In my Fantasy Football for Smart People book series, I advocate maximizing value by seeking safety and scarcity early in drafts, then searching for upside later. That was the general idea behind my early-March article suggesting that Bucs running back Doug Martin deserves consideration as the No. 1 player off of the board in PPR formats. I said it. I meant it. But I didn’t act on it, because I drew the top overall pick in a subsequent mock draft and opted for Adrian Peterson.

But since that time, the Doug Martin bandwagon has gained traction. Recently, Shawn Siegele pointed out that Martin posted the third-most yards from scrimmage for any rookie running back in NFL history, behind only Edgerrin James and Eric Dickerson, with Clinton Portis, Marshall Faulk, and Barry Sanders rounding out the top six. And no one touches Martin on the Running Back Similarity Score app.

Take a look at the median PPR projections for the top 25 non-rookie running backs in terms of current ADP:

Due to massive shifts in workload, we can effectively remove the numbers for Miller, Wilson, and Ivory. But otherwise, analyzing the comps produced by the league’s top runners has a lot of merit. And look at Martin sitting at the top, a full 2.5 points ahead of second-place Ray Rice.

That 20-point PPR median projection is nice, but it doesn’t tell us anything about the distribution of Martin’s comps. There’s a difference between 20 comparable players all centered around that 20-point mark versus just a few such comps who scored close to 20 points (with the rest dispersed either well above the mark or well below it).

To measure the deviation in running back comps, I’ve implemented the same volatility metric I created to determine that Aaron Rodgers is a much better option than Drew Brees. The calculation is simple: (High Projection – Low Projection)/Median Projection. I’m basically looking to simply uncover the deviation in comps for each running back to determine how much volatility he possesses.

A few notes:

• Look at Martin killing it! Not only is he the top-projected running back, but he’s also the safest. The difference between Martin’s Volatility Rating and that for the No. 2 running back—Marshawn Lynch—is larger than that between Lynch and the next four backs.

• It’s interesting that the top backs are all in situations in which they figure to see really heavy workloads. You don’t see players like Murray, McFadden, or McCoy whose workloads aren’t as set in stone. It’s worth noting that all three of those players, although relatively young, missed significant time in 2012.

• Murray in particular is intriguing because he actually has a median PPR projection of 15.6 points—fourth-best among all backs. His ceiling is 18.8 points, but his floor is just 8.1 points. Looking at his comps, though, you see some very interesting names—Frank Gore, Marion Barber, Matt Forte, Marshawn Lynch, Edgerrin James, and Steven Jackson, among others.

• Murray’s wide distribution of potential outcomes fits with what we know about him as a player and with his current situation in Dallas, but I think he’s in for a big year if he can remain healthy. At DallasCowboys.com, I listed four reasons Murray is primed for a breakout year. Among them were his size/speed, the potential mislabeling of him as ‘injury prone,’ and the fact that he’s already proven to be effective when on the field.

• The running back who scares me most is Jones-Drew. With just a median projection of 8.6 points, the 6.2-point deviation between his high and low comps is really large. That would be okay if Jones-Drew were a late-round pick and you were searching for upside, but he’s getting drafted in the middle of the second round—eight spots ahead of Murray. The numbers fit with the idea that Ahmad Bradshaw is a cheaper version of MJD.

Subscribe for a constant stream of league-beating articles available only with a Premium Pass.