2021-22 National Treasures

• G-001 tcdb.com's set-overview page states 'Total Cards: 150' for 2021-22 National Treasures Basketball. This is rejected as base_set_size (which is 100, confirmed via checklistcenter.com's and cardboardconnection.com's explicit '100 Cards' statements and full card-by-card cross-check for cards 1-74). tcdb's 150 figure bundles the 100-card base set with the 50-card Rookie Patch Auto set (same subset-bundling pattern the 2019-20 and 2020-21 siblings caught).
• G-002 checklistcenter.com and cardboardconnection.com both state 46 cards for 'Clutch Factor Signatures'; cardsmithsbreaks.com states 50 cards for a set of the same name. Not resolved with card-level evidence (card-level checklists for auto sets are out of scope per spec); recorded card_count=46 (2 of 3 sources) with the disagreement logged here.
• G-003 checklistcenter.com and cardboardconnection.com list 'NBA Greats Signatures' (29 cards) as a native 2021-22 set; cardsmithsbreaks.com instead labels an identically-sized 29-card set '2020-21 NBA Greats Signatures' as a carryover. Could not determine whether these are the same set (mislabeled by one source) or two genuinely distinct sets that happen to share a card count. Recorded as one set to avoid double-counting.
• G-004 Six carryover sets ('2020-21 Definitive Ink' 11 cards, '2020-21 Logoman Autographs' 10 cards, '2020-21 Signatures' 13 cards, '2020-21 Clutch Factor Signatures' 7 cards, '2020-21 Colossal Material Autographs' 9 cards, '2020-21 Validating Marks' 9 cards) are listed only by cardsmithsbreaks.com and were not corroborated by checklistcenter.com or cardboardconnection.com. A seventh line item, '2020-21 Clutch Factor Signatures Prime (1 card)', was omitted entirely as it looks like a parallel-tier artifact (the Prime-parallel count of the Clutch Factor Signatures carryover) rather than a genuine standalone set; not fully resolved.
• G-005 Cards 1-50 and 51-74 of the base checklist were cross-verified exactly between checklistcenter.com and tcdb.com. Cards 75-100 rely on checklistcenter.com alone; a second tcdb.com fetch aimed at this range returned cards 101-150 (the Rookie Patch Auto set) instead of 75-100, and cardboardconnection.com and beckett.com could not be used as a second full-range source (see G-007). Single-source for cards 75-100 specifically.
• G-006 Individual per-set parallel print runs quoted by checklistcenter.com (e.g. 'Bronze #/25 or #/15', 'Prime #/25 or Less') are imprecise ranges rather than single fixed values and are single-source; per the PARALLEL SCOPE instruction only the colour NAMES (not the print-run figures) were promoted into sets.csv notes, so this ambiguity does not affect sets.csv, but the underlying print-run figures themselves remain unverified card-by-card and single-sourced.
• G-007 beckett.com returned HTTP 403 on every fetch attempt for this product. Its set list and any card-level checklist could not be checked, so a 'Colossal Signatures'-style naming trap (seen for the 2019-20 sibling, apparently bled in from a football article) could not be specifically ruled out for this season, though no such name was encountered from any working source.
• G-008 No numeric pull odds, 'per case' language, or SSP/case-hit terminology was found for this product in any source consulted (checklistcenter.com, cardboardconnection.com, gogts.net, cardsmithsbreaks.com, blog.paniniamerica.net's release post and its linked Product Information Sheet PDF, and cardlines.com's review article). Box-average guarantees ('4 autographs, 4 memorabilia, 1 printing plate, 1 base or base parallel per box') are not numeric pull odds and were not used as scarcity evidence. See scarcity_sets.csv / scarcity_parallels.csv / ssp_cards.csv, all legitimately empty.
• G-009 This release's notes describe base cards or a base subset as serial-numbered, but the figure is only in prose and could not be resolved into structured data. base_print_run is UNKNOWN rather than guessed. Resolving it needs either a base_card_tiers partition for this release or a re-read of the source that stated the run.
• G-010 ABSTAINED ENTIRELY - no card-by-card rows written for 'Treasured Tags Auto' (10 cards). tcdb's subset (sid 319956, https://www.tcdb.com/Checklist.cfm/sid/319956/2021-22-) lists all ten cards with bare card numbers 1-10. A contiguous 1..N on a 1/1 autograph set is the classic renumbering/placeholder smell, and tcdb demonstrably uses bare ordinals as placeholders for the other 1/1 sets in this same product (Logoman Autographs, Rookie Logoman, Colossal Logoman), so those digits cannot be trusted as printed numbers. No marketplace listing quoting a 2021-22 Treasured Tags Autographs card number could be found. Writing 1-10 would risk 10 wrong identifiers; abstained instead.
• G-011 SEVERELY PARTIAL because tcdb stores bare ordinal placeholders instead of printed codes for most cards in the four Logoman sets. Written / stated: Logoman Auto 3/37 (sid 319867), Rookie Logoman 35/50 (sid 319903), Colossal Logoman 12/98 (sid 319835), 2020-21 Logoman Autographs 1/10 (sid 319791). In each of these subsets tcdb shows a real letter code (LA-, RL-, CL-) for some rows and a bare number for the rest; the display order proves the bare numbers are internal ordinals rather than printed numbers (e.g. 'LA-DFX' is rendered between '12' and '14'). Rows with bare ordinals were omitted rather than written. 175 of the 195 cards in these four sets remain unresolved.
• G-012 Cards omitted because tcdb's card_number field literally reads 'TBD' (or 'No.'): Auto Triple 1 (Luka Doncic/Stephen Curry/Trae Young), NBA Greats Signatures 2 (Bill Russell, Kevin Johnson), Signatures 2 (Bill Russell, Kevin Johnson), Notable Nicknames Auto 2 (Jerry West 'TBD', Karl Malone 'No.'), Treasured Signatures 1 (Bill Russell), Treasure Chest Memorabilia 1, Lasting Legacies Auto Relic 2 (Steven Adams, Tobias Harris), Century Materials 1 (Gary Trent Jr.), Jersey Treasures 1 (Ivica Zubac), Colossal Logoman 1 (OG Anunoby), Rookie Logoman 1 (Sandro Mamukelashvili). The player is known in every case; only the printed code is missing. Resolving these needs a marketplace listing per card.
• G-013 tcdb (sid 319839 and its Super Prime parallel sid 319841, checked independently) records THREE rows with the identical card number 'CM-DAY': Deandre Ayton, Dirk Nowitzki and Jamal Crawford. Only the Deandre Ayton row was written. The other two were omitted because (a) 'DAY' is transparently derived from Ayton's name and (b) Dirk Nowitzki's code in the sibling Retro Materials set is RM-DNK per a fanatics.com/store.nba.com listing ('#RM-DNK #14/25'), so CM-DAY for Nowitzki looks like a tcdb data-entry artifact rather than a genuine duplicate printed code. 18 of 20 written. Needs a marketplace listing for the Nowitzki and Crawford Colossal Materials cards.
• G-014 tcdb shows a Kevin Durant row numbered 'NM-JNK' in both the base subset (sid 319879) and the Super Prime parallel (sid 319881); the Super Prime subset ALSO carries a separate NM-JNK row for Jusuf Nurkic. Since NM-JNK plainly belongs to Nurkic, the Durant row was omitted rather than written with a code that would resolve to the wrong player. 29 of 30 written.
• G-015 tcdb (sid 319936) records the Rick Barry card as 'PMS-RBY'. PMS- is the Penmanship prefix in this release; every other row in Timeless Talents Signatures uses TTS-. Rick Barry does not appear in the Penmanship checklist, so this looks like a tcdb prefix typo (probably TTS-RBY), but that is an inference and was not written. 26 of 27.
• G-016 Card-level work surfaced these count disagreements with sets.csv (whose counts came from checklistcenter.com / cardboardconnection.com / cardsmithsbreaks.com): Century Materials - sets.csv 39, tcdb base subset 34 and Super Prime 35; the union of the two comes to 38 coded cards plus one 'TBD', i.e. exactly 39, which CORROBORATES sets.csv. Game Gear Auto - sets.csv 50, tcdb subset total 49 (and its Super Prime parallel is also 49): unresolved, one card either does not exist or is missing from tcdb. 2020-21 Colossal Material Autographs - sets.csv 9 (cardsmithsbreaks.com only), tcdb subset total 8: unresolved. Jersey Treasures - sets.csv 96, tcdb base 84 / Super Prime 95, union 95 coded + 1 'TBD' = 96, CORROBORATES sets.csv.
• G-017 While mapping tcdb's subset block for this product, three named insert sets were found on tcdb that have NO row in sets.csv: 'Treasured Tags' (50 cards, tcdb sid 319955 - distinct from the 10-card 'Treasured Tags Auto' that IS in sets.csv), 'Treasured Threads' (29 cards, sid 319957, with Prime and Super Prime parallels), and 'Tremendous Treasures' (28 cards, sid 319965, with Prime and Super Prime parallels). sportscardchecklist.com's set index for this release lists all three as well. sets.csv was NOT modified (out of scope for this task); recorded here so a sets-level pass can decide whether they belong.
• G-018 SOURCE CALIBRATION FINDING. For this release sportscardchecklist.com and breakninja.com both renumber the coded insert sets, and - unusually - their fake sequence is NOT contiguous. For National Archives Ink both sites print 1..50 with holes at 3, 9, 15, 17, 30 and 31 (44 rows), which superficially looks like a genuine printed run with holes. The real printed numbers are NAI-xxx letter codes, confirmed by comc.com ('#NAI-CAZ - Cazzie Russell /49', '#NAI-VCA') and auctionofchampions.com ('#NAI-DTH'). So the heuristic 'holes in the sequence mean you are reading printed numbers' FAILS on these two sites for this product: they emit a holey pseudo-sequence. Only tcdb.com and marketplace listing titles were used for card_number here.
• G-019 SINGLE-SOURCE numbering. tcdb (sid 319845) gives this 48-card set the numbers 101-150 minus 102 and 146 - i.e. exactly the Rookie Patch Auto block with two holes. The Rookie Patch Auto 101-150 numbering is itself well corroborated (tcdb sid 319916 plus beckett.com '#107 Jalen Green' and goldin.co '#123 Cade Cunningham'), but no independent listing quoting a CROSSOVER card's number was found, so it remains possible that tcdb cloned the RPA numbering onto this subset and the Crossover cards actually carry their own numbering. Rows were written as tcdb prints them; treat the base-mirroring as unconfirmed.
• G-020 Confirmed prefix collisions - (set_name, card_number) is the only safe key for this release. 'CM-' is used by FOUR sets: Century Materials, Colossal Material, Colossal Material Auto and 2020-21 Colossal Material Autographs; e.g. CM-AEW is Anthony Edwards in both Century Materials and Colossal Material Auto, CM-ZWL is Zion Williamson in both Century Materials and Colossal Material, and CM-NKV/CM-SGA appear in the 2020-21 carryover set as well. 'PSI-' is used by both Peerless Signatures (PSI-ADV etc.) and Rookie Private Signings Icon Version (PSI-CWI). Prefixes also change between seasons of this same line: the 2021-22 Signatures set uses SIG- while the 2020-21 carryover Signatures subset inside the same product uses S-, and a sibling season's National Archives Ink uses NA- rather than NAI-.
• G-021 Duplicate card numbers written verbatim because they are genuine printed duplicates (each pair shares plausible initials and both players are real members of the set): Definitive Ink DI-MYT (Mychal Thompson / Myles Turner); Game Gear Auto GGA-SAM (Sam Cassell / Sam Jones); Lasting Legacies Auto Relic LL-JHD (Jrue Holiday / Justin Holiday) and LL-MTB (Matt Barnes / Matt Bonner); Jersey Treasures JT-JHD (James Harden / Jrue Holiday), JT-MKB (Malik Beasley / Mikal Bridges) and JT-TYH (Tyler Herro / Tyrese Haliburton). These make card_number non-unique within a set; flagged rather than de-duplicated.
• G-022 FABRICATION CHECK NOTE, not a defect. Rookie Materials (RMR-), Rookie Dual Materials (RD-), Rookie Triple Materials (RTM-) and Rookie Jumbo Materials (RJM-) each carry the SAME 25 rookies with the SAME three-letter suffixes, differing only in prefix - which trips the 'suspicious uniformity' check. It was investigated and is genuine: these are four material formats of one 25-player rookie subject list, each fetched from its own tcdb subset (sids 319904, 319897, 319928, 319900) and each returning its own distinct prefix. The neighbouring Colossal Rookie Materials (CR-, 46 cards) has a different and longer list, so the block is not a template fill.

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