Why a ledger and not a footnote
Everything in this encyclopedia decays at a different rate. League rules do not move. Projections move slowly. ADP moves weekly and then jumps. Depth charts resolve in a single day at the end of August. A note that mixes all four without saying which is which reads as uniformly authoritative when it is not.
So the rule is: every input gets a row, with a retrieval date and a next-check trigger. If a row cannot be filled in, that is itself the finding — see the not obtained rows below, which are as important as the ones that were.
The ledger
| Source | Last checked | Is it 2026 data | Next check |
|---|---|---|---|
| FFToday projected stat lines | Complete 419-row native-stat snapshot dated 2026-07-30, captured 2026-08-02 | Current projection input; raw stat lines re-scored locally under this league | Draft −48h |
| Fantasy Football Calculator ADP | 10-team PPR feed checked 2026-08-02: 4,121 mocks, 2026-07-26 → 2026-08-02 | Released labelled market proxy, never ESPN ADP | After Aug. 30 cutdown and draft −48h |
| ESPN league settings — scoring, roster, waivers, payout, pick sequence | Confirmed by user 2026-08-02 | 10 teams; full PPR/fractional; 9 starters + 7 bench + 1 IR; no keepers; inverse-standings weekly waivers; 1.01 snake; weeks 1–14 regular season and six-team weeks 15–17 playoffs. Draft date remains unknown. | Confirm exact draft date at draft −48h |
| 2026 bye weeks | 2026-07-25, taken from the ADP and projection pulls, which agree | Yes — byes fall in weeks 5–11, 13 and 14, with week 11 the crunch | Once more at draft − 48h as a consistency check. The schedule itself is fixed; only a transcription error is possible |
Data/players.csv — the derived board input |
Promoted 2026-08-02: 237 canonical rows with 10-team FFC ADP and current projections | Current frozen board input; unmatched ADP keys are explicitly blank; never patch by hand | Rebuild after final cuts and draft −48h |
| Kicker point totals | Modelled 2026-07-25 | Partly. Makes and attempts are 2026 projections; the field-goal distance mix is a league-wide assumption, not a per-player observation | Re-model only if the raw projections change. Treat as ordering, not forecast |
| Defence/special-teams point totals | Retained in the 2026-08-02 promotion; 20 raw FFToday D/ST rows captured for audit | Public FFToday D/ST value brackets are incomplete, so source rows do not replace the validated model | Same as above. This is the one number in the build with no direct scoring replacement |
| ESPN's own ADP | Not publicly extractable as of the Aug. 2 research pass | n/a | ESPN mocks or logged-in export; keep FFC labelled proxy |
| Depth charts and camp roles | Division research receipts created 2026-08-02 | Official/team facts plus named reported role evidence | Padded work, preseason, and cutdown |
| Official injury reports | Not applicable yet. Teams do not file them until game week | n/a | Week 1 of the regular season, then every Wednesday to Friday. See Reading the NFL injury report |
| The exact draft date | Unknown; late August | n/a | Ask Andrew before final refresh |
| Mock drafts run from the 1.01 | July record is historical; current count unverified | n/a | Run from the released 10-team board |
The re-check calendar
Checks are hung on named events, not intervals, because football information does not arrive at a steady rate. It arrives in lumps.
| Event | Approximate date | Why it matters |
|---|---|---|
| Padded practices / first preseason games | Early August | First useful role, route, third-down, and goal-line evidence; reconcile receipts |
| Second preseason game | Mid-August | The one where starters actually play a series or two. Watch who lines up with the first-team offense, not what they do |
| 53-man cutdown | 2026-08-30 | The largest depth-chart resolution of the year. Every team cuts to its final roster in a day, repricing dozens of backup running backs and third receivers at once |
| Final preseason week | Early September | Depth-chart position and injury news only. Ignore the box scores entirely |
| Draft − 48h | TBC | Confirm exact date; pull fresh stat lines and 10-team ADP; rebuild data, board, tiers, and simulation |
The August 30 fork changes the plan's shape. Drafting before the cutdown means a board built on depth charts that will churn immediately afterward — which promotes the in-season waiver layer to load-bearing and argues for late-round upside over insurance picks. Drafting after it means a stable board where the last rounds are worth spending on known backups.
One standing rule against the classic August trap: do not react to preseason statistics. Starters play one or two series. The leaders in preseason rushing yards are, disproportionately, players who will not survive the cutdown. What August produces is depth-chart position and injury news. It does not produce box scores worth reading.
What this ledger commits to
If a row above is stale at draft time and has not been re-pulled, the notes that depend on it are stale too, and the honest move is to say so rather than to draft from them. That is the same rule as everywhere else in this build: a missing input propagates as a missing output, never as a silent default — Evidence and confidence.
This note expires on 2026-08-09 because news changes the board quickly, not because the released 10-team canonical rebuild is pending. Re-run the freeze after final cuts and at draft minus 48 hours before treating it as live advice.
Open questions
- The exact draft date remains pending; do not delay the data rebuild for it, but use it to schedule the final −48h refresh.
- Ten $75 contributions total $750 while confirmed prizes total $900 ($500 / $250 / $150). Preserve this confirmed mismatch as stated; do not infer a financing correction.
- The defence/special-teams model has no direct source scoring replacement. It remains internally validated, while FFToday's 20 raw D/ST rows are retained as an audit input.
- The 2026-08-02 canonical data and dependent prose are rebuilt; the July 12-team board remains historical only.