Which 2025 Fantasy Scores Deserve Your Trust in 2026

In this piece

Puka NacuaWR · LAADP 1.03Christian McCaffreyRB · SFADP 1.06Jeremiyah LoveRB · AZADP 3.04Matthew StaffordQB · LAADP 7.11Brock PurdyQB · SFADP 7.12Michael WilsonWR · AZADP 7.12Jordyn TysonWR · NOADP 8.08Makai LemonWR · PHIADP 10.01

Last season’s fantasy average is evidence, not a promise. The useful adjustment takes a player’s 2025 per-game scoring, pulls it toward his positional average based on how many games he played, then puts a range around the result.

This is not a 2026 forecast or projection model. It is simply last year’s evidence weighted honestly: small samples get less trust, while full seasons retain more of their signal.

A big average can hide a thin sample

Brock Purdy averaged 19.71 PPR points per game last season, but that came across only 9 games. His 2025 scoring, regressed for that nine-game sample, has a midpoint of 17.81 and a broad 9.84-25.78 range.

That is where the evidence shrinks hard. The headline average looks like strong QB production, but the limited sample leaves much more room around what 2025 actually established. His average draft position, or ADP — the typical pick where a player is selected — is 7.10.

Compare that with Matthew Stafford, who scored 20.61 per game across 17 games. Regressing that full-season sample produces a 13.19-25.20 range around a 19.20 midpoint. Stafford’s average barely moves because we saw nearly twice as many games. The Usage Report can help you examine the underlying opportunities before deciding how much of either quarterback’s scoring to carry forward.

Full seasons preserve more of the elite signal

Christian McCaffrey supplied 24.51 points per game over 17 games last season. Even after regression toward the running back average, that evidence lands in a 13.53-27.68 range around a 20.60 midpoint. His 2026 ADP is 1.05.

Puka Nacua has a similar profile: 23.44 points per game across 16 games, followed by a regressed 10.65-28.07 range around 19.36. He is being drafted at 1.03.

Regression does not erase excellent full-season production. It discounts the chance that every extreme repeats exactly, while still giving a long sample substantial weight. Use the Draft Kit to compare that evidence with current prices rather than treating last season’s leaderboard as this season’s ranking.

Price tells you what the room already believes

Here, value means comparing a player’s 2026 draft price against his 2025 production. Michael Wilson averaged 13.0 points over 17 games last season and now carries an 8.04 ADP, which is 25 picks later than that production justifies. Wan’Dale Robinson averaged 13.6 across 16 games and goes at 6.11, 20 picks later than his 2025 production suggests.

Those comparisons do not account for every reason a market may move, but they identify where last season’s evidence costs less. The opposite applies to rookies: Jeremiyah Love is priced at 3.01, but he has no NFL sample, so that price is entirely projection. Jordyn Tyson at 8.01 and Makai Lemon at 9.09 require the same framing.

How to use this: Start with the regressed range, not the raw average. Give full-season performers more evidentiary weight, demand a better price for small samples, and treat rookies separately because no NFL scoring exists to regress. If touchdown expectations drive a close call, consult the Anytime TD Board rather than assuming last year’s scoring pattern simply repeats.