Injury Prone? Why We’re All Wrong

I’m an idiot.

The question was “DeMarco Murray or Arian Foster?”. The year was 2014. In my unwavering support for Killer Tofu, my response was “DeMarco Murray has never played 16 games”. It was a dumb thing to say.

Those who dabble in the depths of football hyperbole know that what I meant was, “because DeMarco Murray had never played a full NFL season in his first three tries, he was unlikely to ever do so.” The same logic is used all the time to rationalize why a player will get injured, or conversely why he won’t. I recently experienced the latter when commenting on Eddie Lacy’s chronic toe issue. I suggested that it is important for dynasty league owners to identify and acquire his potential backup before roster cuts, and I was met with cynicism and sarcastic quips of “how many games has he missed the last two years because of his toe?” The answer, as we all know, is zero.

Both the reasoning I used to incorrectly denigrate Demarco Murray and the logic used by this particular Eddie Lacy enthusiast are fallacious and counterproductive at best, and downright stupid at worst.

After the fall of LaDainian Tomlinson, we have been left with a wake of running-back-by-committees, specialized roles for unique talents, and more passing than we have ever seen. Leagues have widely adopted requiring only starting one running back; Shawn Siegele (@FF_Contrarian) has revolutionized the way people draft RBs; and The Ghost (@TheFFGhost) has implored implementation of Point Per Carry scoring to rejuvenate the relevance of, and save, the position. It’s madness; this is the world Bill Walsh imagined, and we are all just living in it.

When we look at the data for RBs of the past several seasons, we not only see a dropoff in the number of true workhorses, but we see a complete inability to stay relevant and on the field for more than a very short window. Here is the data for all RBs since 2009 to play in X number of consecutive seasons at any point where they were both fantasy relevant (PPR RB36 or better) and played at least 15 games:

While the small numbers on the right are striking, they shouldn’t be all that surprising. To sustain relevancy in the NFL at the RB position, and stay on the field at the same time for several consecutive seasons, is a daunting proposition. Astute fans know the 1’s at the bottom represent the singular active streaks of Marshawn Lynch, who has played through the following injuries over the past few seasons:

This raises several important questions:

• Were the people saying Lynch was likely to get hurt or decline in efficiency the past few seasons wrong?
• Does the data show that it’s fallacious to think Running Backs playing 16 fantasy relevant football games in season N are more likely to do so in N+1? Does it show that it’s actually less likely?
• Is it possible that “injury prone”, if applied blanketly without consideration to specific medical conditions, should be used exactly the opposite way that people currently use it?
• Would it be more intuitive to use the phrase “declination prone” and approach the position with an extreme hesitation to rely on continued relevance?

What jumps out in this chart isn’t the fact that only three running backs have both remained relevant and continually seen the field in any five straight years, it’s the fact that only 16 did it in any three straight years. If you play in a 16-team Dynasty league, that means any team can only reasonably expect to have had one healthy, relevant RB for any three year stretch in the last six seasons. The scarcity and attrition of the position can be overcome and rapidly adjusted to in redraft leagues, but the magnification in the Dynasty format is startling.

The next logical question is, “Does it even matter if RBs play every game since so few continually do it?” There doesn’t seem to be a tangible correlation to the widely held notion that a running back with a history of playing 16 fantasy relevant games makes it more likely he continues to do so. There also doesn’t seem to be a tremendously strong correlation of playing 16 games to being a fantasy relevant running back:

The six active streaks of six consecutive RB36 or better finishes belong to Darren Sproles, Steven Jackson, LeSean McCoy, Fred Jackson, Matt Forte, and Frank Gore. You already realize from the first chart that all six of these players have something in common: they all had a season in the last six years where they played less than 15 games. In fact, there are several instances where they were more valuable despite playing less games.

If their points-per-game was higher it meant a higher likelihood of winning those weeks they played. The weeks they didn’t play weren’t zeroes because of the opportunity to stream replacements. The position having the highest attrition and fragility is part of Siegele’s Zero Running Back thesis: Later in the season you can acquire Running Backs off waivers that have unexpected opportunity. In Frank Gore’s best season, he played 14 games. In Fred Jackson’s 2nd best season, he played 10. That’s not on a point per game basis, that’s based on their season-end ranks for all RBs. When you combine your best bench Flex player with their scores, you have a tremendous output from that spot in your lineup. To then get that player at a discount the following season because of a fear they will again play fewer games, without any medical reason to support the idea, is just stealing. In Dynasty formats, every single one of these players’ shortened seasons was a buying opportunity.

This isn’t permission to be completely reckless and dismissive of RBs missing time. It is concerning that only Willis McGahee and Jamaal Charles have had consecutive top-36 seasons after having broken a previous streak (Arian Foster and Ahmad Bradshaw could each do so this year as well).

Instead, this is permission to question the notion that a player who has not missed time is less likely to miss time in the future. Recency bias fuels the public perception of injury prone, yet a simple review of fantasy relevancy of running backs in consecutive seasons disputes the notion that if a player missed time recently for any reason, that he is necessarily more likely to miss time in the near future. Of the six players with active streaks of six consecutive top-36 seasons, only three (Steven Jackson, Gore, and Forte) have streaks of seven straight years. Only Jackson’s and Gore’s are longer than that, with Gore at nine, and Jackson at an absolutely remarkable 11 straight seasons of RB36 or better finishes; every single one of his career to date.

Look Past the Streaks

Recurring injuries happen; this is a fact.  Hamstring and foot injuries are prone to re-injury; one concussion renders the player more at-risk for a second concussion.  The idea that some players are more susceptible to injury than others is intuitive, but the popular notion that a recent streak of games played or games missed justifies a generic “injury prone” or “durable” label is absurd.

Advice: do not fall victim to streak-induced fallacies. When determining if a player is injury prone, check their injury history and be discerning, not absolute, as you evaluate injury risk.  Does a health-related pattern suggest that a player will continue the pattern, or that he is due to regress?  Only an idiot knows for sure.