It's an extremely misleading claim, for many reasons. First, the ranking applies to both run and pass plays on 3rd and 1. When you remove the pass plays on 3rd and 1, the ranking goes up to 27th. Then when you stop and think, "Why just 3rd and 1? Isn't 3rd and 2 also short yardage?" you see the ranking go up to 22nd. Then you think a little more, and you realize this completely leaves out 4th down -- an even more critical short-yardage situation than 3rd down. Now the ranking goes up to 17th.
All of a sudden, you realize our short-yardage run game wasn't the league's worst after all, but instead was middle of the league. Maybe you still think, "Well that was the McFadden effect. We dropped from the 10th-best short-yardage running team in 2014 down to 17th in 2015 because McFadden replaced Murray." You can find out if that's true by comparing their conversion percentages on short-yardage runs.
2014 Murray 19 of 26 (73.1%)
2015 McFadden 13 of 17 (76.5%)
Add in runs from the 1- or 2-yard line on 1st or 2nd down (which are also short-yardage plays), and it looks like this:
2014 Murray 25 of 32 (78.1%)
2015 McFadden 15 of 21 (71.4%)
I think it's important that the only thing separating the short-yardage performances of the two backs is their 1st- and 2nd-down goal-line runs. It tells me that there must have been something different about the goal line runs, but what? Did Murray just have more of a nose for the goal line? That sounds good, but it doesn't make sense that he wouldn't also have had more of a nose for the first down marker. So it must have been something besides the RB. I think it was the QB, and the different ways that defenses approached short yardage, based on who the QB was. I think that near the goal line in 2015, defenses were able to concentrate their resources on stopping the run because they didn't have a large area of field to cover. In 2014, we had a QB and a #1 WR who were able to exploit defenses near the goal line when those defenses cheated toward the run.
I can test this theory by comparing the early part of the season with the rest of the season. How often did we run early in the season, and how successful were we, before defenses started taking away the run? Was there any difference compared to the rest of the year? Well, yes, there was a huge difference both in how often we ran, and how successful we were.
from 5-yard line or closer
% of runs resulting in TD
2014 (all 16 games) 9 of 17 (53%)
2015 (games 1-6) 5 of 7 (72%)
2015 (games 7-16) 0 of 3 (0%)