Two drafts out · published July 2026
The 2028 way-too-early board.
Two Augusts from now, most of this list will be wrong. We publish it anyway, because the way-too-early board is where every draft argument starts — and nobody else shows you the uncertainty. No analyst has ranked this class. No one has run a 40. Most of these players have not taken a college snap. What honestly exists is the high school recruiting composite, the strength of the program each player signed with, and — for a handful — a freshman season. So that is the whole board: those signals, in the open, with the range each rank swings across defensible weightings printed next to it.
This is not the Draftanomics model. Our model needs consensus boards, measurements, and college production to say anything worth trusting — for 2028 it has none of the three, and we checked: its output on this class is noise. Ranking by noise and calling it a model would be fake precision. The composite is the honest signal, so the composite gets the byline.
What drives every rank
- Recruiting composite — the 247-style 0-to-1 national scale; it sets the order.
- Program talent — the school's roster-talent index. It breaks composite ties and drives the rank range.
- Early production — freshman-season per-play impact, when a played season is on file.
What's missing — and why the ranges are wide
- Analyst consensus: zero boards cover this class yet.
- Measurements: nobody has run a 40 for scouts yet — the composite is carrying that weight.
- Film that matters: the class is the high school class of 2025 — true sophomores this fall. Their draft-relevant seasons have not been played.
- Declarations: anyone in the 2027 pool belongs to the 2027 conversation and is excluded here; if they stay in school, they join this board next spring.
The board didn't load.
The board couldn't load right now — refresh in a minute.
How the order is set: recruiting composite, descending. Where composites tie — and below the five-star line they tie constantly — the program-talent index, then freshman production, breaks the tie. The rank range is not a confidence interval; it is the span of ranks a player takes when program strength is allowed 10-40% of the weight instead of only breaking ties. A tight range means every reasonable weighting agrees; a wide one means the argument is real.
Who is in the pool: the 1,800+players who are first draft-eligible in 2028 — the high school class of 2025. Anyone already in the 2027 candidate pool is excluded, because a 2028 board headlined by this cycle's stars would be a 2027 board wearing a costume.
Once the fall season is in the books, freshman production joins the board automatically; when the first 2028 analyst boards appear (late next summer), this page retires in favor of a real 2028 class hub — the way /year/2027 covers the live cycle today. Until then: arguments, honestly labeled. See methodology for what the model does once real signals exist.