Most of us are probably familiar with the concept of “regression to the mean”. It seems you can’t even start discussing someone who had a big year last year without the phrase being tossed out twice before you stop to take a breath.
According to our good friend Wikipedia, “regression toward (or to) the mean is the phenomenon that if a variable is extreme on its first measurement, it will tend to be closer to the average on its second measurement—and if it is extreme on its second measurement, it will tend to have been closer to the average on its first.” Basically, if someone has an outlier season, their next year is probably going to be closer to normal.
While running queries into NFL history, (what, how do you spend your weekends?), I discovered something startling. Improbably, I stumbled upon a group of players who proved to be virtually regression proof. A group of players who were highly likely to follow up one huge year with another.
The Regression-Proof Eighteen
Who were these players? I don't want to beat around the bush: the key to winning in fantasy football appears to be drafting running backs who had between 357 and 369 carries. Don’t believe that they’re regression proof? Let’s just look at the data.
There have been 18 times in NFL history where a back finished a season with between 357 and 369 carries. Rather than perform any tricks, here's a chart with all 18 backs, the year they finished between 357 and 369, where they ranked that year in fantasy, and where they ranked the following season.
| Player | Year | Rank in Year N | Rank in Year N+1 |
|---|---|---|---|
| Walter Payton | 1979 | RB1 | RB4 |
| Earl Campbell | 1979 | RB2 | RB2 |
| Earl Campbell | 1979 | RB12 | RB27 |
| James Wilder | 1985 | RB8 | RB28 |
| Herschel Walker | 1988 | RB3 | RB11 |
| Emmitt Smith | 1991 | RB3 | RB1 |
| Emmitt Smith | 1994 | RB1 | RB1 |
| Curtis Martin | 1995 | RB2 | RB4 |
| Eddie George | 1997 | RB11 | RB11 |
| Terrell Davis | 1997 | RB2 | RB1 |
| Curtis Martin | 1998 | RB8 | RB8 |
| Curtis Martin | 1999 | RB8 | RB8 |
| Edgerrin James | 1999 | RB2 | RB2 |
| Rudi Johnson | 2004 | RB8 | RB7 |
| Edgerrin James | 2005 | RB5 | RB19 |
| Tiki Barber | 2005 | RB5 | RB7 |
| Adrian Peterson | 2008 | RB3 | RB2 |
| Chris Johnson | 2009 | RB1 | RB4 |
That’s a remarkable 15 out of 18 who repeated as a top-12 running backs, including eight top-5 finishes. Since 1988, that success rate is even higher, with 13 of 14 repeating. The lone exception during that span was Edgerrin James, who that offseason moved from Indianapolis, (one of the best offenses in the NFL), to Arizona, (one of the worst).
Since 1991, James was the only person to see his ranking decline by more than 3 spots. The second-biggest drop came from Chris Johnson, who fell from first in 2009 all the way down to... fourth in 2010. Nine of the last thirteen backs in this group have either held constant or even improved in their second season.
A Conspiracy Emerges?
So why haven’t we heard more about this magical carry range that all-but-assures future success? Because it’s not a real effect. Yes, the data is real enough, but it looks impressive because I designed it to look impressive. In short, I cheated.
There are lots of ways to cheat with data, but the easiest is to peek at it ahead of time. I set my lower limit at 357 because moving it to 356 only adds Stephen Davis’ 2001 season, and Davis was a bust in 2002. A minimum threshold of 355 carries adds Ahman Green in 2003, Jerome Bettis in 2001, and Thurman Thomas in 1993, and all three of them declined in their next year.
Similarly, moving the top threshold to 370 carries only adds two seasons. Both of them, (Christian Okoye in 1989 and Shaun Alexander in 2005), were followed by two of the biggest busts in modern NFL history.
Obviously there’s nothing magical about a specific number of carries that should make it that much different than just one carry more or fewer. A back with 357 really doesn’t receive that much more wear and tear than a back with 356, and we should be skeptical of a theory that suggests otherwise.
So what’s the point of this article? Many who have considered drafting DeMarco Murray have been told about the so-called “Curse of 370”, or the idea that running backs who receive too much work in one year, (in this case, more than 370 carries), are somehow “cursed” and doomed to decline.
The problems with that analysis should be apparent, because they’re the same as the problems with my satirical “Blessing of 357”. Why was 370 chosen as the breakpoint? Because someone peeked at the data ahead of time. As you saw, backs in the 357-369 range were terrific, and including them in the sample would greatly weaken the appearance of a curse.
The Truth Comes Out
At the end of the day, overuse might well be a problem for running backs, but the way to demonstrate it isn’t through manipulating endpoints to generate more compelling data sets.
From 2009 to 2013, there were 60 total top-12 fantasy finishes by running backs in standard scoring. Of those 60 seasons, 19 featured more than 300 carries. Of those 19, just ten, or 53%, managed to repeat as a top-12 fantasy back the next year. Getting a large workload makes it difficult to repeat for fantasy purposes.
At the same time, there were 41 top-12 fantasy finishes by a back with fewer than 300 carries. Of those 41, only 16, or 39%, managed to finish in the top-12 again the next season. So having a smaller workload certainly doesn’t make it any easier to repeat for fantasy purposes.
Indeed, the key takeaway here is that it’s very difficult for running backs to remain in the top 12 year after year. That’s true if you look at guys with high workloads, and it’s true if you look at guys with low workloads. It’s true if you look at guys with high yard-per-carry averages, and it’s true if you look at guys with low ypc averages. No matter how you slice the data, repeats are hard. (Unless, of course, you happen to get between 357 and 369 carries.)
There’s a good chance DeMarco Murray fails to live up to expectations this year. He’s changing teams, which is always risky. He’s playing for a coach who seems intent on rotating his running backs, which caps his upside. And, in general, it’s just tough for running backs to remain at the top of the heap. It's a physically demanding position, and there's always someone new coming to knock you off your perch.
But of all the reasons why DeMarco Murray might fail in 2015, worries about crossing some all-powerful workload threshold last year should be near the bottom of the list.
