NFL Injuries: How Often “Probable” Players Play, And Why Concussions Matter (2 of 4)

October 20, 2010 - by John Ezekowitz

Last week, we introduced our first post in a series of analysis on NFL injury data. In short, we wanted to quantify how likely a player is to play based on his injury status, injury location, and position.

In the last post, we explained our methodology and produced some headline conclusions. In this one, we dive more deeply into what the data says about players listed as “probable” on the injury report, and also shed some light on the raging current debate about concussions in the NFL.

The “probable” injury status is by far the most common. After cutting our data set to starters only, we found that players listed as probable made up over half of the observations. Across all probable observations, players played 73 percent of the time. There was, however, wide variance in the likelihood of playing across the different positions and the various types of injuries.

Probable Playing Percentages: Best And Worst

The likelihood of playing relative to the mean chance of playing was normally distributed, with one key outlier which we will address. In the tables below, I have summarized the five injuries with the highest likelihood of playing and the five with the lowest:

Most Likely to Play
Injury Location	Percent Played
Back	82.76%
Neck	82.14%
Chest	81.58%
Hand	81.40%
Wrist	80.56%

The results as a whole are eye opening. Players listed as probable with these top five injuries played more than eight out of every ten times in the dataset. This seems high, even though we know that probable is the injury status with the greatest likelihood of playing. These players were about 15 percent more likely to play than the average “probable” player, a significant difference.

Least Likely to Play
Injury Location	Percent Played
Calf	66.67%
Lung	66.67%
Hamstring	62.26%
Achilles	62.07%
Concussion	56.67%

Even more interesting than the injuries from which players most often play are the ones where they most often are not able to play. Concussions and hamstring injuries are two of the most common football injuries, and the fact that they are in the bottom decile of likelihood to play implies large consequences for fantasy players looking at deciding who to start.

Hamstring injuries were in fact the most common injury, and even a slight hamstring pull, it seems, is enough to keep a player from playing more often than almost any other injury.

Concussions: The Most Serious Category By Far

The outlier in the “probable” portion of the dataset is concussions. Concussions, and their negative effects on players’ long-term health, have been in the news with increasing frequency over the last few years. Looking at this issue from a purely football perspective (as opposed to a medical one), it appears that there really is no such thing as a minor concussion.

If a player was listed as probable with a concussion, he was 22 percent less likely to play than the average “probable” player, and seven percent less likely to play than a player with an achilles injury, the injury with the 2nd highest incidence of games missed.

We are not medical experts or doctors, and thus cannot conclude if too many players with concussions are still playing, as some have contended. Yet it seems clear that concussions are more serious than any other “probable” injury, by a wide margin. The next time you see a player listed as “probable” with a concussion, remember that his actual chances of playing are as if he were listed as “questionable”: around 50 percent.

Probable Playing Rates By Position

The next table shows the relative likelihood of playing for “probable” players by position:

Position	Percent Played
Quarterback	80.81%
Wide Receiver	78.50%
Fullback	77.65%
Guard	75.93%
Running Back	75.80%
Defensive Tackle	75.00%
Center	73.77%
Defensive Back	71.48%
Tackle	71.09%
Tight End	70.19%
Defensive End	69.34%
Linebacker	65.13%

Somewhat unsurprisingly, quarterbacks were the most likely to play when listed as probable. We suspected this to be true, but it’s always good to have quantitative evidence to back it up. Starting QB’s are the heart of a team’s offense, so it makes sense that if they are at all able to play, they would. Quarterbacks listed as “probable” should play the most often out of any position at any injury status, and indeed they do.

Starting wide receivers have also become more important in the passing-heavy NFL. Thus it makes sense that they would play if they at all could, whereas a running back in a pass-first team may be held out more often in favor of his backup as a conservative measure, even if he could potentially play.

The other noticeable trend here is that defensive positions on average play less often than offensive positions. Only defensive tackles play more often than the average “probable” player, and linebackers and defensive ends are the two positions least likely to play. We might posit that defensive players are more “replaceable” than their skill-position offensive counterparts. Or, it could be that over the last two years, defensive injuries listed as “probable” have been more serious on average than those for offensive players, forcing defenders to miss more games. Without more medical information, we cannot know for sure.

Coming Next

Our upcoming third post next week will examine the specifics for the “questionable” injury status. On average players listed as “questionable” play about half of the time, but again there is wide and interesting variation in their likelihoods of playing.