San Francisco 49ers owner John York claims that a three-game preseason followed by a 17-game regular season may decrease injuries. Is this true?
On Nov. 18, 2019, a story was published by The Athletic in which Dr. John York, the owner of the San Francisco 49ers and chairman of the NFL owner’s health and safety advisory committee, relayed that a 17-game regular season would have “minimal” changes on the health of the league’s athletes.
Dr. York told The Athletic that “the engineers and statisticians…[have looked at] what might happen in a number of different scenarios [including a schedule expansion by one game], the change in injury rates, either a little bit plus, a little bit minus, it was not particularly significant.”
Could this possibly be true? Is it possible that replacing one preseason game with an additional regular-season contest could cut down on injuries league-wide, or at the very least, maintain the status quo?
It instinctively feels counter-intuitive.
Dr. York is quoted as saying during the NFL Helmet Challenge symposium, “Often times we have more injuries in preseason games than during regular-season games.” The author of the The Athletic article continues by writing, “As a result, taking away one preseason game and replacing it with a regular-season game across 32 teams might reduce average injury risk, according to York.”
However, immediately a few counter-points come to mind.
First, the sample size of injury data gathered during the preseason assuredly pales in comparison that that collected during the regular season. In all, there are 64 preseason games played prior to the start of every season compared to 256 regular-season contests.
If the data Dr. York is referring to analyzes preseason and regular season data as individual sets year after year – and it’s unclear if that is the case as the data hasn’t been released publicly at this point – it wouldn’t be that unusual to find that a greater number of injuries occur during the preseason relative to the regular season as the sample size would be significantly smaller. Small sample sizes introduce a higher level of variability to the equation and should decrease confidence when interpreting the data.
To cite numbers provided in The Athletic article, ACL injuries decreased from 28 during the 2018 preseason to 16 during 2019. Additionally, MCL injuries were reduced from 36 to 29. Yes, those numbers represent a decrease in the raw total number of injuries, however, the discussion should then question whether or not that reduction is scientifically significant.
There are two types of scientific significance: statistical and clinical.
To try to simplify a complex topic: Statistical significance occurs when it can be concluded that the change that occurred is highly unlikely to be caused by chance alone. Statisticians determine whether or not a finding is statistically significant if analysis reveals that there are less than 5% odds that the finding was caused by chance (this is known as the alpha-number; an alpha less than .05 signifies statistical significance).
Clinical significance occurs when it is determined that the change that occurred would have a legitimate clinical impact for the individual or community (i.e. the NFL). A finding can be statistically significant, but not clinically significant.
In regard to the numbers provided above, it’s unclear if those values are either statistically or clinically significant. For all we know, those reductions may just be due to chance and chance’s influence increases when sample sizes are small.
Second, exchanging a preseason game for a regular-season game isn’t going to happen in a vacuum. Recent reporting by CBS.com indicates that the additional regular-season game will be played “out-of-market with a heavy emphasis on key international locales, like the United Kingdom, Germany, Mexico, and Brazil.”
Although no studies have been completed – or are publicly available, at least – in regard to the impact of travel on NFL athlete health, it would be safe to assume that an increase would likely result in a detrimental effect on health and risk for injury. Studies published in 2012 and 2018, discussed the negative effects of air travel, particularly when traveling from one coast to the other, on elite athletes as well as players in the NBA.
Increased air travel could negatively affect player health and performance while increasing injury risk secondary to decreased amount of preparation and recovery time during the week, decreased sleep amount and quality, and disruption of circadian rhythms secondary to drastic changes in time zones, to name just a few.
So is it possible that adopting a three-game preseason and 17-game regular season, in which one regular-season game occurs outside of the country, may result in fewer injuries across the NFL? Potentially, though on the surface it seems unlikely. Doing so would almost assuredly increase the already staggering amount of revenue league-wide and would help promote the game worldwide, however, would the decision be in the best interest for the players?
At the very least, further discussion, nuance, and data analysis are needed before a concrete decision should be made.