone just needs to find it.
Dr. Dave Siebert is a resident physician at the University of Washington who plans to pursue fellowship training in Primary Care (non-operative) Sports Medicine.