Saved probe time
Know what the model knew.
Track source coverage, observation time and the path from capture to a published number.
Yahoo Fantasy API permission
This saved API result is historical and does not measure the current browser capture.
The app and a fresh token were refused by the Fantasy API. League, matchup and standings capture use the authenticated browser independently.
200 confirms this read endpoint; 403 can indicate an app refusal
Manual API recheck only; authenticated browser capture continues
Language pipeline · capture and processing
Four independent stages. The Q-01 total is captured text only; it does not include podcasts.
News text fetched
56,817 / 70,871 (80.2%)14,050 remaining. 0 not attempted; 14050 failed or unavailable. Some publisher pages cannot be recovered.
News and pressers processed by Q-01
13,147 / 23,587 (55.7%)10,440 remaining. 11769 accepted by a schema-v2 reader; 90 backfilled; 483 rejected as wrong-page. 9784 news queued for local reading; 656 presser chunks queued for review or capped Haiku. Rejected claims stay pending. New captures can grow this total.
Podcast audio captured · current queue: Locked On 2024+, 6,706 episodes
6,706 / 6,706 (100.0%)Finish: complete
0 remaining. Estimated 0 GB still to download (feed lengths adjusted to 3,735 verified files; actual downloads may differ). 5508 archived in Drive; 20 pending transcription and 5488 filed as transcribed.
Podcasts transcribed · current queue: Locked On 2024+, 6,706 episodes
6,686 / 6,706 (99.7%)Finish: finishes Thu Nov 19, 2026 MT at 9/day (9 done in the last 24.1 h, all machines)
20 remaining. Estimated 20.8 audio hours left; about 4.4 hours of Whisper compute from 6,686 completed episodes. Compute estimate excludes fetching and scheduled waits. Two machines work this queue: the Mac's Whisper runs 12 AM–6 AM daily and 8 AM–5 PM Mon–Fri, Boise time; Nimo runs 24/7 on its current lane (0 episodes in flight). Claims keep them off the same episode.
Podcast audio in the Drive archive · current queue: Locked On 2024+, 6,706 episodes
5,508 / 6,706 (82.1%)Finish: finishes Fri Apr 16, 2027 MT at 8/day (8 done in the last 24.1 h, all machines)
1,198 remaining. Every file is checked against Drive’s md5 before it counts. Nimo uploads its own downloads directly; the Mac files them as transcribed.
Podcasts transcribed · full history back to 2015: 81,553 episodes, about 2,572 GB, about 41,726 audio hours
5,712 / 81,553 (7.0%)Finish: stalled — 0 in 24h
75,841 remaining. All Locked On team dailies in the deep archive index; the current queue and backfill lane are both inside it.
Podcast audio in the Drive archive · full history back to 2015: 81,553 episodes, about 2,572 GB, about 41,726 audio hours
15,822 / 81,553 (19.4%)Finish: stalled — 0 in 24h
65,731 remaining. Drive-verified audio sidecars (md5 checked).
Nimo · Windows laptop, 24/7 transcription
No contact for 25,157 min
Last contact Thu Oct 22 7:57 AM MDT (25156.7 min ago; red after 15). Went live Thu Sep 24 10:05 PM MDT. 201 transcripts collected in 24 h (6.2 audio hours per hour); current: 0 in the last hour, 0 in 24 h, 0 in flight; backfill: 0 in the last hour, 0 in 24 h, 0 in flight. Failures in 24 h: 251; last error Thu Oct 22 7:57 AM MDT: SshRefused: ssh rc=255: ssh: connect to host 192.168.0.26 port 22: Connection refused. Disk: 72 GB free (50 GB floor); — audio files kept locally (0.0 GB); — verified in Drive, — waiting to upload.
Archive lane (download everything, file it in Drive): 8,205 of 8,206 catalogue episodes verified in Drive; 0 queued on Nimo. Download 0.2 MB/s → everything downloaded about Tue Nov 17. Upload 0.0 MB/s, 1 GB in the last 24 h of Drive’s ~750 GB/day limit → all in Drive about not yet measured (MT). Rates are the last 10–15 minutes.
The Desk · GTX 1070, GPU transcription (PW-193)
No contact for 4 min
Last contact Thu Oct 22 7:58 AM MDT (3.8 min ago; red after 15). Went live Tue Oct 6 7:32 PM MDT. 674 transcripts collected in 24 h (22.74 audio hours per hour); current: 0 in the last hour, 0 in 24 h, 0 in flight; backfill: 0 in the last hour, 0 in 24 h, 0 in flight. Failures in 24 h: 0; last error Sat Oct 10 11:22 AM MDT: RuntimeError: ssh rc=255: ssh: connect to host 192.168.0.16 port 22: Network is unreachable. Disk free on D: 1,363 GB.
Combined throughput (Nimo + Desk, Mac mlx not counted the same way): 28.94 audio hours per hour.
Q-01: Allowed now. 12 AM–6 AM daily; 8 AM–5 PM Mon–Fri, Boise time. Podcast policy: Download + scheduled transcription. Audio capture clock: every 55 min, loaded. Whisper schedule: 12 AM–6 AM daily, 8 AM–5 PM Mon–Fri; loaded. Mac audio spool limit: 53.7 GB; 42.9 GB free-space floor. Nimo keeps its audio until its disk nears a 50 GB floor. GB = 10^9 bytes throughout.
Q-01 checkpoint Sep 30 · 11:33 AM MT; podcast queue Nov 17 · 10:16 AM MT; audio job Nov 17 · 10:16 AM MT; build reading Nov 17 · 10:41 AM MT.
Source inventory
Age is relative to collection. Different feeds have different expected cadences; an old offseason record is not automatically a failed sensor.
| Source | Recorded value | Age | Evidence |
|---|---|---|---|
| Market quotes (odds_quote) | 448,126 priced rows | 1.1h | next kickoff in 55.6h; threshold 80h. Spreads, totals and moneylines with prices. |
| Play-by-play cache (2026) | week 10 · 26,246 plays | 2.6 h | last game 2026-11-16 |
| Warehouse · player_week 2026 | week 10 · 7,618 rows | 4 min | 1919 MB |
| Active data generation | 71063a1fc7d9 | 1 h | schema v1 |
Capture clocks
Saved capture audit: 2026-11-17T10:22:57 (source clock). Status is reproduced from that audit; negative ages require clock reconciliation.
| Source | Recorded status | Rows | Newest source time | Age at audit (h) | Allowed interval (h) |
|---|---|---|---|---|---|
| odds_quote | FRESH | 448,126 | 2026-11-17 16:32:59 | 0.8 | 25.5 |
| projections | ICED | 90,630 | 2026-10-30 01:20:31.175649 | 448.0 | 192 |
| early_lines | FRESH | 13,102 | 2026-11-17 16:32:59 | -6.2 | 192 |
| inj_snapshot | FRESH | 26,441 | 2026-11-14 18:00:01 | 64.4 | 108 |
| dc_snap | FRESH | 1,328,072 | 2026-11-17 16:52:03 | -6.5 | 40 |
| roster_state | FRESH | 3,072 | 2026-11-17 15:13:01.195437 | 2.2 | 30 |
| presser | FRESH | 1,446 | 2026-11-14 01:03:58.180014 | 88.3 | 192 |
| workload_claim | FRESH | 17,532 | 2026-11-15 15:00:57.011245 | 50.4 | 96 |
| avail_snap | FRESH | 15,501 | 2026-11-15 16:00:05 | 49.4 | 192 |
| coach_week_snap | FRESH | 15,624 | 2026-11-17 15:03:01.071002 | -4.7 | 192 |
| news | FRESH | 70,871 | 2026-11-17 17:19:26.029958 | 0.1 | 6 |
| pred_lock | FRESH | 0 | 0 unstamped past lock | — | — |
| odds_quote[pinnacle] | FRESH | 25,404 | 2026-11-17 16:32:57 | 0.8 | 25.5 |
| odds_quote[draftkings] | FRESH | 53,978 | 2026-11-17 16:32:31 | 0.8 | 25.5 |
| odds_quote[fanduel] | FRESH | 46,840 | 2026-11-17 16:32:59 | 0.8 | 25.5 |
| odds_quote[williamhill_us] | FRESH | 46,770 | 2026-11-17 16:32:59 | 0.8 | 25.5 |
| early_lines[pinnacle] | FRESH | 966 | 2026-11-17 16:32:57 | -6.2 | 192 |
| early_lines[draftkings] | FRESH | 978 | 2026-11-17 16:32:31 | -6.2 | 192 |
| early_lines[fanduel] | FRESH | 978 | 2026-11-17 16:32:59 | -6.2 | 192 |
| early_lines[williamhill_us] | FRESH | 978 | 2026-11-17 16:32:59 | -6.2 | 192 |
| inj_snapshot | FRESH | 26,441 | 2026-11-14 18:00:01 | 64.4 | 108 |
| presser | FRESH | 1,446 | 2026-11-14 01:03:58.180014 | 88.3 | 192 |
Presser exact-name review
The unresolved count stays open until source adjudication.
Review coverage, including unresolved labels
New rows awaiting review
Needs source or audio adjudication
Transcript-context verdicts
Excludes broadcast-only mentions
Read 2026-11-17T16:47:37+00:00. Labels are review coverage, not a correctness rate. Exact labels, excerpts and pending evidence.
Inspected in the saved validation
Includes withheld historical products
No reading is not a clean pass
Gaps requiring interpretation
| Kind | Location | Detail | Available from |
|---|---|---|---|
| field coverage | week 2 · DST · projection | none of the 2 DST rows carry a projection, while 28 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 2 · K · projection | none of the 2 K rows carry a projection, while 28 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 2 · DST · floor | none of the 2 DST rows carry a floor, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 2 · K · floor | none of the 2 K rows carry a floor, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 2 · DST · ceiling | none of the 2 DST rows carry a ceiling, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 2 · K · ceiling | none of the 2 K rows carry a ceiling, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 2 · DST · baseline | none of the 2 DST rows carry a baseline, while 28 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 2 · K · baseline | none of the 2 K rows carry a baseline, while 28 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 2 · DST · why | none of the 2 DST rows carry a why, while 28 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 2 · K · why | none of the 2 K rows carry a why, while 28 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 3 · DST · projection | none of the 2 DST rows carry a projection, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 3 · K · projection | none of the 2 K rows carry a projection, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 3 · DST · floor | none of the 2 DST rows carry a floor, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 3 · K · floor | none of the 2 K rows carry a floor, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 3 · DST · ceiling | none of the 2 DST rows carry a ceiling, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 3 · K · ceiling | none of the 2 K rows carry a ceiling, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 3 · DST · baseline | none of the 2 DST rows carry a baseline, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 3 · K · baseline | none of the 2 K rows carry a baseline, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 3 · DST · why | none of the 2 DST rows carry a why, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| field coverage | week 3 · K · why | none of the 2 K rows carry a why, while 27 of 32 players in the week do | the producer covers other positions and not this one |
| week shape | week 1 · projection | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 2 · projection | 88% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 3 · projection | 84% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 4 · projection | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 5 · projection | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 6 · projection | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 1 · floor | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 2 · floor | 84% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 3 · floor | 84% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 4 · floor | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 5 · floor | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 6 · floor | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 1 · ceiling | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 2 · ceiling | 84% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 3 · ceiling | 84% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 4 · ceiling | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 5 · ceiling | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 6 · ceiling | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 1 · baseline | 0% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
| week shape | week 2 · baseline | 88% of rows carry it against 100% in a typical week -- the weeks are meant to be copies of one shape | the other weeks of the same payload |
The weather ladder already stores lead days. Research now requires an identifiable forecast run and a defensible availability time. A realized historical weather observation cannot substitute for a forecast made one week earlier.
Payload validation · Nov 17 · 10:38 AM MT
Build commitments
Source: executable build plan. A completed item is not proof of betting edge.
| Item | State | Work |
|---|---|---|
| C-13 | done | C-13 — Fill `published_at` and `effective_at` on the 9,039 receipts, then on every table without a cloc |
| C-10 | done | C-10 — Coach and coordinator press conferences within 24 h. Podcast priority queue over the 461,382 ind |
| C-14 | done | C-14 — The injury-report ladder and depth charts "cannot be walked backwards" only through one endpoint |
| C-17 | done | C-17 — College play-by-play and player seasons from collegefootballdata.com (free key) for rookie prior |
| P-08 | done | P-08 — 2026 in-season participation does not publish until after the season. Impute on-field sets from |
| P-09 | done | P-09 — `game_context_asof`: roof, surface, elevation, travel_km, tz_shift, rest days, short week, offic |
| C-12 | done | C-12 — Parse the FTN payloads already sitting in `evidence.sqlite` into `ftn_charting`: play-action, sc |
| C-01 | done | C-01 — Per-source staleness measured by the *source's own timestamp*, not the job's exit code. It write |
| C-02 | done | C-02 — Fix the `-3≤hours_to_kickoff≤6` window bug. Turn the eight one-shot captures into standing ones: |
| C-05 | done | C-05 — Props for all markets, not 6, hourly; de-vig method unchanged. |
| P-06 | done | P-06 — Load all 372 pbp columns into `plays_wide`; `plays` stays as a stable view. Extend the banned li |
| C-03 | done | C-03 — Captures append to `data/land/*.parquet` and never touch DuckDB. One single-writer loader owns t |
| C-11 | done | C-11 — One refresher for everything on the nflverse release path: pbp, participation, FTN charting, NGS |
| C-09 | done | C-09 — Daily roster diff: signings, IR, practice-squad elevations (the Saturday tell), waiver claims. |
| P-10 | done | P-10 — One identity: GSIS canonical through `dp_ids` (35 id systems, 100% populated) in place of `xwalk |
| P-07 | done | P-07 — Participation 2016–2025 → `participation(game_id, play_id, gsis_id, side, pos)` and `play_person |
| C-06 | done | C-06 — Fix the Eastern-into-UTC index bug. Capture the forecast ladder (lead 7→0 days, hourly) for ever |
| C-04 | done | C-04 — **The number you actually face.** Hourly from Monday 00:00 to Wednesday 23:59 MT: spreads, total |
| C-07 | done | C-07 — Referee crew per game, historical from nflverse `officials` plus the weekly assignment release. |
| C-15 | done | C-15 — Every public Big Data Bowl tracking release (10 Hz, all 22 players) into `data/tracking/*.parque |
| P-11 | done | P-11 — `asof()` covers every new table. | 7/7 → n/n. |
| C-08 | done | C-08 — Two to four named beat writers per team through RSS, Bluesky and team sites, via `playerweek/bro |
Desk roster
4 staffed, 1 partial, 0 struck. Every job the operating model charters for this desk (docs/WHAT-I-WAS-MISSING.md #4), against what actually runs in this repo -- not what the tab above implies.
| Job | Status | Evidence |
|---|---|---|
| Data Acquisition & Rights Lead Finds the highest-value missing observations and secures lawful, durable access. | staffed | loop/source_probe.py checks terms of use before a new source is captured (the same check C-16 cites before touching coaches film). |
| Football Event & Ontology Engineer Gives every entity, event, state, clock, and revision one canonical meaning. | partial | playerweek/knowability.py's as-of clocks and playerweek/gameserving.py's serving contract give canonical timing and shape, but no dedicated entity/ontology module exists. |
| Streaming Capture Engineer Preserves information that backtesting can never reconstruct. | staffed | bin/odds-capture.sh and the C-band sensor cadence (buildsite/machine.py) preserve odds and language state that cannot be reconstructed after the fact. |
| Data Reliability Engineer Detects stale, missing, revised, duplicated, contradictory, or impossible data. | staffed | data/capture_health.jsonl, playerweek/watchdog.py, and bin/lane-watchdog.py detect stale or missing sources by their own clock. |
| Feature, Label & Provenance Engineer Builds leakage-safe model inputs and event-keyed outcomes. | staffed | playerweek/features.py and research/playgrain/loop/registry.py build leakage-safe inputs, though registry.sql_columns() is defined but not yet wired into fit_models.py's training set -- a known, recorded gap. |