2018-19 Impeccable

• G-001 INDEPENDENT CORROBORATION IS THINNER THAN THE SOURCE COUNT SUGGESTS. beckett.com is unreachable from this environment (CAV-033); every beckett.com/news URL surfaced by search was left unfetched because the site 302s to maintenance. So this release rests on exactly two lineages: tcdb, and cardboardconnection - with checklistcenter used once (the 101-150 block names) and being the same editorial lineage as cardboardconnection, not a third party. Where this report says "two sources agree" it means tcdb plus that one editorial lineage.
• G-002 NO MARKETPLACE CORROBORATION WAS OBTAINED FOR ANY PRINT RUN IN THIS RELEASE. The scope memo records 2018-19 as the season with the richest marketplace corroboration, but a targeted search for a serial off a card in hand ("2018-19 Impeccable" Silver "/49") returned only catalogue pages - cardboardconnection and unreachable beckett.com articles - and no goldin, comc or eBay listing. Every integer print run written here therefore rests on tcdb serial tokens plus cardboardconnection, with no seller-transcribed third party.
• G-003 ROOKIE FLAG: a within-product positive discriminative test FIRED and was deliberately not acted on. tcdb renders no RC token on any of base cards 1-100 while rendering RC on every one of cards 101-150 in the same Checklist.cfm rendering of the same parent sid 192094, and independently no player in the 1-100 run is a 2018-19 rookie (the 2018 draft class sits entirely in the 101-150 autograph blocks). That is the strongest form of the absence test the specs describe, and it supports rookie_flag=no. It was still written as UNKNOWN on all 100 rows because EXTRACTION_SPEC 2.4 licenses yes/no only where the source explicitly shows an RC designation, and all five sibling Impeccable seasons already in the catalogue carry UNKNOWN on all 100 base rows. This gap exists so the held fact is not lost to the UNKNOWN. https://www.tcdb.com/Checklist.cfm/sid/192094/2018-19-Panini-Impeccable?PageIndex=2
• G-004 FIFTEEN OF THE TWENTY-NINE SET-SCOPE PARALLEL TIERS ARE PROVABLY SERIAL NUMBERED WITH A PER-CARD VARIABLE RUN, AND THE SCHEMA CANNOT SAY SO (EXTRACTION_SPEC 2.3). Each carries print_run=UNKNOWN and serial_numbered=UNKNOWN, which is indistinguishable from "nothing was found" - the exact collapse unknown_semantics.csv exists to prevent. The enumerated token sets are in each row's visual_notes. The fifteen: Elegance Retired Jersey Autographs Holo Silver (SN10 x8, SN5 x1); Elegance Veteran Jersey Autographs Holo Gold (SN5 x9, SN3 x1) and Holo Silver (SN10 x9, SN5 x1); Illustrious Ink Holo Gold (SN10 x33, SN5 x5) and Holo Silver (SN25 x31, SN10 x8); Immortal Ink Holo Gold (SN10 x25, SN5 x3, SN3 x1) and Holo Silver (SN25 x26, SN10 x2, SN5 x1); Impeccable Shots Signatures Holo Gold (SN10 x18, SN5 x8, SN3 x1) and Holo Silver (SN25 x19, SN10 x6, SN5 x2); Impeccable Stars Signatures Holo Gold (SN5 x14, SN10 x13, SN3 x1) and Holo Silver (SN25 x15, SN10 x10, SN5 x3); Impeccable Victory Signatures Holo Gold (SN10 x15, SN5 x11, SN3 x3) and Holo Silver (SN25 x17, SN10 x10, SN5 x2); Indelible Ink Holo Gold (SN10 x24, SN5, SN3 x1) and Holo Silver (SN25 x26, SN10 x2, SN5 x1). MEASUREMENT WORTH RECORDING: 15 of 29 is 52 percent, against the 34 percent the 2018-19 Immaculate build measured and the 22 percent measured on 2018-19 Noir. A three-row sample of any of these fifteen would have produced a clean-looking sourced integer that was wrong. https://www.tcdb.com/Inserts.cfm/sid/192094/2018-19-Panini-Impeccable
• G-005 SETS.CSV CANNOT HOLD A SET'S OWN PRINT RUN AND ON THIS PRODUCT ALMOST EVERY SET HAS ONE. Seventeen of the twenty-six sid-bearing sets carry a PER-CARD VARIABLE run enumerated off their own rows (the Elegance pair, Illustrious Ink, Immortal Ink, all seven team sets, Assists, Championships, Jersey Number, Points, Rebounds, Shots, Stars, Stats, Victory, Indelible Ink, Stainless Stars Autographs), and five have a UNIFORM set-level run that the schema still cannot carry: Gold NBA Logo /10, Silver NBA Logo /22, Stainless Stars /99, Impeccable Rookie Signatures /99, plus the two parent-filed blocks Rookie Autographs /99 and Elegance Rookie Jersey Autographs /99. All figures are recorded in words in each set's notes, per EXTRACTION_SPEC 2.2, and none was put in card_count. Impeccable Jersey Number Autographs is additionally the CAV-023 scheme - each card numbered to that player's own jersey number.
• G-006 CARD COUNT CONTESTED, ILLUSTRIOUS INK: 39 written. tcdb's base sid 198227 states "Total Cards: 39" and renders 39 rows, and its Holo Silver parallel sid 198229 independently states and renders 39. cardboardconnection states "38 cards", and tcdb's OWN Holo Gold (198228) and Platinum (198230) tiers both state 38 and render 38. So the disagreement is not simply tcdb against an aggregator - it is inside tcdb, two sids saying 39 and two saying 38, with the aggregator siding with 38. 39 was written because the set's own base page and one parallel agree on it and a tier roster is not required to be complete; the contest is real and unresolved. https://www.tcdb.com/ViewSet.cfm/sid/198227/2018-19-Panini-Impeccable-Illustrious-Ink ; https://www.tcdb.com/ViewSet.cfm/sid/198228/2018-19-Panini-Impeccable-Illustrious-Ink-Holo-Gold
• G-007 CARD COUNT CONTESTED, SILVER NBA LOGO: 100 written against cardboardconnection's "50 cards. Serial numbered #/22". tcdb sid 198267 states "Total Cards: 100" and rendered 77 rows before truncating, every one flagged SN22 - 77 rendered rows refutes a 50-card set outright, so this is an editorial figure disproved by enumeration rather than a standoff. Note the sibling season impeccable-bkb-2019-20 records Silver NBA Logo at 49 cards, so the set roughly doubled or the 2019-20 figure has the same defect; not investigated, outside this slice. https://www.tcdb.com/ViewSet.cfm/sid/198267/2018-19-Panini-Impeccable-Silver-NBA-Logo
• G-008 THREE OF CARDBOARDCONNECTION'S STATED PRINT-RUN RANGES ARE REFUTED BY ENUMERATION OF THE SAME SETS' ROWS. It states Impeccable Assists Autographs as "#/14 through #/11" while tcdb renders SN14, SN13, SN11, SN10, SN9 and SN8; Impeccable Championships Autographs as "#/7 through #/2" while tcdb renders SN11 and SN8 as well; Impeccable Points Autographs as "#/34 through #/26" while tcdb renders SN24, SN22 and SN20 as well. In all three cases the stated range understates the low end. Nothing was written from any of these ranges - a range is not a print run in any case - but the pattern is worth recording, because the same source's unhedged flat figures (the five base parallels) WERE relied on after they matched tcdb's tokens five for five.
• G-009 TEAM AUTOGRAPH SET TOTALS DISAGREE WITH CARDBOARDCONNECTION BY ONE. The seven single-team sets have tcdb Total Cards of 8 (76ers), 15 (Celtics), 17 (Knicks), 20 (Lakers), 8 (Pistons), 9 (Rockets) and 10 (Warriors), summing to 87. cardboardconnection does not break them out and states one combined "Impeccable Autographs Team Set" of "86 cards". The difference is exactly the Rockets ambiguity in the next gap: at Rockets=8 the sum is 86 and the two sources reconcile.
• G-010 IMPECCABLE ROCKETS: TWO ENDPOINTS OF ONE SID GAVE DIFFERENT ROW COUNTS AND 9 WAS WRITTEN. ViewSet.cfm on sid 198244 states "Total Cards: 9" and self-reported 8 rendered rows with four tokens summing to 8; Checklist.cfm on the same sid under ?sort=0 rendered 9 named rows (Charles Barkley, Hakeem Olajuwon, Tracy McGrady, Clyde Drexler, then RO-CCP, RO-RSP, RO-RHY, RO-RAL, RO-RTJ). Two of three signals say 9 and the header agrees with the fuller render, so 9 was written; the third signal is what would have made the team sum reconcile with cardboardconnection at 86. Unresolved. https://www.tcdb.com/Checklist.cfm/sid/198244/2018-19-Panini-Impeccable-Impeccable-Rockets?sort=0
• G-011 TWO AUTOGRAPH SETS HAVE NO tcdb SID AND ARE FILED INSIDE THE PARENT CHECKLIST PAST THE BASE RUN (EXTRACTION_SPEC 5.3). Rookie Autographs is cards 101-110 (AU, RC, SN99, no MEM) and Elegance Rookie Jersey Autographs is cards 111-150 (AU, MEM, RC, SN99). Both are in sets.csv and both are EXCLUDED from base_set_size, which is 100. The 110/111 boundary was confirmed on three independent renderings (the parent's own PageIndex=2, and the Holo Silver and Platinum parallel sids' PageIndex=2). This matches how all five sibling Impeccable seasons were built. It does NOT match contenders-bkb-2019-20, which sets base_set_size=165 including its Rookie Ticket autographs 101-165, or contenders-bkb-2021-22 at 165; the catalogue-wide convention split documented in EXTRACTION_SPEC 5.3 is therefore still live and this release is on the exclusion side of it.
• G-012 NO CLAIM IS MADE ABOUT VARIATION SETS, AND Errors.cfm WAS NOT USED AS EVIDENCE OF ABSENCE. Per CAV-020 its silence discriminates nothing, so it was not fetched for this release. What is asserted about the base run rests only on contiguous enumeration: cards 1-100 rendered contiguously with no b-suffixed or duplicate number seen on either Checklist.cfm or Gallery.cfm, and the parent's "Total Cards: 150" reconciles exactly with 100 base plus the 50-card block. A variation set filed as b-suffixed rows would not necessarily surface in that enumeration, so its absence is unsupported rather than disproved.
• G-013 THE PARALLEL TIERS ARE NOT NESTED AND SEVERAL TIER ROSTERS ARE SHORTER THAN THEIR BASE SET. Elegance Veteran Jersey Autographs is 10 cards but its Platinum tier states "Total Cards: 9"; Impeccable Stars Signatures is 28 but its Platinum tier states 27; Illustrious Ink is 39 (or 38, see the count gap) but its Holo Gold and Platinum tiers both state 38. Whether a player is simply absent from the scarcer tier or the tier page is short was not established - it would need a roster diff per tier, which is a chase-pass job. parallels.csv carries no roster, so nothing was written either way.
• G-014 SERIAL TOKENS WERE DROPPED BY THE FETCH LAYER IN BOTH DIRECTIONS INSIDE THIS PRODUCT, AS EXTRACTION_SPEC 3.2 warns. Impeccable Championships Autographs renders 18 rows but its per-row tokens account for 17; Impeccable Jersey Number Autographs renders 19 rows against 18 tokens; Impeccable Rebounds Autographs self-reported 9 rendered rows while its own token tally sums to 10 and its header states 10. Separately the MEM flag moved between renderings of the same 50 rows: the parent and the Holo Silver and Platinum parallel sids all put the MEM boundary at 110/111, while the Holo Gold sid put it at 106/107. No value was written off any single fetch's silence, and the MEM boundary was taken from the three that agree.
• G-015 BASE PARALLEL COVERAGE IS NOT 100 PERCENT OF ROWS AND THE UNSEEN WINDOWS ARE NAMED RATHER THAN IGNORED. Serial tokens were read on cards 1-78 and 91-100 for Silver, 1-84 and 91-100 for Gold, 1-76, 81-90 and 101-150 for Holo Silver, 1-74, 81-90 and 101-150 for Holo Gold, and 1-78, 81-90 and 101-150 for Platinum. Cards 79-90 (Silver), 85-90 (Gold), 77-80 (Holo Silver), 75-80 and 91-100 (Holo Gold) and 79-80 and 91-100 (Platinum) were never re-rendered. A token change hiding inside one of those interior windows is not excluded by enumeration; it is excluded only by cardboardconnection stating each figure flatly and by both ends of each tier agreeing.
• G-016 DERIVED-COLUMN DIVERGENCE AGAINST THE SIBLING SEASONS, NOT RESOLVED HERE. parallel_family was derived by lifting FAMILIES and family_of out of pipeline/fix_schema.py by AST, as this round's brief requires, which gives Silver=other, Gold=other, Platinum=Platinum, Holo Silver=Holo, Holo Gold=Holo. impeccable-bkb-2019-20 carries "Metal" on its Silver, Gold and Platinum rows, a family the repo's ordered FAMILIES list does not contain and family_of cannot produce. So the Impeccable line now answers the family question two different ways across adjacent seasons. Nothing was added to fix_schema.py. This is the CAV-021 taxonomy question in a new instance.
• G-017 CHANNEL BASIS DIVERGENCE AGAINST THE SIBLING SEASONS, DELIBERATE. Every parallel row here carries channel=hobby with channel_basis=sourced-presence, cited to Steel City Collectibles' product page titled "2018-19 Panini Impeccable Basketball Hobby Box" - a source that places the product in the hobby channel without claiming exclusivity, which is what sourced-presence means per CAV-008. impeccable-bkb-2019-20 carries the same channel with channel_basis=from-note and no citation at all. The two seasons therefore make the same claim at different evidentiary strengths. No sibling row was touched. https://www.steelcitycollectibles.com/i/2018-19-panini-impeccable-basketball-hobby-box
• G-018 THE 2018-19 SCOPE MEMO FILES THIS PRODUCT UNDER A HEADING THAT IS STALE, AND THIS IS THE FOURTH PROVEN INSTANCE. specs/2018-19_SCOPE_MEMO.md section 1 lists "Panini Impeccable" under "No brands.csv row (11)". reference/brands.csv carries brd_01M2B8M6BJKF4BVCZ1A77FRZ26,Impeccable,impeccable,Panini Impeccable Basketball,basketball,Panini, and five Impeccable releases (2019-20 through 2023-24) are already built against it. The memo's own dated corrections already record the heading as stale for Contenders Optic, Absolute Memorabilia and Dominion; Impeccable is a fourth. Nothing was minted for this release and the season-independent product name matched byte for byte.
• G-019 SET NAMES FOR THE TWO PARENT-FILED BLOCKS COME FROM cardboardconnection, NOT FROM tcdb, AND checklistcenter PRINTS DIFFERENT ONES. sets.csv writes "Rookie Autographs" and "Elegance Rookie Jersey Autographs", matching cardboardconnection's section headings and matching the names impeccable-bkb-2019-20 already uses for the structurally identical blocks. checklistcenter renders the same two blocks as "2018-19 Panini Impeccable - Rookie Auto Set" and "2018-19 Panini Impeccable - Elegance Rookie Jersey Auto Set". tcdb gives these blocks no name at all because they have no sid. A resolver matching on checklistcenter's strings will miss both rows. https://www.checklistcenter.com/2018-19-panini-impeccable-basketball/
• G-020 SID ENUMERATION IS COMPLETE AND BOTH BOUNDARIES OF BOTH BLOCKS WERE PROVED FOREIGN, BUT ONE SET SITS OUTSIDE THEM. The base parallels occupy 198192-198196 (198191 is "2018 Panini Plates & Patches - Upper Echelon Red", 198197 is "1989 Pacific Cards & Comics Team USA") and the inserts occupy 198218-198271 (198217 is "1998 Playoff Prestige Retail - Award Winning Performers (Blue Foil)", 198272 is "2017-18 Extreme Kingston Frontenacs (OHL)"). Every interior sid in both ranges is accounted for by Inserts.cfm - 5 parallels plus 54 inserts - so nothing is hiding inside them. Impeccable Rebounds Autographs is sid 198641, far outside both, found only because Inserts.cfm listed it; the 198273-198640 stretch was NOT swept, so a further out-of-block Impeccable sid that Inserts.cfm also omitted would not have been found.
• G-021 CARD-LEVEL CHECKLISTS ARE OUT OF SCOPE FOR THIS PASS (CHASE_CHECKLIST_SPEC governs them) and insert_cards.csv was not written. One observation for whoever does that pass: Impeccable Rockets renders a MIXED page - four bare ordinals (1 Charles Barkley, 2 Hakeem Olajuwon, 3 Tracy McGrady, 4 Clyde Drexler) alongside five real codes (RO-CCP, RO-RSP, RO-RHY, RO-RAL, RO-RTJ) - which is the dominant premium-line pattern, and Elegance Retired Jersey Autographs prints ER- codes (ER-CBK, ER-RPS). The bare ordinals are placeholders, not printed numbers.
• G-022 NO base_card_tiers.csv WAS WRITTEN AND NONE IS REQUIRED. The base run is one untiered block of 100 cards at a single print run of /99 - no named tier, no sub-brand block, no per-tier print run, and no unissued interior number. base_print_run is the integer 99 rather than varies-by-tier, so the tier file would have nothing to hold and card_tier is empty on all 100 base rows.

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