Why Brandon Marshall Is Undervalued, Julio Jones Overvalued
The distinguishing feature of the similarity score apps—the trait that gives them an advantage over traditional projections—is that they provide a range of potential outcomes for any given player. There’s a certain level of uncertainty in any prediction, and the apps account for that.
When implementing uncertainty into projections, the hurdle many fail to overcome is establishing a definitive prediction at all. It can be tempting to examine a player’s spectrum of potential outcomes and say, “Well, this guy could score 22 points per game, or he could score 11, but he’ll probably be somewhere in between. And this player might score 20 points per game, or he could score 9 points per game, but he’ll also probably be somewhere in between. So what the hell do I do?”
The most obvious solution is to generate an average for each player based on their comparables. That certainly has its advantages; taking the mean production of 20 very similar players will generally give you a quality projection.
However, doing that ignores a feature of the similarity scores that really makes them so useful—the magnitude of their range of outcomes. A player whose floor is 10 points and whose ceiling is 20 points could very well have the same mean projection as a player with a 13-point floor and 17-point ceiling. But there’s a difference between the two, and it needs to be captured.
That’s why I recently calculated the ceilings and floors for the elite players at the quarterback and running back positions, discovering that Peyton Manning is a sneaky safe pick and Doug Martin could very well be fantasy football’s top running back. Today, I’ll turn my attention to the wide receivers.
Below, I charted the ceilings for the top half-dozen wide receivers in terms of current ADP—the only six who are getting selected in the first two rounds. These are the ceiling projections for each player based on their top four comparables in each statistical category for PPR leagues—receptions, yards, and touchdowns. I threw out any comps who didn’t participate in at least six games.
It’s not too surprising to see Calvin Johnson with the highest projected ceiling. In reality, his ceiling is probably even higher than listed because it’s difficult to find comps for a player coming off of a 122/1,964/5 season.
While A.J. Green, Demaryius Thomas, Julio Jones, and Dez Bryant all possess similar upside, it’s interesting to see Brandon Marshall towering above them with a ceiling of 21.5 points per game. Marshall’s 118 receptions in 2012 don’t hurt, but he’d still be the clear No. 2 if his peak receptions per game—currently at 6.69—were closer to the average of the group (which is barely less at 6.57).
Marshall’s ceiling as a highly-targeted No. 1 option with Jay Cutler at quarterback is probably higher than that for Thomas and Jones, at least, simply because they have to share looks with other talented receivers. Jones in particular probably doesn’t have the sort of upside everyone who is drafting him in the middle of the second round is expecting. Unless the Falcons completely shift their game plan to emphasize Jones over Roddy White and defenses tailor their schemes to allow Jones to see more single-coverage, he might not have top three potential.
Since all of these receivers are currently getting selected in the first two rounds, it might be more valuable to examine their floors. The easiest way to acquire value in the first few rounds is to minimize risk; everyone has awesome upside, so it’s just as easy to hit a home run by simply trying to make contact as it is by swinging for the fences.
Quite surprisingly, Marshall has a higher floor than each of the other top-rated receivers. Again, Johnson’s numbers are deflated due to a lack of truly similar comps. You could say he’ll see increased defensive attention coming off of one of the premiere seasons in NFL history, but how much more coverage could he really see?
To get a decent sense of the risk/reward surrounding each player, I added the ceiling and floor production—the average points per game for the top four comparables plus the average for the bottom four comparables in each statistical category.
1. Brandon Marshall: 34.1
2. Calvin Johnson: 33.1
3. A.J. Green: 31.8
4. Dez Bryant: 31.7
5. Demaryius Thomas: 30.6
6. Julio Jones: 29.8
There seems to be a bigger difference between Marshall and Jones than their respective ADPs (2.02 for Marshall and 2.05 for Jones) suggest. Although Jones is entering the sought-after third year of his career, wide receivers at his age have still historically produced slightly worse numbers than those at Marshall’s. Plus, their ages are already factored into their comps. I wouldn’t specifically target Marshall in a dynasty league, but for 2013 alone, there’s probably only one receiver in fantasy football who should be rated higher.