NFL Injuries Before and After the 2011 Collective Bargaining Agreement (CBA)

TL;DR

This study analyzed NFL injury data from 2007-2016 to assess if the 2011 CBA's practice restrictions led to increased injuries, finding no significant detrimental impact attributable to the CBA itself.

Contribution

It provides an empirical evaluation of injury trends before and after the 2011 NFL CBA, addressing concerns about practice restrictions affecting player safety.

Findings

Injury counts increased from 2007 to 2016 but were not significantly linked to the CBA.

Conditioning-dependent injuries stabilized after an initial rise, showing no long-term increase.

No statistically significant injury increase was associated with the 2011 CBA according to Poisson models.

Abstract

The National Football League's (NFL) 2011 collective bargaining agreement (CBA) with its players placed a number of contact and quantity limitations on practices and workouts. Some coaches and others have expressed a concern that this has led to poor conditioning and a subsequent increase in injuries. We sought to assess whether the 2011 CBA's practice restrictions affected the number of overall, conditioning-dependent, and/or non-conditioning-dependent injuries in the NFL or the number of games missed due to those injuries. The study population was player-seasons from 2007-2016. We included regular season, non-illness, non-head, game-loss injuries. Injuries were identified using a database from Football Outsiders. The primary outcomes were overall, conditioning-dependent and non-conditioning-dependent injury counts by season. We examined time trends in injury counts before (2007-2010) and after (2011-2016) the CBA using a Poisson interrupted time series model. The number of game-loss regular season, non-head, non-illness injuries grew from 701 in 2007 to 804 in 2016 (15% increase). The number of regular season weeks missed exhibited a similar increase. Conditioning-dependent injuries increased from 197 in 2007 to 271 in 2011 (38% rise), but were lower and remained relatively unchanged at 220-240 injuries per season thereafter. Non-conditioning injuries decreased by 37% in the first three years of the new CBA before returning to historic levels in 2014-2016. Poisson models for all, conditioning-dependent, and non-conditioning-dependent game-loss injury counts did not show statistically significant or meaningful detrimental changes associated with the CBA. We did not observe an increase in injuries following the 2011 CBA. Other concurrent injury-related rule and regulation changes limit specific causal inferences about the practice restrictions, however.