Thursday, September 10, 2026
How Much of 2025 Fantasy Scoring Should You Trust?
Sample size changes how seriously drafters should treat last season’s per-game scoring, especially when current prices assume a repeat.
Bobby's AI Desk·written from our own draft-market and 2025 production data·how this works
Takeaways
- → Discount strong per-game scoring when it comes from a short sample.
- → Compare 2026 ADP directly with 2025 production before calling a player valuable.
- → Treat rookie prices as projections, not extensions of NFL evidence.
In this piece
Last season’s per-game scoring needs a sample-size check before it informs a 2026 draft. The method here takes each player’s 2025 scoring, pulls it toward his positional average according to games played, then places a range around that adjusted figure. This is not a 2026 forecast; it is last year’s evidence, weighted honestly.
The small-sample number takes a haircut
Brock Purdy averaged 19.71 points across 9 games last season. Once that limited sample is regressed toward the quarterback average, the central reading falls to 17.81, with a wide 9.84–25.78 range and medium confidence.
That is the clearest warning against treating a big per-game average as settled performance. Purdy’s 19.71 came from barely more than half a season, so the adjustment shrinks it hard and leaves substantial uncertainty on both sides.
A full season holds up better. Drake Maye scored 20.70 per game across 17 games; the adjusted reading remains 19.27, with a tighter 14.63–23.90 range and high confidence. Matthew Stafford also played 17 games, moving from 20.61 to 19.20 with a 13.19–25.20 range. The Usage Report can help you inspect what supported those averages, but the basic lesson starts with games observed.
Draft price is a separate question
ADP, or average draft position, tells you where a player is typically selected. Value means getting production at a cheaper draft cost, and this comparison specifically measures a player’s 2026 price against his 2025 production.
Jauan Jennings has a 13.08 ADP after averaging 11.6 points over 15 games last season. His 2026 price is 32 picks later than that 2025 production justifies. Josh Jacobs, at 8.11, averaged 15.8 across 15 games and is going 26 picks later than last season’s production suggests.
The reverse applies to Justin Jefferson. His 1.12 price is 16 picks earlier than his 11.9 points per game from 2025 justify. That does not settle his 2026 result; it shows how much the market is asking you to pay beyond last season’s output. Track those costs in the Draft Kit.
Averages can hide weekly volatility
Boom/bust describes players whose weekly scores frequently swing between large and small results. Joe Burrow averaged 16.8 points over only 8 games last season, scoring at least 20 in 50% of them but fewer than 8 in 25%.
Garrett Wilson reached 20 points in 43% of his games and finished below 8 in 29%. Those splits are reminders that one per-game figure cannot describe roster-level risk. The Anytime TD Board is useful when checking whether touchdown expectations are doing too much work in your evaluation.
Rookies belong in a different bucket
Rookies have no NFL sample, so their prices are projections rather than weighted evidence. Jeremiyah Love costs 3.04, Jadarian Price goes at 5.12, and Carnell Tate goes at 7.08. None can be evaluated with the same 2025 regression exercise, which makes assumptions more important than demonstrated NFL scoring.
How to use this: Start with the adjusted range, not the raw average. Demand a discount for short samples and wide ranges, then compare that uncertainty with current ADP. Treat rookies separately because their draft prices rest entirely on projection.
Get the ADP movers by email
One email a week through draft season: who's climbing, who's falling, and which players the market is pricing furthest from what they actually produced last year. Free — no account needed.
Weekly through Week 1. No spam, unsubscribe any time.