Keep ADP on the screen
PRICE SIGNAL
ADP is useful for predicting the room and estimating which players may reach your next turn. It becomes dangerous only when it is treated as a projection or universal answer.
Home / 2026 Research / ADP vs Model
Original OverADP Research
ADP vs Model: What 1,000 Drafts Found
A paired historical simulation tested whether a roster-aware decision system could build stronger starting rosters than drafting from market ADP first.
Answer
Across 1,000 paired 12-team simulations using true 2023 and 2024 Fantasy Football Calculator ADP, Target Intel produced a 50.6% top-three regular-season finish rate versus 25.7% for ADP-first drafting when Week-1 lineups were frozen. That is evidence for the combined decision system—not a guarantee or proof that raw model ranks always beat ADP.
50.6%
Target Intel top-three regular-season rate with the Week-1 lineup frozen.
25.7%
ADP-first top-three regular-season rate under the same league conditions.
+120.9
Average frozen-lineup point difference for Target Intel versus ADP first.
The result by historical season
| Season | Strategy | Simulations | Avg. rank | Top-three rate | Frozen-lineup points |
|---|---|---|---|---|---|
| 2023 | ADP first | 500 | 6.63 | 26.4% | 1,102.5 |
| 2023 | Target Intel | 500 | 4.26 | 48.8% | 1,223.0 |
| 2024 | ADP first | 500 | 6.72 | 25.0% | 1,129.0 |
| 2024 | Target Intel | 500 | 4.04 | 52.4% | 1,250.4 |
Pooled headline: 1,000 simulations per strategy across the two true-ADP seasons. Percentages are regular-season results. Download the summary CSV.
What actually beat ADP?
Not a raw projection sort. The strongest policy used ADP as a market price and selectively overrode it when position-specific projection value, roster need, positional scarcity and the probability a player would disappear before the next turn aligned.
The practical difference
ADP tells you who the room is drafting. OverADP helps decide who your roster should draft now and who can wait. That decision layer is the part supported by the simulation.
How the paired simulation worked
- Each strategy entered 500 paired 12-team, 15-round snake drafts in 2023 and another 500 in 2024.
- Each pair used the same season, draft slot, random seed, opponent logic, schedule and actual weekly player outcomes.
- Opponents drafted from noisy ADP plus roster need and never used OverADP projections.
- The primary result froze the Week-1 lineup for Weeks 1-14 to isolate the roster created by the draft.
- Injuries, missed games and breakouts entered through real historical weekly fantasy points.
What the study does not prove
This was a retrospective exploratory backtest, not an untouched future-season experiment. Opponents approximate home-league behavior rather than replaying observed draft rooms. The simulation omits trades, FAAB auctions, kickers, defenses and explicit injury designations. Championship rates did not improve, so the evidence supports stronger drafted rosters and regular-season position—not guaranteed titles.
The 2025 sensitivity season is intentionally excluded from the headline because it uses an ESPN preseason-rank proxy rather than true historical ADP. Read the complete test design and limitations.
How to use the result in a 2026 draft
Compare position-adjusted value
NOT RAW POINTS
Use VBD and position-specific projections to compare players whose raw scoring scales are different.
Draft for this roster
LIVE CONTEXT
Starter holes, flex paths, existing volatility and positional cliffs can change the best selection after every pick.
Frequently asked questions
Does a fantasy football model beat ADP?
This simulation found that the combined Target Intel strategy beat an ADP-first control across the 2023-2024 true-ADP sample. It did not find that a raw model-only sort universally beats ADP, and it cannot guarantee future performance.
Why freeze the Week-1 lineup?
Freezing the initial lineup separates the roster created at the draft from later start/sit and waiver decisions. That makes the headline result more directly about draft quality.
Did Target Intel win more championships?
No reliable championship improvement appeared in this test. A short playoff is highly variable, and the study is not evidence for guaranteed titles or prize money.
Related Rankings
The Draft Tool
Get the Full Model in Your Draft Room
Every projection, 80% confidence interval, risk tier, and sleeper/bust call — live, in a single-page draft command center. Free to browse. $6.99 for a single draft, $24.99 for the full season.
ENTER THE WAR ROOM →