← Research · Above Replacement

Do Preseason Rankings Actually Predict Anything?

Above Replacement Research · August 2026

Every August the industry publishes rankings; every January nobody checks them. We did: preseason expert-consensus draft boards for eight seasons (2015, 2018, 2020–2025 — several recovered from the Internet Archive after their URLs were overwritten), scored against what players actually did, position by position.

The scoreboard

PositionRank↔result correlation (pooled)Read
WR0.58Reliable — trust the board
RB0.54Reliable — trust the board
TE0.42Decent, noisier
QB0.37Volatile but real (positive all 8 seasons, p = 0.008)
DST−0.16 ± 0.12No signal (p = 0.73)

What this means on draft day

Spend your early picks where the rankings work. WR and RB are the positions where preseason opinion most reliably converts to points — which, combined with the flex spot's extra demand, is the quantitative case for building drafts around them.

QB rankings are real but jumpy. The signal never inverted in eight seasons, but its strength swings year to year. Paying up is only defensible at the very top; the QB middle class returned less than late-round starters in our tests.

Drafting a defense early is paying for a coin flip. Preseason DST rankings showed no predictive power at all — the correlation's sign flips season to season and pools to indistinguishable-from-zero. Every draft slot spent on an "elite" defense before the final rounds buys nothing measurable. (In-season, defenses are predictable — one week ahead, from Vegas lines. See our streaming study.)

A note on honesty

An earlier internal version of this analysis described DST as having an anti-signal (rankings inversely predictive). Proper significance testing killed that claim — the truth is "no signal," which is less fun and more useful. We publish the correction because the whole point of scoring rankings is that someone checks.

Methodology: preseason boards joined to end-of-season fantasy points under a fixed scoring rule; Spearman correlations within position, pooled across seasons; sign tests across seasons for robustness. The findings drive the position strategy inside the Above Replacement draft tool.