2022-23 NBA Hoops Premium Box Set

• G-001 EXTRACTION_SPEC 5.2 BASE-CARD IDENTITY COLLISION, naming both releases: hoopspremium-bkb-2022-23 (this release) and paninihoops-bkb-2022-23 describe THE SAME 300 physical cards. This release is the /199 Panini-Direct online-exclusive Premium Box Set printing of that checklist; the parent is the pack-issued unnumbered printing (paninihoops-bkb-2022-23 carries base_print_run=unnumbered, base_print_run_basis=assumed). A source states the repackaging in words rather than leaving it to be inferred: cardboardconnection, "An online-exclusive series, 2022-23 NBA Hoops Premium Box Set offers all 300 cards from the main 2022-23 NBA Hoops checklist on a premium foil stock." (https://www.cardboardconnection.com/2022-23-nba-hoops-premium-box-set-basketball-cards); Panini America, "Hoops Premium Box Sets include a complete 300 card set which includes Base, Base Rookie, and Base Hoops Tribute all numbered to 199!" (https://blog.paniniamerica.net/available-tomorrow-2-2-2022-23-hoops-nba-basketball-premium-box-set/). MEASURED, NOT ESTIMATED, against data/paninihoops-bkb-2022-23/base_cards.csv with csv.DictReader: POSITIONAL overlap (row index against row index, comparing card_number+player_name+team_or_school) = 296 of 300. BY-CARD-NUMBER overlap (keyed on card_number, comparing player_name+team_or_school) = 296 of 300. Zero card numbers appear on only one side; all 300 player_name values match position for position and number for number; the 4 differences are ALL and ONLY team_or_school, on cards 197, 198, 217 and 221, and are enumerated with their quotes in G-009. The (player_name, team_or_school) multiset is therefore NOT identical on the two sides, although the player_name multiset is. BOTH FIGURES ARE REPORTED DELIBERATELY, because they can diverge badly and a single figure misleads: the 2020-21 pair of this same line reads 175/270 positionally against 269/270 by card number (270/270 after its 3b-to-98 correction), purely because one row was numbered "3b" and shifted every row after it, and a later marker on that release then wrongly asserted the POSITIONAL figure had become 270/270 when only the by-number figure had moved. In 2022-23 the two measures agree at 296/300 because both files are a clean contiguous 1-300 in the same order, so no shift can arise. Built per 5.2 rather than deduplicated or double-counted. Note the same physical object is ALSO recorded twice inside paninihoops-bkb-2022-23: as a base-set-scope parallel row "Premium Box Set" print_run 199 (its gap G-006) and as a sets.csv row "Base Hoops Tribute Premium Box Set Autograph" (its gaps G-009 and G-023). This release IS the separate release record; a resolver pass should decide which side keeps them. paninihoops-bkb-2022-23 gap G-006 states the modelling question in its own words - "whether the Premium Box Set should be modelled as a parallel of this release or as its own product - cardboardconnection maintains a separate page for it while beckett lists it among this release's base parallels" - and this release is the catalogue answering "its own product" for the fourth consecutive season.
• G-002 parallels.csv is written with its header and ZERO data rows, which is a finding and not an unfinished file (EXTRACTION_SPEC 2.3). VERIFIED FOR THIS SEASON RATHER THAN ASSUMED FROM THE PRIOR THREE. EVIDENCE FOR THE ABSENCE: https://www.tcdb.com/Inserts.cfm/sid/335315/2022-23-Hoops enumerates the parent set's complete inventory under the verbatim headings "Insert Sets (81)" and "Parallel Sets (22)". A whole-page search for the string "Premium" returns exactly ONE entry in the entire inventory - "Premium Box Set" (sid 345819) among the 22 parallels. There is no "Premium Box Set Gold", no "Premium Box Set Holo", no sub-parallel of it of any kind, and no entry containing the word "Tribute" either. The page was read TWICE under two distinct URLs (plain, and with a harmless &sort=0 style parameter appended) so the second reading is not a 15-minute cache hit on the first, and both readings returned the identical heading counts and the identical single Premium entry. A narrow verbatim re-fetch of the set page itself under a cache-busted URL confirms the word "parallel" does not appear on it (https://www.tcdb.com/ViewSet.cfm/sid/345819/2022-23-Hoops---Premium-Box-Set). cardboardconnection's product page describes only the /199 base cards and the one Tribute autograph and names no parallel (https://www.cardboardconnection.com/2022-23-nba-hoops-premium-box-set-basketball-cards); checklistcenter's page likewise names none (https://www.checklistcenter.com/2022-23-panini-nba-hoops-premium-box-set-basketball-card-checklist/); Panini America's own launch post describes only the complete 300-card set /199 plus one autograph and mentions no parallel (https://blog.paniniamerica.net/available-tomorrow-2-2-2022-23-hoops-nba-basketball-premium-box-set/). This matches hoopspremium-bkb-2019-20, hoopspremium-bkb-2020-21 and hoopspremium-bkb-2021-22, all also parallel-free, making four consecutive seasons. It remains true that no source says in words "this box set has no parallels"; the absence is inferred from a complete enumeration read twice plus three product descriptions, not from a statement. One contamination shape was investigated and nothing was written from it - see G-014.
• G-003 EXTRACTION_SPEC 5.4 UNIT CHECK AT RELEASE LEVEL - THE TRAP THIS LINE HAS NOW SPRUNG IN ALL FOUR SEASONS, AND IT RECURS HERE WITH THE SAME SHAPE AND A DIFFERENT NUMBER. cardboardconnection prints "**Product Configuration:** 301-card factory set" on the same page as "300 cards. Serial numbered #/199. Full set per box." and "An online-exclusive series, 2022-23 NBA Hoops Premium Box Set offers all 300 cards from the main 2022-23 NBA Hoops checklist on a premium foil stock." (https://www.cardboardconnection.com/2022-23-nba-hoops-premium-box-set-basketball-cards). 301 is CARDS PER SEALED BOX: the 300-card base set plus the one Tribute autograph. THIS SEASON A SECOND SOURCE NAMES THE UNIT OUTRIGHT, which the prior seasons did not have: checklistcenter prints "301 Cards Per Box" (https://www.checklistcenter.com/2022-23-panini-nba-hoops-premium-box-set-basketball-card-checklist/). 301 is therefore written into NO field; base_set_size is 300, sourced from tcdb Total Cards "300" (https://www.tcdb.com/ViewSet.cfm/sid/345819/2022-23-Hoops---Premium-Box-Set), from Panini America's "a complete 300 card set", from cardboardconnection's "300 cards", and corroborated by counting 300 rendered rows and tiles one by one. Both load-bearing strings were re-fetched under narrow verbatim prompts that demanded them inside quotation marks with their surrounding words (EXTRACTION_SPEC 4.1) before either was believed; the tcdb re-fetch, issued under a cache-busted URL, additionally confirms that neither "301" nor "271" appears anywhere on the structural source. FOR THE RECORD OF HOW THIS TRAP HAS BEHAVED ACROSS THE LINE: 2019-20 = 305 and 303, and those two belonged to a DIFFERENT product (Premium Stock); 2020-21 = 271, belonging to this product; 2021-22 = 271 again, same product, same page, same words; 2022-23 = 301, same product, same page, same words, scaled to the larger base set. A fifth season should expect it again and should expect the figure to be base_set_size + 1.
• G-004 TASK TRAP 3 TESTED: "NBA Hoops Premium Stock" IS A DIFFERENT, ONGOING RETAIL LINE AND NOTHING FROM IT IS MERGED HERE. Nothing in this release derives from a Premium Stock source. THE NEGATIVE IS STRONGER THIS SEASON THAN IN 2020-21 OR 2021-22 BECAUSE IT RESTS ON AN ENUMERATION AND NOT ONLY ON A SEARCH: the parent set inventory https://www.tcdb.com/Inserts.cfm/sid/335315/2022-23-Hoops, read twice under two distinct URLs, lists "Insert Sets (81)" and "Parallel Sets (22)" and contains no entry whose name includes the word "Stock". A targeted search for a 2022-23 Premium Stock product returned only 2023-24 items - "2023-24 Panini NBA Hoops Premium Stock Basketball Factory Sealed Blaster Box" (amazon) and "2023-24 Panini NBA Hoops Premium Stock Basketball Bundle Box" (ebay) - with nothing for 2022-23. That agrees with this project's own already-gated data/paninihoops-bkb-2022-23/release.csv, which states in its notes "There is no 2022-23 Hoops Premium Stock product; Premium Stock begins with 2023-24.", and with hoopspremium-bkb-2020-21 G-004 and hoopspremium-bkb-2021-22 G-004, which found no such product in their seasons either. reference/brands.csv still carries no row for a Premium Stock line; that escalation is already open as G-003 on hoopspremium-bkb-2019-20 and is NOT re-decided here.
• G-005 license_status WAS NOT SUPPLIED BY THE TASK BRIEFING and the catalogue is inconsistent on it, so the value written here is a convention choice and is recorded as one. EXTRACTION_SPEC 2.1 says license_status comes "as given in your task prompt"; this prompt gave product_name, season, manufacturer and league only, and settled the question by instruction: match this line's own prior seasons and gap it, do not reconcile unilaterally. This release therefore writes "licensed (NBA/NBPA)", matching hoopspremium-bkb-2019-20, hoopspremium-bkb-2020-21 and hoopspremium-bkb-2021-22 byte for byte. The parent product for the SAME season, data/paninihoops-bkb-2022-23/release.csv, writes "licensed (NBA)" for what is physically the same checklist under the same licence (measured: 296/300 identical, differing only on four team attributions - see G-001 and G-009). WHAT THE SOURCES ACTUALLY SAY: nothing. No source consulted in this pass - tcdb sid 345819, tcdb sid 335315, cardboardconnection, checklistcenter, blog.paniniamerica.net - prints the word "NBPA", the phrase "NBA Players Association", or any licensing statement at all about either product. The distinction between the two values is not sourced on either side and is a catalogue convention, not a fact either release found. The columns that actually gate the loader (licensed=yes, league=NBA) are unambiguous and identical on both sides. A repair pass should pick one value for the pair hoopspremium/paninihoops per season, or source the difference. This is the fourth consecutive season of this line to record the identical divergence.
• G-006 CHANNEL IS KNOWN AND SOURCED BUT HAS NOWHERE TO LIVE IN THE SCHEMA, so it is recorded here. Panini America's launch post is headed "AVAILABLE TOMORROW (2/2) | 2022-23 Hoops NBA Basketball Premium Box Set ONLINE EXCLUSIVE" and states the product goes on sale "Thursday, February 2nd at 11am (CST) at the Panini Direct Store" and that "there are only 199 sets produced" (https://blog.paniniamerica.net/available-tomorrow-2-2-2022-23-hoops-nba-basketball-premium-box-set/). cardboardconnection calls it "An online-exclusive series" (https://www.cardboardconnection.com/2022-23-nba-hoops-premium-box-set-basketball-cards). In the taxonomy of EXTRACTION_SPEC 2.3 that is channel=panini-direct with channel_basis=sourced-exclusive and, by the derived map, channel_class=promotion. release.csv has no channel column and parallels.csv has no rows, so NO machine-readable channel value exists anywhere in this release and a consumer filtering by channel will not find it. Identical disposition to hoopspremium-bkb-2020-21 G-007 and hoopspremium-bkb-2021-22 G-006. Note for a schema pass: this is the FOURTH consecutive season of this line where a fully sourced channel is unrepresentable, and the parent release records the mirror-image problem from the other side - data/paninihoops-bkb-2022-23/parallels.csv writes channel=UNKNOWN on its "Premium Box Set" row and its gap G-006 says "The allowed channel values ... have no online/direct-to-consumer option". NOTE the allowed-value list quoted in that gap is STALE against the current spec, which does list panini-direct; the value it needed exists now.
• G-007 card_tier IS WRITTEN ON 281-300 ONLY, IT IS SOURCED FROM THE PARENT RATHER THAN FROM THIS PRODUCT, AND IT DIVERGES FROM THE PARENT RELEASE FILE; all three facts are recorded rather than smoothed. This release writes card_tier="Hoops Tribute" on cards 281-300 and leaves it empty on 1-280. BASIS: the PARENT set renders the literal flag token "HT" on exactly those 20 tiles and on no others - https://www.tcdb.com/Gallery.cfm/sid/335315/2022-23-Hoops?PageIndex=29 renders "281 - Paolo Banchero - Orlando Magic *HT*" through "290 - Anthony Davis - Los Angeles Lakers *HT*" and https://www.tcdb.com/Gallery.cfm/sid/335315/2022-23-Hoops?PageIndex=30 renders "291 - Trae Young - Atlanta Hawks *HT*" through "300 - Vince Carter - Toronto Raptors *HT*", ten of ten tiles flagged on each page, while parent pages 8, 9, 10, 19, 20 and 28 render no HT token at all. THIS PRODUCT'S OWN RENDERER IS SILENT: the Premium Box Set's own gallery pages 29 and 30 and its own Checklist page 3 render only "SN199" on 281-300 and carry no HT or TRIB token, which is a REGRESSION from 2021-22, where the Premium Box Set's own checklist did render "TRIB" on its Tribute block (hoopspremium-bkb-2021-22 G-007), and matches 2020-21, which had to read the token off the parent only. NOTE the expansion of the token "HT" to the words "Hoops Tribute" comes from Panini America ("Base Hoops Tribute") and cardboardconnection ("The Hoops Tribute Premium Box Set Autographs set"), NOT from tcdb - EXTRACTION_SPEC 3.3 warns that the fetch layer invents glosses for tcdb flag tokens, so the token is recorded here as the literal string it is. TWO DIVERGENCES: (a) data/paninihoops-bkb-2022-23/base_cards.csv leaves card_tier EMPTY on all 300 rows including 281-300, although its release notes describe the same three segments in prose ("Base #1-230 (veterans), Base Rookies #231-280 (RC), Base Hoops Tribute #281-300"); (b) the 231-280 rookie block gets no card_tier here, following hoopspremium-bkb-2019-20 G-009, hoopspremium-bkb-2020-21 G-010 and hoopspremium-bkb-2021-22 G-007, which all left their rookie block empty for the same reason - even though Panini America DOES name the segment in words ("Base, Base Rookie, and Base Hoops Tribute"), so a uniform repair pass across the line could now fill it with a citation. I did not change the line's convention unilaterally mid-line. The base run is NOT treated as tiered for print-run purposes: it is uniformly /199, so base_print_run is the integer 199 and not varies-by-tier, and base_card_tiers.csv is correctly absent (EXTRACTION_SPEC 2.5).
• G-008 rookie_flag RESTS ON THE PARENT SET, NOT ON THIS PRODUCT'S OWN CHECKLIST - with a positive control and COMPLETE row-or-tile coverage of all 300 numbers, which is the strongest position any season of this line has reached. tcdb renders NO "RC" token anywhere on the Premium Box Set's own rows or tiles (it renders "SN199" and nothing else, on all 300), the same behaviour recorded in hoopspremium-bkb-2019-20 G-008, hoopspremium-bkb-2020-21 G-009 and hoopspremium-bkb-2021-22 G-008: a parallel gallery is a good print-run source and a bad rookie-flag source. rookie_flag was therefore read off the PARENT set sid 335315, for the identical card numbers. POSITIVE CONTROL RUN FIRST, because "no RC found" is worthless if the renderer never prints RC: https://www.tcdb.com/Gallery.cfm/sid/335315/2022-23-Hoops?PageIndex=24 was fetched and DOES render it, verbatim "231 - Paolo Banchero - Orlando Magic RC" and "240 - Johnny Davis - Washington Wizards RC", on all ten tiles 231-240, with the example caption quoted back as "[231](...) - [Paolo Banchero](...) *RC*". COVERAGE ACTUALLY OBSERVED ON THE PARENT, number by number: cards 1-73 via https://www.tcdb.com/Checklist.cfm/sid/335315/2022-23-Hoops?PageIndex=1 (answered "RC does not appear on any row of this page"); 71-100 via Gallery.cfm PageIndex 8, 9 and 10 (10 tiles each, all three answered RC absent); 101-184 via Checklist.cfm PageIndex=2 (answered "RC appears on no row of this page"); 181-200 via Gallery.cfm PageIndex 19 and 20 (10 tiles each, both answered RC absent); 201-272 via Checklist.cfm PageIndex=3 which renders RC beginning at 231 and quoted "231 | Paolo Banchero | RC | Orlando Magic" and "272 | Luke Travers | RC | Cleveland Cavaliers", so 201-230 are positively observed WITHOUT RC on the same page that carries it; 271-280 via Gallery.cfm PageIndex=28, ten of ten tiles flagged RC; 281-300 via Gallery.cfm PageIndex 29 and 30, both answering RC absent and both flagging all ten tiles HT instead. Every one of the 300 numbers is therefore covered by at least one endpoint and no range rests on block structure alone. RESIDUAL WEAKNESS, STATED RATHER THAN HIDDEN: for cards 241-270 the RC designation rests on Checklist.cfm PageIndex=3 summarising the contiguous range 231-272 rather than on 30 individually quoted rows; the range's two endpoints were quoted verbatim and its tail 271-280 was independently confirmed tile by tile. Corroborated structurally by Panini America's "Base, Base Rookie, and Base Hoops Tribute" (https://blog.paniniamerica.net/available-tomorrow-2-2-2022-23-hoops-nba-basketball-premium-box-set/) and by data/paninihoops-bkb-2022-23, which independently records rookie_flag=yes for exactly 231-280 and no elsewhere - MEASURED with csv.DictReader, zero rookie_flag disagreements across all 300 cards.
• G-009 CONTRADICTS EXISTING DATA ON FOUR TEAM ATTRIBUTIONS, and this release wrote what its own source prints. Measured with csv.DictReader against data/paninihoops-bkb-2022-23/base_cards.csv: 296 of 300 cards match exactly on card_number, player_name and team_or_school, both positionally and by card number; all 300 match on card_number and player_name; the only four differences are team_or_school, and in every one of the four the parent release says "Utah Jazz" and tcdb says something else. THE FOUR, with this release's source quoted verbatim from the Premium Box Set's OWN gallery, each re-fetched under a cache-busted URL as a narrow verbatim check (EXTRACTION_SPEC 4.1) before being written: "197 - Malik Beasley SN199 - Minnesota Timberwolves" and "198 - Jarred Vanderbilt SN199 - Minnesota Timberwolves" (https://www.tcdb.com/Gallery.cfm/sid/345819/2022-23-Hoops---Premium-Box-Set?PageIndex=20 ; parent file says Utah Jazz for both); "217 - Bojan Bogdanovic *SN199* - Detroit Pistons" (https://www.tcdb.com/Gallery.cfm/sid/345819/2022-23-Hoops---Premium-Box-Set?PageIndex=22 ; parent file says Utah Jazz); "221 - Royce O'Neale *SN199* - Brooklyn Nets" (https://www.tcdb.com/Gallery.cfm/sid/345819/2022-23-Hoops---Premium-Box-Set?PageIndex=23 ; parent file says Utah Jazz). tcdb is internally consistent on all four - its Checklist and Gallery renderers agree on 217 and 221, and Gallery agrees with itself across two differently-parameterised reads on 197 and 198. WHAT THE PATTERN ACTUALLY IS, so a later pass does not "fix" it in the wrong direction: cards 194-200 sit in the Minnesota Timberwolves team block and cards 215-222 in the Utah Jazz team block, and all four players changed teams in the 2022 offseason (Beasley and Vanderbilt Minnesota-to-Utah, Bogdanovic Utah-to-Detroit, O'Neale Utah-to-Brooklyn). tcdb keeps the BLOCK team on 197 and 198 and gives the NEW team on 217 and 221, so tcdb is not applying one rule; the parent release resolves all four to Utah Jazz, which is the new team for 197/198 and the old/block team for 217/221, so it is not applying one rule either. The two catalogues are inconsistent in OPPOSITE directions on different cards. NEITHER SIDE IS OBVIOUSLY RIGHT AND I DID NOT PICK A WINNER SILENTLY: I wrote what this product's own source prints and recorded the contradiction here. Cross-check available for a repair pass: this release and the parent agree exactly on card 73 ("73 - Collin Sexton - Utah Jazz", the new team for a Cleveland-block card), card 114 (Kentavious Caldwell-Pope, Denver Nuggets, new team in a Washington block), card 215 (Donovan Mitchell, Cleveland Cavaliers) and card 216 (Rudy Gobert, Minnesota Timberwolves) - so the disagreement is confined to four cards, not to a convention. I did not edit paninihoops-bkb-2022-23 (one agent, one release).
• G-010 TWO tcdb RENDERERS DISAGREED WITH EACH OTHER ON ONE CARD AND THE DISAGREEMENT WAS RESOLVED BY THREE INDEPENDENT READS; recorded because it calibrates how far Checklist.cfm can be trusted on the team column. https://www.tcdb.com/Checklist.cfm/sid/345819/2022-23-Hoops---Premium-Box-Set?PageIndex=1 rendered card 73 as "73 | Collin Sexton | SN199 | Cleveland Cavaliers", and the PARENT's Checklist.cfm PageIndex=1 independently rendered it "73 | Collin Sexton | Cleveland Cavaliers" when asked for that row by name. https://www.tcdb.com/Gallery.cfm/sid/345819/2022-23-Hoops---Premium-Box-Set?PageIndex=8 rendered "73 - Collin Sexton - Utah Jazz", and the parent's Gallery.cfm PageIndex=8 returned the caption quoted verbatim as "73 - Collin Sexton - Utah Jazz". data/paninihoops-bkb-2022-23/base_cards.csv, built from a different source mix, also carries Utah Jazz. UTAH JAZZ IS WRITTEN: two Gallery reads on two different sids plus this project's own already-gated parent release, against two Checklist reads. Cards 66-73 are the Cleveland Cavaliers team block and Sexton was traded to Utah in the 2022 offseason, so the most economical explanation is that the Checklist renderer - or the summarising fetch layer reading it - carried the team-block header down onto a row whose own team differs from the block. THAT EXPLANATION IS A HYPOTHESIS AND IS WRITTEN INTO NO FIELD. It is also not universally true on this page: the same Checklist read correctly gave 26 Alec Burks as Detroit Pistons, 34 Danny Green as Memphis Grizzlies, 54 Malcolm Brogdon as Boston Celtics, 60 Jerami Grant as Portland Trail Blazers and 69 Lauri Markkanen as Utah Jazz, all of which are non-block teams and all of which match the parent release. CONSEQUENCE FOR THIS PASS: every team value taken from a Checklist page and NOT confirmed by a Gallery tile is weaker than the rest. The ranges that rest on Checklist alone are 1-70, 101-180 and 201-280 (the Gallery top-ups cover 71-100, 181-200 and 281-300); within those ranges the parent release agrees on every card except 217 and 221 (G-009), which is the strongest available corroboration short of re-reading 230 rows through a second renderer.
• G-011 THE AUTOGRAPH SET IS NOT IN tcdb AT ALL THIS SEASON AND ITS SERIAL NUMBERING IS CONTESTED; card_count=20 is written, no print run is written anywhere. (1) NOT CATALOGUED: https://www.tcdb.com/Inserts.cfm/sid/335315/2022-23-Hoops was read twice under two distinct URLs and contains no entry whose name includes "Tribute" and exactly one whose name includes "Premium" (the base set, sid 345819). Every prior season of this line had a dedicated sid for this set - 2019-20 sid 218337 "Premium Box Set Autographs" Total Cards "20"; 2021-22 sid 296224 "Hoops Tribute Premium Box Set Autographs" Total Cards "20" - so there is NO row-level reading and NO header count for it in 2022-23 and card_count rests entirely on aggregator prose. (2) card_count=20 rests on two sources stating twenty: cardboardconnection, "The Hoops Tribute Premium Box Set Autographs set is home to 20 top players with each card limited to 10 copies." (https://www.cardboardconnection.com/2022-23-nba-hoops-premium-box-set-basketball-cards), and checklistcenter, which states the set is 20 cards (https://www.checklistcenter.com/2022-23-panini-nba-hoops-premium-box-set-basketball-card-checklist/). Those two are NOT clearly independent - see G-012. (3) THE NUMBERING IS GENUINELY CONTESTED AND IS THEREFORE IN NO FIELD: cardboardconnection says "limited to 10 copies"; Panini America says "all numbered to as low as /10" (https://blog.paniniamerica.net/available-tomorrow-2-2-2022-23-hoops-nba-basketball-premium-box-set/), which is a CEILING-shaped phrase and per EXTRACTION_SPEC 2.3 is not a print run; checklistcenter says the set is "Serial Numbered #/20", which is also exactly its card count and is therefore a strong candidate for the EXTRACTION_SPEC 5.4 unit confusion rather than a real third value. sets.csv has no print_run column in any case (EXTRACTION_SPEC 2.2), so nothing could have been written even had the sources agreed. (4) CONTRADICTS EXISTING DATA, in the same direction the prior two seasons did: data/paninihoops-bkb-2022-23/sets.csv carries this set as "Base Hoops Tribute Premium Box Set Autograph" with card_count=UNKNOWN, and that release's own gap G-023 already states "the set is 20 cards" and "sets.csv not modified (out of scope for this pass); flagged so the count can be corrected". This release writes the 20 that gap asked for. Note the 2021-22 pair had the same argument with the opposite defect - paninihoops-bkb-2021-22 wrote card_count=10, the PRINT RUN, for the same 20-card set (hoopspremium-bkb-2021-22 G-009). I did not edit paninihoops-bkb-2022-23; the repair belongs to a pass that owns it. [CORRECTED 2026-09-17: POINT (4) ONLY IS RESOLVED; THIS GAP STAYS OPEN. paninihoops-bkb-2022-23 was set to card_count=20 on 2026-09-13, so the contradiction described in point (4) no longer exists and that release's G-023 is now resolved. Points (1), (2) and (3) are untouched and remain the reason this gap is open: the autograph set is still absent from tcdb for this season so there is no row-level reading and no header count; card_count=20 still rests on two aggregators that are not clearly independent (see G-012); and the serial numbering is still genuinely contested three ways and is therefore still in no field. Resolving this gap would need a source tcdb does not carry.]
• G-012 WHICH PARTS OF THIS RELEASE REST ON A SINGLE SOURCE, AND WHERE "TWO SOURCES" MAY BE ONE OPINION. (a) THE ENTIRE BASE CHECKLIST - all 300 card numbers, players and teams - rests on tcdb sid 345819 alone as a first-hand reading; no other site was fetched card by card. Its strongest corroboration is structural rather than independent: data/paninihoops-bkb-2022-23/base_cards.csv, built in an earlier pass from a different source mix (beckett, steelcitycollectibles), agrees on 296 of 300 rows both positionally and by card number (G-001, G-009), and checklistcenter independently prints the first and last rows, "1 Jayson Tatum - Boston Celtics" and "300 Vince Carter - Toronto Raptors" (https://www.checklistcenter.com/2022-23-panini-nba-hoops-premium-box-set-basketball-card-checklist/). (b) THE AUTOGRAPH SET rests entirely on aggregators plus the manufacturer with NO structural renderer at all (G-011). (c) cardboardconnection and checklistcenter agree closely on this product and their independence CANNOT be assumed - the same caution hoopspremium-bkb-2019-20 G-010, hoopspremium-bkb-2020-21 G-011 and hoopspremium-bkb-2021-22 G-010 raised. They do diverge on one value, which is mild evidence for independence: cardboardconnection says the autographs are numbered to 10, checklistcenter says /20. (d) THE GENUINE THIRD SOURCE IS THE MANUFACTURER: Panini America's own launch post (https://blog.paniniamerica.net/available-tomorrow-2-2-2022-23-hoops-nba-basketball-premium-box-set/) is first-party and not derived from either aggregator, and it independently states the 300-card set, the /199 numbering, the one autograph "numbered to as low as /10", the 199 sets produced and the February 2 on-sale date. So base_set_size, base_print_run and release_date each rest on three sources of which one is the manufacturer; the base CHECKLIST does not. (e) beckett was NOT fetched in this pass and no beckett page was opened. (f) MARKETPLACE CORROBORATION, seen as search-result titles rather than fetched pages and therefore load-bearing for nothing: "2022-23 Hoops Premium Box Set #94 Cody Martin /199" (https://www.ebay.com/itm/315775238160) matches this checklist at 94 Cody Martin Charlotte Hornets; "2022-23 Panini NBA Hoops Premium Box Set 170/199 Gabe Vincent #103 Rookie RC k1r" (https://www.ebay.com/itm/336764175932) matches at 103 Gabe Vincent Miami Heat and confirms the /199 numbering by showing a copy number, BUT ITS "Rookie RC" IS FALSE - card 103 is outside the 231-280 rookie block and Gabe Vincent first played in the NBA in 2020-21, so the EXTRACTION_SPEC 4.3 temporal check fires; nothing was written from that title. "2022-23 Panini NBA Hoops Complete Basketball Set 1-300 All RC's Tribute Banchero" (https://www.ebay.de/itm/388253753046) restates the 1-300 shape with a Tribute segment.
• G-013 ENDPOINT AND FETCH-COST MEASUREMENTS FROM THIS PASS, confirming EXTRACTION_SPEC 3.2 and correcting the task briefing's budget for a 300-card run. (1) The corrected 3.2 recipe is right and the old flat 10:1 claim would have lost the tail of every page. Measured on sid 345819: Checklist.cfm?PageIndex=1 delivered complete rows 1-73 of 100 and stopped; ?PageIndex=2 delivered 101-184 of 101-200 and stopped; ?PageIndex=3 delivered 201-281 of 201-300 and stopped. Truncation therefore sat at rows 73, 84 and 81 of 100 - inside the 74-86 band six agents measured, with row 73 falling one below its floor. (2) THE TASK BRIEFING'S FIGURE OF "3 Checklist + 6 Gallery = 9 fetches for 270 cards" DOES NOT TRANSFER TO 300 CARDS, and the reason is structural, not statistical: a 270-card run has a third Checklist page of only 70 rows, which sits under the truncation point and arrives complete, so only TWO page-tails need topping up. A 300-card run has three FULL 100-row pages and therefore THREE tails. Measured cost here: 3 Checklist + 7 Gallery (pages 8, 9, 10, 19, 20, 29, 30) = 10 fetches for 300 cards, against 30 for Gallery alone. The rule that generalises is ceil(N/100) Checklist plus about 3 Gallery pages per FULL 100-row Checklist page. (3) NO Gallery tile was silently dropped this pass - all seven top-up pages and all nine parent pages returned 10 of 10 - so the dropped-tile defect hoopspremium-bkb-2020-21 G-012 recorded is intermittent, not systematic. The cache-busting discipline was still applied: every re-check was issued under a URL varied by a harmless extra parameter, never as a second call to an identical URL. (4) NO 503 was returned on any of the ~26 tcdb fetches in this pass, at a concurrency of 3. (5) NO fetch refused to transcribe on copyright grounds this pass; the prompt shape used was "transcribe the rows in the form number | player | flags | team", which the briefing predicted would fare better than a range-bounded ask, and it did.
• G-014 CONTAMINATION SHAPE FOUR WAS TESTED FOR THIS SEASON; ITS STRUCTURAL PRECONDITION IS PRESENT BUT NO 2022-23 INSTANCE WAS FOUND, and nothing was written either way. In 2021-22 an eBay seller title read "2021-22 Panini Hoops Premium Box Set Silver Kevin Huerter /199 #148 Hawks SP", inventing a "Silver" parallel OF the Premium Box Set; the explanation was that "Silver" is a SEPARATE /199 base parallel of the MAIN Hoops set, so a seller holding one /199 Hoops card had two true-sounding labels (hoopspremium-bkb-2021-22 G-014). THE SAME PRECONDITION HOLDS IN 2022-23: data/paninihoops-bkb-2022-23/parallels.csv carries BOTH a base-set-scope "Silver" at print_run 199 channel hobby AND a base-set-scope "Premium Box Set" at print_run 199, two distinct /199 tiers of the same 300-card checklist, exactly the configuration that produced the false label last season. A targeted search for a 2022-23 instance returned NONE - the "Premium Box Set Silver" titles it returned are "2024-25 Panini NBA Hoops Premium Box Set Silver Foil VETS #/199 CHOOSE YOUR CARD" (https://www.ebay.com/itm/398072766815) and the known 2021-22 one (https://www.ebay.com/itm/188488755273, "2021-22 Panini Hoops Premium Box Set Silver Lauri Markkanen /199 #55 Bulls SP"). So the shape is LIVE ON THIS LINE in at least two other seasons and simply did not surface for 2022-23 in this pass. NEGATIVE RESULT CAVEAT: absence in a search is weaker than absence in an enumeration, so this records "no 2022-23 instance was found", not "none exists". parallels.csv stays at zero rows (G-002), and a later pass meeting a "2022-23 Hoops Premium Box Set Silver" title should read it as the main set's Silver parallel and not as a parallel of this release. NOTE the 2024-25 title additionally invents a tier word this catalogue does not carry - "Silver Foil" - which is a fifth variant of the same shape.
• G-015 PRODUCT-NAME VARIANTS SEEN IN PROSE, NONE WRITTEN, NO BRAND ESCALATION NEEDED (EXTRACTION_SPEC 2.1/5.1 - report, do not decide). product_name is written as "NBA Hoops Premium Box Set Basketball", which is what reference/brands.csv carries for brand hoopspremium (brd_01M2B8M6BJKF4BVCZ1A77FRZ27) and what hoopspremium-bkb-2019-20, hoopspremium-bkb-2020-21 and hoopspremium-bkb-2021-22 all carry, byte for byte. The sources agree on the identity: tcdb "2022-23 Hoops - Premium Box Set" (https://www.tcdb.com/ViewSet.cfm/sid/345819/2022-23-Hoops---Premium-Box-Set), cardboardconnection page heading "2022-23 NBA Hoops Premium Box Set Basketball Cards Checklist" (https://www.cardboardconnection.com/2022-23-nba-hoops-premium-box-set-basketball-cards), checklistcenter "2022-23 Panini NBA Hoops Premium Box Set - Basketball Card Checklist" (https://www.checklistcenter.com/2022-23-panini-nba-hoops-premium-box-set-basketball-card-checklist/), Panini America "2022-23 Hoops NBA Basketball Premium Box Set" and "Hoops Premium Box Set" (https://blog.paniniamerica.net/available-tomorrow-2-2-2022-23-hoops-nba-basketball-premium-box-set/). VARIANTS DELIBERATELY NOT WRITTEN ANYWHERE: none of this season's sources used the 2021-22 prose variant "Premium Collector's Series Set", and no dacardworld "Premium Collectors Set" listing was found for 2022-23, so the variant recorded in hoopspremium-bkb-2021-22 G-015 did not recur; it is repeated here only so a later agent meeting it first does not mint a second brand row from it. Also not written: the parent release data/paninihoops-bkb-2022-23/sets.csv names the autograph set "Base Hoops Tribute Premium Box Set Autograph" (singular, with a leading "Base"), while this release uses cardboardconnection's plural form "Hoops Tribute Premium Box Set Autographs", which also matches hoopspremium-bkb-2021-22 byte for byte. That is a set-name divergence between two releases describing one object, not a brand question, and is recorded rather than reconciled.
• G-016 No pack odds and no scarcity tier are recorded for either set. This is a sealed complete factory box set with fixed contents sold as a single unit online, not a pack product - cardboardconnection says "Full set per box" and "1 per box" and checklistcenter says "Each box set includes the full 300 card base set, as well as 1 of the exclusive signed cards." - so per-set pack odds do not exist in the ordinary sense; no source consulted states any odds figure. stated_odds, odds_evidence_url, odds_evidence_quote and scarcity_tier are therefore empty on both sets.csv rows. Same disposition as hoopspremium-bkb-2019-20 G-012, hoopspremium-bkb-2020-21 G-013 and hoopspremium-bkb-2021-22 G-012.
• G-017 PRINT-RUN FIGURE AVAILABLE THAT THE SCHEMA CANNOT CARRY, recorded here so it is not lost. Panini America states "there are only 199 sets produced" (https://blog.paniniamerica.net/available-tomorrow-2-2-2022-23-hoops-nba-basketball-premium-box-set/). That is the number of SEALED BOX SETS made, which is a property of the product and happens to equal the per-card serial numbering /199 - unsurprising, since one set contains exactly one copy of each card. base_print_run=199 is written on the strength of the per-card statements ("each base card will be numbered to /199"; tcdb "Serial numbered to 199"; cardboardconnection "Serial numbered #/199"; the per-row flag "SN199" on all 300 rows and tiles), NOT on the strength of the 199-sets-produced figure, because those are two different facts that coincide. THE COINCIDENCE IS A HAZARD WORTH RECORDING: a later pass reading only "199 sets produced" would have a number of the right size for the wrong reason, which is the same class of error as the 301-cards-per-box trap in G-003 running in the opposite direction. base_print_run_basis is "sourced" and the quotes above are what sources it.

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