ADP Is a Promise at Receiver and a Suggestion at Tight End

Every ADP is an average, and averages hide their spread. Two players can share a 40th-overall ADP and be nothing alike at your pick: one of them goes 38th in nearly every room, and the other goes anywhere from the late 20s to the mid 50s depending on who else is in the draft.

So I measured the spread instead of the average. Eight drafting seasons, 2018 through 2025, 1,493 player-seasons, half-PPR. For every player I have how far his actual draft slot strays from his own average, and I grouped it by where he goes off the board, because spread grows with depth at every position and comparing a third-rounder to a 15th-rounder tells you nothing. Every number below is the median player inside that range of picks.

Receivers are the position ADP describes best

From pick 25 on, wide receiver is the tightest position on the board. Not usually, not on average. Every one of the five ranges I measured from pick 25 to pick 180.

In picks 25 through 48, a receiver's draft slot moves about 3.9 picks. In 49 through 72, 5.6. In 73 through 96, 7.2. In 97 through 120, 8.5. Even out in 121 through 180, where everything is loose, receivers run 12.2 against a field that is wider at every other position.

The first 24 picks are the exception. There the typical slot moves about two picks, and running back at 2.0 is actually tighter than receiver at 2.2. It is also the one range where none of this helps you, because rooms barely disagree about anything in the first two rounds.

Everywhere else, when you are looking at a receiver and wondering whether the number under his name means anything, it does. That is the single most useful thing on this list.

Quarterback is the wide one early

The gap that surprised me sits in picks 25 through 48. A quarterback drafted in that range moves about 6.0 picks against a receiver's 3.9, roughly one and a half times as much, at the same cost, in the same rooms.

It settles down after that. By picks 49 through 72 quarterbacks are at 6.7 against 5.6, and by 73 through 96 it is 8.8 against 7.2. The disagreement is concentrated exactly where the first quarterbacks come off the board, which is also the range where most drafters think they have a read on the room.

I am not going to tell you why, because this data cannot tell me. It measures where players landed, not the order picks happened in, so any story about a run igniting is a story, not a finding. What it does say is that in that range the number under a quarterback's name carries less information than the number under anybody else's.

Tight end blows out, and it blows out late

Through pick 96, the tight end market is not especially wild. It runs about 1.1 to 1.3 times a receiver's spread, which is close enough to background noise that I would not draft differently for it.

Then it detonates. From pick 97 on, a tight end's draft slot moves about 1.5 to 1.6 times as much as a receiver's at the same ADP: 13.4 picks against 8.5 in the 97 to 120 range, and 18.7 against 12.2 from 121 to 180.

An 18.7-pick swing in a 12-team league is a round and a half in either direction. The tight end you have penciled in for the 11th round is a live candidate to go in the 9th, and just as live a candidate to sit there until the 13th. Both of those are normal rooms. Neither of them is the room the ADP was built from.

A receiver's ADP is close to a promise. A tight end's is closer to a suggestion.

What I do with it at the table

I trust receiver ADP and I plan around it. "He'll make it back to me" is the read this data supports best, from the third round to the last one.

I stop treating a late tight end as a schedulable pick. If there is a specific one I want after pick 97, waiting one more round is a much bigger gamble than his ADP implies. The flip side is the part people forget: that same width means reaching for him at market price is often paying for nothing, because the room that lets him fall a round and a half exists too. Either commit early or genuinely be fine with the next guy. What does not work is planning to take him exactly where the number says.

I hold quarterback plans loosely in the third and fourth rounds. Not because a run is coming, which I cannot measure, but because that is the one range where the market's own number is least like what any given room does.

The fine print

These are medians of what actually happened. There are no error bars here and I am not going to invent any: the underlying data has no usable count of how many drafts each player appeared in, so nothing is weighted and nothing gets a probability attached. I can tell you a tight end's slot moves 1.5 times as much as a receiver's. I cannot tell you his odds of lasting to your pick, and anybody who quotes you that number off ADP data is guessing.

I ran the spread two ways, once from the extremes of each player's draft range and once across the whole distribution. Both rank the positions identically. The numbers above are the second, which is the one a single weird draft cannot move.

Two limits worth stating. These are end-of-preseason snapshots, so this describes how drafts finished, not how boards looked in early August. And it is half-PPR, 2018 through 2025. Superflex is a different market with a different quarterback problem, and I did not measure it here.

One bias runs the other way, in my favor. A player who shows up in fewer drafts has fewer chances to post an extreme pick, so he measures tighter than he really is. If deep tight ends appear in fewer rooms than deep receivers, and they almost certainly do, then the tight end gap is wider than what I measured, not narrower.