2021-22 One and One

• G-001 Blue base parallel print run disputed: cardboardconnection.com, gogts.net, cardsmithsbreaks.com, and first-party blog.paniniamerica.net all state #/35; checklistcenter.com and cardlines.com instead state #/30. Resolved to #/35 on 4-source-vs-2-source majority including the first-party blog post; the #/30 figure is rejected but logged here per the hard rule to record disagreements.
• G-002 sportscardchecklist.com describes this set with uneven per-color print counts (Black/Gold/Green/Red 28 cards each, Blue 7 cards, Purple 19 cards) rather than a uniform 28-card run across all parallel colors. This is single-source and not corroborated elsewhere; the 28-card base count and Blue-Purple-Red-Gold-Green-Black color ladder are used in sets.csv, but the uneven per-color subset counts are not asserted anywhere in our files.
• G-003 dacardworld.com's FOTL case listing states a FOTL-exclusive 'Timeless Moments Auto Pink #/3' -- rarer than the FOTL Pink #/15 tier used by the Rookie/Prime Rookie Jersey Autograph sets. This #/3 figure is single-source (only the one FOTL case product page); not independently corroborated.
• G-004 The tail of the extended 151-180 checklist (cards 176-180: Wilt Chamberlain, Larry Bird, Pete Maravich, Oscar Robertson, Chris Webber) is confirmed only by gogts.net; tcdb.com's fetched page content stopped at card 175 (Kevin Garnett) without displaying 176-180 in the fetched excerpt, though the two sources agree exactly through card 175 (both list Kevin Garnett at #175), giving confidence the sequence is genuine and not fabricated. Card-level Legends checklist is out of scope for base_cards.csv regardless (Legends is not the base set).
• G-005 cardboardconnection.com's page text folds the 30-card Legends subset into a stated 'Base Set Size: 180 cards' framing, and sportscardchecklist.com indexes all base parallels as '180 cards' each. This matches the base-set-inflation pattern already confirmed on the 2019-20 and 2020-21 One and One Basketball releases and was rejected here in favor of the 150-card resolution, which is backed by explicit 'Rookies (RC, SN99)' / 'Legends (SN99)' subheadings on tcdb.com's card-level checklist (corroborated by gogts.net and cardsmithsbreaks.com showing the same three-way Base/Rookies/Legends content break).
• G-006 All 10 Noir Sneaker Spotlight card_numbers were written as the plain run 1-10, but NO marketplace seller title naming a player at a specific number could be located to validate the ORDER (the check that was available for Downtown, where goldin.co /item/ lots name '#1 LeBron James' and '#9 Nikola Jokic'). Repeated searches returned only 2021-22 Panini NOIR Sneaker Spotlight cards -- a DIFFERENT product; the ebay.de listing 'https://www.ebay.de/itm/315539331133' titled '2021-22 Panini Noir LeBron James Sneaker Spotlight #6 /99 Miami Heat' carries the structured attribute Set='2021-22 Panini Noir' and Card Number='699', so despite the tempting coincidence that our set also has LeBron/Miami at #6 it is NOT evidence for this set and was rejected. An ebay.com search path returned an unrelated category page (relevance sanity check failed) rather than listings. The numbers rest on tcdb https://www.tcdb.com/Checklist.cfm/sid/338115/2021-22-Panini-One-and-One-Noir-Sneaker-Spotlight plus cardboardconnection https://www.cardboardconnection.com/2021-22-panini-one-and-one-basketball-nba-cards, which agree position-for-position and were shown to be mutually independent (cardboardconnection orders the CODED sets differently from tcdb, so it is not copying tcdb) and where cardboardconnection passed per-set-type calibration on this release's plain inserts. The residual risk is the prizmdp-Manga failure mode: two catalogues both showing ordinals where the card prints a code. This set has no parallel page on tcdb, so the parallel-page test was unavailable. Players affected, in written order: LeBron James (Cleveland), Giannis Antetokounmpo, Devin Booker, James Harden, Tim Duncan, LeBron James (Miami), Donovan Mitchell, Chris Paul, Kawhi Leonard, Rasheed Wallace.
• G-007 Card numbers for seven of this release's coded sets rest entirely on tcdb pages -- the set's own base checklist plus its own colour-parallel checklists, which are the same SITE and therefore not independent corroboration. Affected sets: 75th-Team Signatures (75- prefix), Jersey Autographs (JA-), Dual Jersey Autographs (DJ-), Rookie Autographs (RA-), Timeless Moments (TM-), Timeless Moments Autographs (TM-), Rookie Dual Jersey Autographs (RD-). Only two of the nine coded sets obtained independent marketplace confirmation of a printed code: Rookie Jersey Autographs via the goldin lot https://goldin.co/item/2021-22-panini-one-and-one-rookie-jersey-autograph-rj-jgr-jalen-greenbqvr3 ('#RJ-JGR Jalen Green'), and Prime Rookie Jersey Autographs via the free-text eBay listing https://www.ebay.com/itm/297323859413 ('#PR-DML' Davion Mitchell). Every aggregator checked (cardboardconnection) FAILS calibration on coded sets in this release, so none of them can corroborate these numbers. The scheme itself is well evidenced -- all nine coded sets use a two-character prefix and tcdb's rendering is internally consistent across base and parallel pages -- but each individual suffix in the seven sets above is single-site.
• G-008 PREFIX COLLISION INSIDE ONE RELEASE: the non-autograph insert 'Timeless Moments' and the autograph set 'Timeless Moments Autographs' BOTH use the TM- prefix, and they share at least one identical card_number on different cards -- TM-SCY is Stephen Curry in both sets (a Timeless Moments insert and a Timeless Moments Autographs card). Any resolver keyed on card_number alone will mis-hit; (set_name, card_number) is the only safe key here. This is the same hazard the spec documents for 2024-25 Select's JRS-. Recorded so the collision is visible as data rather than only as prose in the affected rows' notes.
• G-009 CORROBORATION FOUND for the previously single-source claim in G-002. G-002 records that sportscardchecklist.com alone described 75th-Team Signatures with uneven per-colour counts (Black/Gold/Green/Red 28 each, Blue 7, Purple 19) and that this was not corroborated elsewhere. tcdb independently confirms the uneven shape on its own per-colour checklist pages: Blue https://www.tcdb.com/Checklist.cfm/sid/338095/2021-22-Panini-One-and-One-75th-Team-Signatures-Blue renders exactly 7 rows (75-BWT, 75-EHY, 75-ITH, 75-LWL, 75-RBY, 75-RPS, 75-WFZ), Purple https://www.tcdb.com/Checklist.cfm/sid/338098/2021-22-Panini-One-and-One-75th-Team-Signatures-Purple renders exactly 19 rows, and Black https://www.tcdb.com/Checklist.cfm/sid/338094/2021-22-Panini-One-and-One-75th-Team-Signatures-Black renders all 28. Blue and Purple are strict subsets of the 28 base codes. G-002 is left open and unedited per the append-only rule; this row records the new evidence additively. No change was made to sets.csv or parallels.csv.

If you know a source that answers one of these, tell us — name the question's label and where to look.