We don't draft players. We draft deltas.
A delta is a measurable distance between two values. We find the ones that matter: the gap between where consensus has a player and where he actually belongs.
Three positions, three studies, roughly 1,350 drafted players — and per-opportunity quality is flat across draft capital at all three. The draft decides who plays. It does not decide who is good.
Data scientist. Makes every call with the numbers… the studies, the models, the base rates. Looks into data you've never even thought of.
Makes every call with the vibes — the film feel, the locker-room reads, the gut. Some things the numbers can't see yet.
The Dashboards
One tab per position. Each is a live dashboard carrying every current finding from that position's study — the tables, the charts, and the base rates, all interactive. When a study is updated, the dashboard updates with it.
Getting on the field, and getting the ball. 580 drafted WRs since 2008, both gates measured.
The draft decides who plays. It does not decide who is good. 212 drafted QBs, 2008–2025.
388 drafted RBs — and every down they played — on talent, volume, and third down.
Everything Else
Where the studies turn into calls you can actually draft on.
Every week, two rankings per position: Wilson's built purely from the data, MC's built purely from the vibes — plus the delta between them.
Two players in, one decision out — plus who wins you the week, who keeps you out of trouble, and why those aren't the same man.
Which of last week's outliers meant something. Each miss split into the work a man was given and what he did with it, judged against four seasons of what followed.
The season-long redraft boards — PPR, Half PPR, and Superflex — every player, tier, bye, and the delta note behind each call.
Build any category from the studies — draft day, age, target band, moves, vacancies — and get the actual list of players in it.
Every NFL kicking venue since 2002, and what the buildings themselves do to field goals.
One question, three positions
| Finding | RB | WR | QB |
|---|---|---|---|
| Magic round exists? | No | No (p = 0.16) | No (p = 0.59–0.87) |
| Per-opportunity quality flat across draft days? | Yes (0 of 9 sig.) | Yes (0 of 15 sig.) | Yes (0 of 16 sig. at primary bar) |
| Wider outcome spread on Day 3? | Yes (p = 0.02–0.05) | not tested | No (all Holm p = 1.00) |
| Elite conversion once volume is earned | n/a | Same by round (p = 0.61) | Same by round (CI straddles zero) |
| Early-1st buys floor or ceiling? | n/a | Floor (ceiling OR 0.99, p = 0.79) | Marginal ceiling only |
| “Ever” outcomes explained by duration? | not tested | not tested | Yes — 103% shrinkage on log(pick) |
| Early Day 3 sweet spot? | Slight floor edge | Yes (15.7% vs 3.9%) | No (11.9% vs 7.7%, n.s.) |
| Hard prerequisite gate | 50 carries | 80 targets | 10 starts (0 of 104) |
| Late-round elite share, era trend | n/a | Flat (21.7% → 21.3%) | Flat (14.8% → 14.8%) |
| Middle class, era trend | n/a | Collapsed (top-24 32.5% → 17.5%) | Day 2 marginal (p = 0.058); undrafted eliminated |
| Vacancy fills from inside? | n/a | Yes (OR 2.79, 12.5% first-feed) | No — 7.6%, and 4.3% for Day 3 backups |
| Does the role travel on a move? | n/a | Alpha yes, fringe no (58.7% → 32.4%) | Full-season yes (87% → 87%), part-season no (74.6% → 46.8%) |
| Pre-draft signal that survives | College receiving | Draft age (Day 3 only) | College rushing — STRENGTHENS under control |
| College-to-NFL skill transfer | r = 0.27 (receiving) | r = 0.27 (receiving) | ρ = 0.83 (rushing) |
Cross-study callbacks: where the three studies agree, disagree, and why. ~1,350 drafted players across RB, WR, and QB.