You’ve probably been sitting in a fantasy football draft before and thought, “Do I need to reach for a second running back here? Is taking a running back with a middling ranking like Carlos Hyde really worth passing on a more highly ranked wide receiver like Keenan Allen?”
The fantasy football world has gotten very smart over the years, and nowadays, even the casual fan knows about positional scarcity and why running backs are more valuable than quarterbacks despite scoring less points. But sometimes our dealings with positional scarcity are more subjective than data-driven. For example, I haven’t seen many studies that show what value you might expect from a RB10 versus a WR10 and what that might mean for your second-round pick in a snake draft.
So, I conducted my own research: a meta-analytical study examining historical value for every fantasy player since 2002. The results of that study produced a metric that I believe is the key to unlocking your full draft potential: Leverage, which you can read more about in my article Introducing the Fantasy Football Leverage Statistic.
In this article, I’ll walk through exactly how you can optimize your draft no matter your draft position.
What Is Draft Leverage Differential?
In examining 16 years of fantasy data, the need arose for that data to be normalized based on things like player population size, distributions, and relative value. That process of normalization led to the development of the Leverage metric. Leverage is a statistical concept designed to improve understanding of relative worth across both time and position in fantasy football. By adjusting for changes in the NFL game over time, Leverage more accurately represents an individual player or position’s relative value in past seasons and in the coming season.
The single most important quality of Leverage is its precision in predicting future fantasy production based on historical data. Because of this, I saw an immediate application to draft strategy, and I developed a metric utilizing Leverage to improve player evaluation and selection in snake drafts called Draft Leverage Differential.
Draft Leverage Differential measures the magnitude of difference between the projected value of player you could draft with your current pick and the projected value of a player you could draft with your next pick.
How It Works
Let’s say you have the No. 1 overall pick in a 10-person snake draft. With your No. 1 overall pick, you obviously want to take the best player available, right? But what does “best” mean? Does it mean the player who will score the most points that season? Of course not — if that was the case, you would always draft a quarterback first. The “best” player really means the most valuable player. What you need to do with your No. 1 overall pick is to select the player who will provide the most relative value at his position.
When you make your No. 1 overall selection in a 10-man league, you will have to wait 18 more picks before you get to select another player. So, when you make your first-round pick, you also need to consider what options you will likely have available for your second-round pick. Thankfully, we have plenty of publicly available ADP reports that tell us exactly that.
This is where Draft Leverage Differential comes in. It utilizes historical data for positional leverage and computes the magnitude of difference between players expected at different draft positions, based on ADP reports for the current year. Keep in mind that Draft Leverage Differential does not compare the fundamental value of two specific players; rather, it measures the historical average difference between two kinds of players. For example, Draft Leverage Differential will tell you that with the fifth overall pick you should select RB3 if available, but it will not tell you who the RB3 should be. However, you can create your own ranked list however you’d like (such as by using our fantasy rankings from experts Sean Koerner, Chris Raybon, and Matthew Freedman) and then use Draft Leverage Differential to identify which specific position to focus on.
So if, for example, eight straight running backs go off the board to start the draft, don’t worry: Draft Leverage Differential can auto-adjust and still tell you how to approach your pick and the rest of the draft.
One more note before we dive in to an example: You might be thinking, “How is this different than something like Value Based Drafting (VBD)?” I’m glad you asked. While the fundamental ideas are the same — essentially, maximizing the value of your draft — the way in which VBD and Draft Leverage Differential are calculated are very different. Whereas VBD uses season-long projections, which inherently rely on accurately predicting the future (which can be fragile since they’re highly susceptible to small errors), Draft Leverage Differential uses a huge sample size of historical data to determine what you can expect the RB1, WR6, TE8, etc., to produce.
In a nutshell: VBD relies exclusively on accurate projections while Draft Leverage Differential takes other variables into account to improve its accuracy as a system, and thus, its value to you as a fantasy drafter.
Putting It Into Action
Let’s break this down. First, here’s a recent ADP report for 2018 from fantasyfootballcalculator.com. Our first- and second-round draft positions assuming a 10-man league where we’re drafting from the No. 1 slot are highlighted in red:
1. Ezekiel Elliott (RB1)
2. Le’Veon Bell (RB2)
3. Todd Gurley (RB3)
4. Antonio Brown (WR1)
5. Alvin Kamara (RB4)
6. David Johnson (RB5)
7. Leonard Fournette (RB6)
8. DeAndre Hopkins (WR2)
9. Saquon Barkley (RB7)
10. Kareem Hunt (RB8)
11. Melvin Gordon (RB9)
12. Odell Beckham Jr. (WR3)
13. Dalvin Cook (RB10)
14. Julio Jones (WR4)
15. Devonta Freeman (RB11)
16. LeSean McCoy (RB12)
17. Michael Thomas (WR5)
18. Davante Adams (WR6)
19. Jordan Howard (RB13)
20. Keenan Allen (WR7)
21. Christian McCaffrey (RB14)
22. A.J. Green (WR8)
23. Mike Evans (WR9)
24. Joe Mixon (RB15)
Most drafters will end up taking a running back first overall, and to evaluate if we should do the same, we have to compare the historical Leverage Score of RB1 vs. the historical Leverage Score of whatever running back would likely be available at pick No. 20.
Based on the ADP report above, we could realistically expect Christian McCaffrey (RB14) to be available, so we’d calculate his Draft Leverage Differential score:
RB1 historical Leverage Score = 290.8
RB14 historical Leverage Score = 158.3
Draft Leverage Differential = 142.5
Computing the difference between RB1’s historical Leverage Score and RB14’s historical Leverage Score yields our Draft Leverage Differential, which in this case is 142.5.
Since we also have the option to take WR1 rather than RB1, let’s compute the Draft Leverage Differential between WR1 and the wide receiver likely available at pick No. 20, which in this case is WR7:
WR1 Leverage Score = 209.5
WR7 Leverage Score = 163.0
Draft Leverage Differential = 46.5
Given the current results for ADP for the 2018 Fantasy Football season, no quarterbacks or tight ends are taken in the first and second rounds. Therefore, the Draft Leverage Differential for taking a QB1 or TE1 with the No. 1 overall pick is zero — you gain nothing by selecting a QB1 or TE1 first overall that you couldn’t get with your second round pick. Having ruled out QB1 or TE1 with our first overall pick, we’re down to just comparing RB1 to WR1. In that comparison, RB1 wins by a landslide, with a Draft Leverage Differential 96 points greater than selecting WR1 first. Therefore, the wise selection would be to take a running back first overall.
If you’ll indulge me, I’d like to present this data in a slightly different way that is perhaps a little more intuitive:
Let’s say you take RB1 in the first round. In the second round, you select WR7. Let’s calculate the combined expected leverage value returned by those two players:
RB1 expected leverage = 290.8
WR7 expected leverage = 163.0
Total expected value from both players = 453.8
Now let’s say you didn’t listen to me, and you took WR1 and RB14 instead:
WR1 expected leverage = 209.5
RB14 expected leverage = 158.3
Total expected value from both players = 367.8
Because the Leverage metric is essentially fantasy points scored adjusted for positional scarcity in a given season, it essentially serves as a perfect proxy for fantasy points scored. Therefore, by following Draft Leverage Differential’s advice and taking the running back in the first round, you’ve effectively increased your fantasy team’s scoring expectation by 86 fantasy points.
This kind of analysis is already swimming in your head when you draft in the form of big boards, projections, ADP data, positional scarcity, and those few articles you read on specific players. Draft Leverage Differential reduces all this information down into an easy-to-use metric you can utilize to dominate your draft.
What to Expect In Part 2
In Part 2, I’ll dive even deeper, showing you how the effects of Draft Leverage Differential become magnified through the length of a full fantasy draft and demonstrating its effectiveness even if you’re stuck with an unfavorable draft slot. I’ll also give you the full data on each position and show you exactly how to begin using it immediately.
Pictured above: Christian McCaffrey
Photo credit: Bob Donnan – USA TODAY Sports