Fantasy Football: The Myth of Overworked Running Backs
Page 1 Page 2
A more practical example of ‘regression toward the mean’ deals with the attitudes of coaches. Although the mindset of NFL coaches seems to be altering of late, it is still not uncommon for coaches to be disciplinarians. Why? Because yelling at players does seem to produce better results.
Coaches tend to holler at players after poor results. However, NFL players are professionals–the majority of the time they don’t screw up. So, if Player X catches 98% of all passes thrown his way, he is very likely to follow up a dropped pass with a catch. When the coach yells as Player X after each drop and then notices he follows it up with a catch, he attributes the recent success to his discipline.
The problem, though, is that Player X is probably no more likely to catch the ball after being yelled at than after receiving no discipline. If Player X “is” a 98% catch rate guy, he is very likely to catch any given pass, regardless of the result of the last one thrown his way. He is incredibly likely to catch a pass after receiving discipline not because of the discipline, but because of ‘regression toward the mean’–the discipline is an irrelevant byproduct of the previous drop.
The fact that the player followed the drop with a catch is because he generally follows each previous play with a catch. If the coach doesn’t attribute the majority of Player X’s success to his silence (the result of catches), then he is not justified in attributing a relationship between Player X’s success and his own discipline (which comes after drops).
It may seem as though I am rambling here (and perhaps I am), but this situation is directly analogous to the effectiveness of a running back following a season with a heavy workload. Is the running back likely to be less effective? Of course. But is this a direct result of the high number of touches? Not at all.
In the same way that Player X is likely to regress toward the mean (catch the ball) following a drop (regardless of the coach’s discipline), a running back is likely to regress toward the mean (less touches). Less touches generally means less production.
Further, running backs who garner 370+ carries in a season could not possibly have been injured for more than a game or two. They are more likely to get injured the following year (as compared to the season with a lot of touches) not as a result of the heavy workload, but because of regression toward the mean–running backs often get injured. A running back who missed zero games one season can, at best, only tie this mark the following season.
Thus, while the production of a running back coming off a season with a heavy workload is likely to decrease, it is not a legitimate reason to avoid that player in fantasy drafts. The (probable) decrease in production is due to the previous season being a statistical outlier (a result which is unusually far from the mean).
The best way to look at the situation is this: what is the running back’s chance of having production which is comparable to the previous year? It is actually the same as it was prior to the start of the previous season, i.e. the workload has no noticeable effect on his ability to produce.
For example, if a running back has a 20% chance of garnering 2,000 total yards in a season, that percentage remains stable (assuming his skills level does the same) from year to year. Thus, the chance of this player following a 2,000 yard season with another is unlikely, but not due to a heavy workload (almost a necessity for such productive output), but rather the fact that he only had a 20% chance to do so from the start. We wrongly (and ironically) attribute the decrease in production to the player’s prior success when, in reality, no such causal relationship exists.
So you can draft Chris Johnson (358 carries), Steven Jackson (324 carries), and Maurice Jones-Drew (312 carries) with confidence. Their 2010 production may not match that of 2009, but you can be sure it isn’t due to the number of touches they garnered last season.
Page 1 Page 2