This article is part of our According to the Data series.
We recently released our Top 200 Overall Composite Rankings. I think there's a ton of value in aggregate rankings because they factor out individual biases. Generally, such an approach — "wisdom of the crowds"— will beat individual rankings. Each of the guys put in lots of work to generate their individual rankings, yet chances are the average of their rankings will outperform most of them on an individual basis. That's not a knock on the RotoWire experts at all - they're some of the best in the business - but it's just a likelihood given that aggregates typically beat out individuals.
Let's look at the rankings of Drew Brees, for example. Two of the experts - Mark Stopa and Kevin Payne - have Brees ranked 21st overall, while Luke Hoover has him all the way up at No. 13 and Chris Liss has him down at No. 31. Luke and Chris could certainly be correct in their rankings - perhaps they know something the others don't - but it's more likely Brees will fall between the spots where Luke and Chris have him ranked. Brees actually checks in at No. 21 overall - right where Mark and Kevin ranked him. Since the rankings were formulated independently of one another, Brees is most likely to check in close to that mean.
To see just how the RotoWire expert rankings compare to public opinion, I charted the composite rankings against current ADP. I noted the
We recently released our Top 200 Overall Composite Rankings. I think there's a ton of value in aggregate rankings because they factor out individual biases. Generally, such an approach — "wisdom of the crowds"— will beat individual rankings. Each of the guys put in lots of work to generate their individual rankings, yet chances are the average of their rankings will outperform most of them on an individual basis. That's not a knock on the RotoWire experts at all - they're some of the best in the business - but it's just a likelihood given that aggregates typically beat out individuals.
Let's look at the rankings of Drew Brees, for example. Two of the experts - Mark Stopa and Kevin Payne - have Brees ranked 21st overall, while Luke Hoover has him all the way up at No. 13 and Chris Liss has him down at No. 31. Luke and Chris could certainly be correct in their rankings - perhaps they know something the others don't - but it's more likely Brees will fall between the spots where Luke and Chris have him ranked. Brees actually checks in at No. 21 overall - right where Mark and Kevin ranked him. Since the rankings were formulated independently of one another, Brees is most likely to check in close to that mean.
To see just how the RotoWire expert rankings compare to public opinion, I charted the composite rankings against current ADP. I noted the difference in the rankings in terms of overall draft spots and the percentage change.
Rank | Overall | ADP | Difference | % Change |
4 | C.J. Spiller | 7 | 3 | 42.9 |
26 | Lamar Miller | 43 | 17 | 39.5 |
15 | Jimmy Graham | 23 | 8 | 34.8 |
17 | Demaryius Thomas | 24 | 7 | 29.2 |
32 | Victor Cruz | 39 | 7 | 17.9 |
5 | Jamaal Charles | 6 | 1 | 16.7 |
30 | Cam Newton | 36 | 6 | 16.7 |
27 | David Wilson | 32 | 5 | 15.6 |
34 | Montee Ball | 40 | 6 | 15 |
18 | Chris Johnson | 21 | 3 | 14.3 |
36 | Reggie Bush | 42 | 6 | 14.3 |
22 | Larry Fitzgerald | 25 | 3 | 12 |
39 | Jordy Nelson | 44 | 5 | 11.4 |
10 | LeSean McCoy | 11 | 1 | 9.1 |
13 | Dez Bryant | 14 | 1 | 7.1 |
14 | Brandon Marshall | 15 | 1 | 6.7 |
28 | Roddy White | 30 | 2 | 6.7 |
1 | Adrian Peterson | 1 | 0 | 0 |
2 | Arian Foster | 2 | 0 | 0 |
3 | Doug Martin | 3 | 0 | 0 |
8 | Ray Rice | 8 | 0 | 0 |
9 | Trent Richardson | 9 | 0 | 0 |
12 | A.J. Green | 12 | 0 | 0 |
19 | Julio Jones | 19 | 0 | 0 |
35 | Vincent Jackson | 35 | 0 | 0 |
29 | Andre Johnson | 28 | -1 | -3.5 |
40 | Darren McFadden | 37 | -3 | -8.1 |
37 | DeMarco Murray | 34 | -3 | -8.8 |
24 | Maurice Jones-Drew | 22 | -2 | -9.1 |
11 | Alfred Morris | 10 | -1 | -10 |
20 | Steven Jackson | 17 | -3 | -17.6 |
31 | Peyton Manning | 26 | -5 | -19.2 |
33 | Frank Gore | 27 | -6 | -22.2 |
16 | Aaron Rodgers | 13 | -3 | -23.1 |
25 | Stevan Ridley | 20 | -5 | -25 |
23 | Matt Forte | 18 | -5 | -27.8 |
38 | Rob Gronkowski | 29 | -9 | -31 |
21 | Drew Brees | 16 | -5 | -31.3 |
7 | Calvin Johnson | 5 | -2 | -40 |
6 | Marshawn Lynch | 4 | -2 | -50 |
The players at the top of the list are those who are ranked higher in the composite rankings than where they're getting drafted by the public. There's a good chance you can acquire value on those names because 1) the experts are high on them and 2) the public hasn't caught on, so you can get them at a good price.
You can see that running back C.J. Spiller is the biggest value in terms of percentages, ranked fourth overall by the RotoWire experts but getting drafted three spots later, on average. In terms of overall movement, no one touches Lamar Miller. The guys have him ranked 17 spots higher than his ADP. I'm also quite high on Miller, so there's a really high probability that you can acquire value on him this fall.
At the other end of the spectrum, we see Marshawn Lynch and Calvin Johnson as the worst values in terms of the percentage change, although that's primarily due to their high ADP. The biggest overall mover in terms of draft spots is Rob Gronkowski, whom the guys have rated nine spots lower than his ADP.
Again, I'm on the side of the staff. Gronkowski is currently getting drafted as if there's absolutely nothing wrong with his health; even though he's scarce, he never should have been in the first or even second-round conversation simply because the opportunity cost of drafting a tight end in that range is so massive. He's just now getting drafted where he should have been if healthy.
None of this is to say that you should blindly accept aggregate rankings. You can and should alter rankings as you see fit. That's part of the fun. But if you have a player ranked drastically higher or lower than everyone else, you need to take a step back and examine the situation. You could very well be right, but we don't want to approach our projections and rankings as if our decision-making is flawless. By comparing our rankings to composite rankings from other experts, we're basically admitting that we're prone to error.
The first step - and probably the most important - in creating your fantasy football rankings this year is understanding your own biases and potential errors. By admitting you're susceptible to error, and that there's a whole lot you don't know, you can paradoxically enhance the accuracy of your predictions.
Jonathan Bales is the author of the Fantasy Football for Smart People book series. He also runs the "Running the Numbers" blog at DallasCowboys.com and writes for the New York Times.