Aaron Jones led the NFL (much to my dismay) in yards per carry in 2018, averaging 5.48 yards per carry courtesy of 729 yards on 133 carries. But there are many problems with focusing too much on a simple metric like yards per carry, which doesn’t tell you very much.
Every year, I like to conduct a study on all running backs who had above average YPC averages and then figure out how many carries we would need to remove to drop his YPC to below average. Why do I like to do this? Because it’s the offseason, and why not! But also because it helps to highlight and remind us how sensitive YPC is to outlier runs.
When I looked at the top running backs in the NFL last year, as a group, they rushed 8,153 times for 35,989 yards, a 4.41 YPC average. So for all running backs who had an above-average YPC number last year, how many carries would we have to remove to get them below 4.41?
Jones had a 67-yard run last year; take that away, and his YPC drops to 5.02. Remove his next 3 best runs — 33, 30, and 29 yards — and his YPC drops to 4.42. You need to take away his 5th best carry, an 18-yard rush, to drop his YPC to 4.31. So for Jones, the answer is 5 carries is how many you need to take away to get his YPC to below the threshold.
You might think, hey, this is biased against running backs with a low number of carries! Well, that’s the point! We should be skeptical of placing too much emphasis on a player’s YPC average if that running back doesn’t have a lot of carries. The “leader” in this statistic is actually a two-way tie between Todd Gurley and Gus Edwards. For Gurley, you need to take away his top 6 runs — which were only 36, 29, 26, 24, 24, and 23 yards — to get his YPC down to 4.36. This serves as a good way of noting that Gurley actually had a “down” year when it comes to YPC: he was only a couple of long runs away from an even more dominant season. Edwards had only 127 carries, but he also didn’t have many big runs: even if you take away his best 5 runs, his YPC would drop to 4.45; you need to remove his 6th best carry to get him to 4.34.
The graph below shows the results for each player who had an above-average YPC last year. The first four columns should be self-explanatory. The “TakeAway” column shows how many carries you need to take away to get that player’s YPC to below 4.41. Note that for Barkley, McCaffrey, and Chubb — despite each topping 5.0 YPC — that answer is only 3! And for Chubb and Barkley, removing their best 5 runs brings both players to a sub-4.00 YPC average. The next 10 columns show the 10 longest runs that player had. Then, the final 10 columns show that player’s YPC average once you remove each of his best X carries.
What stands out to you?