There is an urban legend, known as “The Sinking Library,” which has been told and retold on college campuses across this country for decades. The story goes like this: when the architect set out to design the university’s library, she or he forgot to account for the weight of the books. Because of this oversight, the library is slowly sinking into the ground, growing ever-closer to being consumed by the Earth with each passing year.
If this story was told on your campus, you may have heard it differently depending on which school you attended. In some versions, bookshelves in the library are kept empty in order to keep the facility from sinking more quickly. In others, it isn’t the library, but the athletic center that is sinking (because the architect didn’t account for the weight of the water in the swimming pool).
No matter which variation you’ve heard, none of these stories have been proven to be anything more than campus mythology, a tale kept alive as it’s passed down through generations of students. But I think this story persists not because its orators truly believe it to be fact, but because it’s a parable – one that gets to a central truth of the human condition.
And there’s a lesson here, one that explains why we inadvertently sabotage our chances of success on fantasy football draft day.
Run custom mocks with your league settings with our Draft Simulator
First, Let’s Talk about Fantasy Rankings
You’ve probably seen lists of player rankings so often that you’re completely normalized to their cadence and structure. But I want to take a second look at them, with one question in mind: how do they influence our strategy for lineup construction?
For example, look at this year’s (2020) consensus expert rankings. In particular, I want you to pay attention to the first 72 spots (rounds 1-6 in a twelve-team league’s conventional draft). Within those spots, you will find a total of 33 wide receivers, compared to 26 running backs.
The high volume of receivers and backs at the top of this list seems to insinuate that, essentially, it is equally important to load up on both positions early in your draft. Further, these rankings suggest that more wide receivers are worthy of draft consideration than running backs in the first six rounds. Now, I should note that there is a deliberate method to this list’s construction. For instance, the 36th ranked wide receiver (Will Fuller) is recommended to be drafted right around the 30th ranked running back (Derrius Guice). This, on the surface, makes sense: the WR36 and RB30 each generated about the same amount of total points in 2019.
But here’s my question: is this how it actually plays out? If we fast-forward ahead to the end of the season, will we find that these positional values held true to where they were suggested to be drafted?
In a word: no.
Take a Second Look at the 2019 Fantasy Rankings (and Results)
Consider last year’s (2019) consensus player rankings. I took this list and layered each player’s final 2019 fantasy performance over it. In other words, if you drafted a running back in the first round of 2019, you didn’t ultimately end up with the RB1, RB2, or RB3 in some perfect order based on where you drafted. No, you might have ended up with the RB1 (Christian McCaffrey), the RB17 (LeVeon Bell), or the RB38 (David Johnson).
But, here’s the problem: you didn’t know that when you drafted that player. We know that intuitively. Yet, rather counter-intuitively, that’s not how we draft. Instead, we draft based on this belief that wide receivers and running backs are equally important to select early because that’s what the fantasy rankings tell us. So, if you followed a conventional draft strategy, you probably moved on to selecting a wide receiver in the second or third round because you assumed you were “set” at your RB1 slot when you drafted Christian McCaffrey (true!) or David Johnson (decidedly not true!).
You see, fantasy rankings tend to suffer from hindsight bias. They stack up all of the outputs from last season and rank them in order as if we’ll all be perfect in our selections in the following season. As if we should choose the projected WR36 in the same round that we choose the projected RB30 because the outputs of those two positions were nearly identical last year.
But the real data turns out to be far messier. Once I layered 2019 actual player performance over the projected 2019 rankings, I found that, on average, 29 of the top 35 running backs ended up being drafted within the first 72 picks last season. Compare that to wide receivers: only 21 of the top 35 wide receivers were drafted, on average, within that range. In fact, if you plotted out these 2019 pre-draft rankings against a conventional twelve-team league’s draft, there were precisely zero running backs drafted in rounds 7-10 that returned value.
Here’s another way of looking at my findings: based on last year’s player ADPs, if you selected a wide receiver in rounds 7-12 of your 2019 fantasy draft (again, assuming a twelve-team league), there was about a 35% chance that player returned top 40 positional value. But if you drafted a running back in rounds 7-12? That selection returned top 40 positional value only 13% of the time.
Put these findings together with what we already know: that the difference between the average weekly fantasy output for the top 10 running backs in 2019 was about 4 points greater than the average for the top 10 wide receivers. And that the difference between the 2019 RB20 (Marlon Mack) and RB60 (Brian Hill) was 112 fantasy points, while the difference between the 2019 WR20 (John Brown) and WR60 (Kenny Stills) was only 83 fantasy points.
In review: (1) the best running backs are the most valuable players in fantasy; (2) it is far more difficult to source roster-worthy running backs after the sixth round than it is wide receivers; and, (3) the delta between good and average running backs is greater than the delta between good and average wide receivers.
So: Why do fantasy rankings treat the positions equally? Why do they actually rank more wide receivers in the top 72 draft positions than running backs?
The Lesson of The Sinking Library
One interpretation of the parable of “The Sinking Library” is that it’s a lesson in hubris. We have a difficult time imagining that we’ll be wrong about something, and thus, we often don’t account for a margin of error in our work. This is what ultimately doomed the fabled architect in the story: the measurements were so precise, he or she didn’t leave any room for an oversight (in this case, the weight of the books).
This is the state of fantasy football draft rankings. We line up players based solely on expected output, without taking into account the only certainty of draft day: that we will be wrong more often than we will be right.
Last season, only about 48% of running backs and wide receivers drafted in the first four rounds (or first 48 positions) of fantasy drafts ended up finishing within the top 35 players at their position to end the season. The data is shouting at us: the chances your early-round selection will pan out is tantamount to a coin flip.
Now, that would be immaterial if running backs and wide receivers were equally accessible to source later in the draft (or on the waiver wire). But they aren’t. Remember, in 2019 only six running backs that finished in the top 35 at the position were still available after the first 72 draft picks (compared to fourteen wide receivers).
You see, the fatal flaw of fantasy rankings is that they don’t account for a margin of error. They don’t assume we will be wrong – and they don’t give us room to correct for failure.
If You Baked in a Margin for Error, The Only Choice is a Zero-WR Draft Strategy
In most leagues (certainly in deeper leagues, where the waiver wire isn’t brimming with stellar week-to-week options), the data is shouting at us: Over-drafting the running back position is essential to building a championship lineup. You cannot afford to miss on running backs on draft day. This is the position that is far more valuable to get right, and far less replaceable when you’re wrong.
Remember, we human beings are pretty bad predictors of the future. This means you can’t just overconfidently draft a few backs and hope things work out. No, you need to draw a larger circle by which to search for them. My recommendation: in conventional-lineup, standard twelve-team leagues, drafting at least five RBs in your first six picks is paramount.
Don’t trick yourself into believing that you can predict the future, particularly in what will be, perhaps, the most unpredictable NFL season in decades. Over-draft the position that affords you the highest probability for success. Focus on the rest later in the draft, where good options at wide receiver, tight end, and quarterback still exist. You don’t need the best players at every position. You just need the best ones where it’s most valuable.
You will be wrong more often than you will be right on draft day. Bake in a margin for error.
Don’t forget to account for the weight of the books.
Try the only fantasy football draft software that syncs with your draft
Subscribe: Apple Podcasts | Spotify | Google Podcasts | Stitcher | SoundCloud | iHeartRadio
Beyond our fantasy football content, be sure to check out our award-winning slate of Fantasy Football Tools as you prepare for your draft this season. From our free mock Draft Simulator – which allows you to mock draft against realistic opponents – to our Draft Assistant – that optimizes your picks with expert advice – we’ve got you covered this fantasy football draft season.
David Giardino is a featured writer at FantasyPros. For more from David, check out his archive and follow him @davidgiardino.