How the Numbers Work

The 0–100 rating behind the Power Rankings, answering one question: who has the strongest team right now? It is not the standings. A lucky 5–1 team can sit below an unlucky 3–3 team.

From Week 3 on, the score is about 70% results and 30% structure. Results are what actually happened: the record, the all-play record (every score against all eleven other teams, every week), and points scored. Structure is what the roster looks like going forward: trades, waiver pickups, injuries, role changes, and the schedule ahead. Weeks 1 and 2 leaned the other way, mostly on the preseason analysis, because two games aren't enough evidence. The preseason board is the post-draft analysis of roster strength, ceiling, and construction, before any games existed to score.

The rankings are authored, not just computed. When the columnist disagrees with a mechanical ordering, the column says so and says why.

A different question: who actually wins this thing? Twelve probabilities that always sum to exactly 100%. Because eight of twelve teams make the playoffs and every playoff round is a one-week, single-elimination matchup, the market stays compressed — no preseason roster earns some absurd 30% share. The American-style odds are a conversion for entertainment. This is not a sportsbook.

A frozen artifact, graded with information available on August 29, 2026, and never revised. The components: value vs ADP (30%), starting lineup (25%), championship ceiling (20%), roster construction (15%), and bench optionality (10%).

Two recorded ADP sources exist, and they are used for different things. The per-pick prices shown across the site come from ESPN Live Draft Results (retrieved August 29, 2026) — the platform this league drafts on, which makes it the market the room actually faced. The frozen team grades were computed before that correction, against Fantasy Football Calculator's 12-team half-PPR data (3,302 drafts, August 24–29). Per the immutability rule, the grades stay as computed; the Corrections Dept. on the rankings page tells the full story. Players outside a source are shown without ADP — an unknown stays unknown.

The grades and the power board are allowed to disagree, and they do. A C+ draft can be the fourth-best team. An A- draft can rank seventh. The columns explain the gaps.

Nothing is invented: no fabricated ADP, statistics, scores, injuries, or transactions. Historical snapshots are immutable — once a week publishes, it stays exactly as it was believed at the time, wrong takes included. That is the point.