2018-19 Contenders Optic

• G-001 THE SCOPE MEMO UNDERCOUNTS THE MISSING RELEASES AND THIS RELEASE IS THE PROOF. specs/2018-19_SCOPE_MEMO.md section 6 tabulates Contenders Optic as missing from 2019-20, 2020-21, 2021-22, 2022-23 and 2023-24 - five seasons - and does NOT list 2018-19, while its own section 1 table files "Panini Contenders Optic / 2019-07-31" among the eleven lines with no brands.csv row and calls it "borderline", quoting cardboardconnection to the effect that it is "not a standalone product but rather a variant application". Both readings are wrong about the same product. tcdb indexes "2018-19 Panini Contenders Optic" on the 2018 season index with its OWN parent sid 206591, distinct from "2018-19 Panini Contenders" (188543) and "2018 Panini Contenders Draft Picks" (178771); it has its own release date (July 31, 2019 against Contenders' December 26, 2018), its own hobby box configuration (1 pack of 6 per box against Contenders' 12 packs of 10), its own 100-card base checklist that coincides with Contenders' in exactly ONE of 100 slots, its own 30-card Rookie Ticket block, and its own inventory of 18 sid-bearing insert and autograph sets and 67 sid-bearing parallel subsets. It is a standalone product. The memo's section 6 count of "roughly 28 basketball releases missing from seasons already treated as finished" is understated by AT LEAST ONE for a different reason than the ones it lists: 2018-19 is not a season the catalogue treats as finished, so this release does not increase that 28, but the memo's per-line statement "Contenders Optic ... missing from 2019-20 through 2023-24" is wrong at its lower bound - the line starts at 2018-19, so the Contenders Optic backfill is SIX seasons, not five. A human should correct the section 6 table's Contenders Optic entry and the section 1 "borderline" classification. Note also that brands.csv ALREADY carries the contendersoptic line (brd_01M2B8M6BJKF4BVCZ1A77FRZ21, product_name "Panini Contenders Optic Basketball"), so the memo's "no brands.csv row" column is stale for this line and nothing needed minting. This build was also flagged by data/contenders-bkb-2018-19 gap G-019, which found the same hole two days earlier from the other side. Evidence: https://www.tcdb.com/ViewAll.cfm/sp/Basketball/year/2018 ; https://www.tcdb.com/ViewSet.cfm/sid/206591/2018-19-Panini-Contenders-Optic ; https://www.cardboardconnection.com/2018-19-panini-contenders-optic-basketball-cards
• G-002 CONVENTION DIVERGENCE UNDER SECTION 5.3, RECORDED NOT RESOLVED. This release writes base_set_size=100 and files the 101-130 Rookie Ticket autograph block as two separate auto sets, applying the spec's exclusion rule. TWO EXISTING RELEASES IN THIS CATALOGUE DISAGREE, both in the sibling Contenders line: data/contenders-bkb-2019-20/release.csv carries base_set_size=165 and 165 rows in base_cards.csv with card_type values base and rookie-auto, and data/contenders-bkb-2021-22/release.csv likewise carries 165. Six releases agree with this one: contenders 2018-19 = 100, contenders 2020-21 = 100, contenders 2022-23 = 100, contenders 2023-24 = 100, contendersoptic 2019-20 = 100 and contendersoptic 2020-21 = 100. Verified against the repo, not presumed. WHAT THIS RELEASE ADDS TO THE ARGUMENT ALREADY MADE BY data/contenders-bkb-2018-19 G-001: that release could not carry its block as base cards because tcdb renders its numbers "101a"/"101b" and cards.card_number is INTEGER NOT NULL. THIS release cannot carry its block as base cards for a DIFFERENT and independent reason - tcdb renders no letter suffix at all here, so the plain and variation rows of card 101 are two distinct cards sharing the single integer 101, and 30 numbers render 60 rows. A 130-card reading would need 60 rows for 30 numbers and a 160-card reading would need numbers nobody prints. So in the earliest season of BOTH lines of this family, under two different renderings, the include convention is unavailable in principle. That is further evidence that 165 is a per-agent judgement about one continuous numbering rather than a house convention for the line. Recommended repair, for a human to gate: contenders 2019-20 and contenders 2021-22 move to base_set_size=100 with their 101-165 rows re-filed as auto sets. NEITHER RELEASE WAS EDITED BY THIS BUILD. Evidence: https://www.tcdb.com/Checklist.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=2 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=11&sort=0 ; https://www.tcdb.com/Errors.cfm/sid/206591/2018-19-Panini-Contenders-Optic
• G-003 STRUCTURAL DIFFERENCE FROM THE SAME-SEASON SIBLING, WORTH ITS OWN RECORD. data/contenders-bkb-2018-19/release.csv states that its Rookie Ticket block renders as "101a" and "101b" - a plain row and a photo-variation row per rookie, distinguished by a letter suffix on the card number - and its parent header "Total Cards: 145" counts the 45 DISTINCT numbers against 190 rendered rows. This product, the same manufacturer in the same season with the same block design, renders BOTH members of each pair under the same plain integer: Checklist.cfm?PageIndex=2 returns "101, Collin Sexton, Cleveland Cavaliers, (AU, RC)" immediately followed by "101, Collin Sexton, Cleveland Cavaliers, (AU, RC, VAR)", and Gallery.cfm?PageIndex=11&sort=0 renders the same pair as two tiles both numbered 101, with the fetch explicitly noting "Card numbers lack letter suffixes in the source material". Errors.cfm confirms the pairing from a third endpoint, returning 60 rows for 30 numbers annotated "BASE:" and "VAR:". So tcdb uses two different disambiguation schemes for the same structure in the same season across two products of one family. Consequences: (1) an agent carrying a letter-suffix expectation across from the sibling would look for suffixes here and conclude the variation block is missing; (2) the integer-column argument that makes the 165 convention impossible in the sibling takes a different form here, namely a uniqueness collision rather than a parse failure; (3) any downstream de-duplication keyed on (release, card_number) will silently merge the plain and variation autographs of this product. Evidence: https://www.tcdb.com/Checklist.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=2 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=11&sort=0 ; https://www.tcdb.com/Errors.cfm/sid/206591/2018-19-Panini-Contenders-Optic
• G-004 THE PARENT HEADER COUNTS DISTINCT CARD NUMBERS, WHICH IS A THIRD THING AGAIN. "Total Cards: 130" on sid 206591 equals neither the base set size (100) nor the count of rendered rows (160). It counts distinct card numbers across the base run and the excluded autograph block, because each of the 30 block numbers renders twice. Section 3.2 warns that Total Cards counts different things on different lines and section 5.3 forbids sizing a block off it; this release is another instance, and the briefing's remark that the header "has now been seen counting four different things" is corroborated rather than extended - this is the same unit data/contenders-bkb-2018-19 found (145 distinct numbers against 190 rendered rows). Anyone sizing the base set off this header would have written 130 and anyone sizing the block off it would have written 30 numbers as 30 cards. The block was sized off the last rendered page instead: Gallery PageIndex=16 ends at 130 and PageIndex=17 renders no tiles, and Checklist PageIndex=3 renders no rows. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/206591/2018-19-Panini-Contenders-Optic ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=16&sort=0 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=17&sort=0 ; https://www.tcdb.com/Checklist.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=3
• G-005 BASE PRINT RUN IS ASSUMED, NOT SOURCED. base_print_run=unnumbered with base_print_run_basis=assumed. No source states in words that the base Season Ticket cards carry no serial numbering. The evidence is strong and indirect and section 2.1 names exactly this case: tcdb prints literal serial tokens on the parallel rows of this very set - "SN99" on Blue, "SN49" on Orange, "SN10" on Gold, "SN1" on Gold Vinyl and the four Printing Plates, and "AU, SN149"/"AU, SN99"/"AU, SN25"/"AU, SN10"/"AU, SN1" on the autograph block - and prints NO token on any of the 100 base rows on Checklist.cfm or on Gallery.cfm pages 8, 9 and 10. checklistcenter and cardboardconnection both list the base ladder with Silver and Red carrying no figure, in a list where every other tier carries one. That is good evidence and it is still not a statement, so the basis is assumed. Evidence: https://www.tcdb.com/Checklist.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=1 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=8&sort=0 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=9&sort=0 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=10&sort=0 ; https://www.checklistcenter.com/2018-19-panini-contenders-optic-basketball/ ; https://www.cardboardconnection.com/2018-19-panini-contenders-optic-basketball-cards
• G-006 RELEASE DATE RESTS ON TWO SOURCES AND BECKETT COULD NOT BE REACHED FOR A THIRD. The tcdb parent renders "Release Date: July 31, 2019" and cardboardconnection renders "July 31, 2019". Neither labels it an estimate, so 2019-07-31 is written rather than UNKNOWN. beckett.com remains unreachable from this environment - the scope memo records six consecutive passes and data/contenders-bkb-2018-19 records a seventh, and nothing in this pass reached it either. Note also that the date is a JULY 2019 date on a product carrying a 2018-19 season prefix, seven months after the sibling Contenders released on 2018-12-26; that is consistent with the memo's own listing of this product at 2019-07-31 and with the next season's Contenders Optic releasing 2020-09-23, and it is not an error. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/206591/2018-19-Panini-Contenders-Optic ; https://www.cardboardconnection.com/2018-19-panini-contenders-optic-basketball-cards
• G-007 A SYSTEMATIC ONE-ROW SHORTFALL ON THE PLAIN TIER OF THREE AUTOGRAPH SETS - A COUNT PATTERN, NOT THREE INDEPENDENT CONTESTS. On three sets tcdb's PLAIN subset renders exactly one row fewer than tcdb's OWN Gold subset of the same set, and checklistcenter states the higher figure while naming in a parenthetical exactly the player the plain subset lacks. Measured: (1) Veteran Season Ticket Autographs - plain sid 210846 states "Total Cards: 29" and its 29 rows were enumerated by hand with no VT-GAN row, Gold sid 210850 states "Total Cards: 30" and its 30 rows were enumerated by hand and include "VT-GAN - Giannis Antetokounmpo", Gold Vinyl and Printing Plate Black also state 30, and checklistcenter states "30 Cards (*No Giannis)"; (2) NBA Ink - plain sid 210862 states 29, Gold sid 210861 states 30 and renders "NI-KKZ" Kyle Kuzma, checklistcenter states "30 Cards - Serial Numbered #/99 or as Noted (*No Kuzma)"; (3) Sophomore Contenders Autographs - plain sid 210864 states 4 and its four rows were enumerated with no Kuzma, Gold sid 210865 states 5 and its five rows were enumerated and include SC-KKZ Kyle Kuzma, checklistcenter states "5 Cards - Serial Numbered #/99 (*No Kuzma)". sets.csv writes 30, 30 and 5 - the figure two independent renderings support. WHAT IS NOT RESOLVED: whether the missing plain-tier row is a tcdb cataloguing hole or a real product fact (a signer whose card exists only in the numbered tiers). checklistcenter's parenthetical can be read either way and its placement is ambiguous - on the Veteran line it sits immediately after the count, on the NBA Ink and Sophomore lines after the serial statement. The reading is left open on purpose, because the written count is the same under both. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/210846/2018-19-Panini-Contenders-Optic-Veteran-Season-Ticket-Autographs?sort=0 ; https://www.tcdb.com/ViewSet.cfm/sid/210850/2018-19-Panini-Contenders-Optic-Veteran-Season-Ticket-Autographs-Gold?sort=0 ; https://www.tcdb.com/ViewSet.cfm/sid/210862/2018-19-Panini-Contenders-Optic-NBA-Ink ; https://www.tcdb.com/ViewSet.cfm/sid/210861/2018-19-Panini-Contenders-Optic-NBA-Ink-Gold ; https://www.tcdb.com/ViewSet.cfm/sid/210864/2018-19-Panini-Contenders-Optic-Sophomore-Contenders-Autographs ; https://www.tcdb.com/ViewSet.cfm/sid/210865/2018-19-Panini-Contenders-Optic-Sophomore-Contenders-Autographs-Gold ; https://www.checklistcenter.com/2018-19-panini-contenders-optic-basketball/
• G-008 THE FOUR CONTENDERS TRIBUTE YEAR SUBSETS AND CHECKLISTCENTER'S ONE SECTION ARE THE SAME TEN CARDS, AND THIS IS THE SECTION 5.4 CONFIRMATION TO LOOK FOR RATHER THAN A CONTEST. tcdb splits the Contenders Tribute autographs into four year subsets with their own sids and states "Total Cards: 5" (1981, sid 210856), "Total Cards: 3" (1998, 211140), "Total Cards: 1" (2003, 211142) and "Total Cards: 1" (2007, 211144) - and every one of those ten rows was enumerated individually, not taken from the headers. checklistcenter carries a single section "2018-19 Panini Contenders Optic - Contenders Tribute Auto Set" at "10 Cards - Serial Numbered #/10", and every tcdb row renders the literal token "AU, SN10". Two catalogues decomposing the same number the same way. sets.csv files the four tcdb subsets separately because they carry distinct sids and distinct printed prefixes (81-, 98-, 03-, 07-), which a resolver needs; a consumer summing this line must sum the four rows and must NOT also count a tenth row for the checklistcenter grouping. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/210856/2018-19-Panini-Contenders-Optic-Contenders-Tribute-Autographs-1981 ; https://www.tcdb.com/ViewSet.cfm/sid/211140/2018-19-Panini-Contenders-Optic-Contenders-Tribute-Autographs-1998 ; https://www.tcdb.com/ViewSet.cfm/sid/211142/2018-19-Panini-Contenders-Optic-Contenders-Tribute-Autographs-2003 ; https://www.tcdb.com/ViewSet.cfm/sid/211144/2018-19-Panini-Contenders-Optic-Contenders-Tribute-Autographs-2007 ; https://www.checklistcenter.com/2018-19-panini-contenders-optic-basketball/
• G-009 SET-LEVEL PRINT RUNS CANNOT BE STORED AND ARE RECORDED IN WORDS ONLY, per section 2.2 - sets.csv has no print_run column and card_count must not carry one. Four autograph sets in this release are serial numbered at SET level and none of those figures is a parallel tier, so none of them is in parallels.csv either: Legendary Contenders Autographs renders "AU, SN99" on its rows and checklistcenter states "Serial Numbered #/99 or as Noted"; Up and Coming Contenders Autographs renders "AU, SN99"; NBA Ink renders "AU, SN99", "AU, SN49" and "AU, SN25" on different rows of one sid, i.e. a per-card variable run; all four Contenders Tribute year subsets render "AU, SN10" uniformly and checklistcenter states "Serial Numbered #/10". Sophomore Contenders Autographs renders "AU, SN99" and checklistcenter states "#/99". These figures exist in the notes of the relevant sets.csv rows and nowhere structured. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/210858/2018-19-Panini-Contenders-Optic-Legendary-Contenders-Autographs ; https://www.tcdb.com/ViewSet.cfm/sid/210867/2018-19-Panini-Contenders-Optic-Up-and-Coming-Contenders-Autographs ; https://www.tcdb.com/ViewSet.cfm/sid/210862/2018-19-Panini-Contenders-Optic-NBA-Ink ; https://www.tcdb.com/ViewSet.cfm/sid/210856/2018-19-Panini-Contenders-Optic-Contenders-Tribute-Autographs-1981 ; https://www.tcdb.com/ViewSet.cfm/sid/210864/2018-19-Panini-Contenders-Optic-Sophomore-Contenders-Autographs ; https://www.checklistcenter.com/2018-19-panini-contenders-optic-basketball/
• G-010 THE SCHEMA CANNOT EXPRESS "SERIAL NUMBERED, RUN VARIES BY CARD", AND THIS RELEASE HAS TWO INSTANCES THAT ARE NOT WRITTEN ANYWHERE STRUCTURED. (1) NBA Ink is provably serial numbered per card - the same sid renders "AU, SN99", "AU, SN49" and "AU, SN25" on different rows - and that is a property of the set, not of a parallel tier, so section 2.2 keeps it out of sets.csv and it is not a parallel, so it stays out of parallels.csv. (2) Legendary Contenders Autographs is stated "#/99 or as Noted", the classic shape of a base figure with per-card exceptions, and the exceptions were not enumerated. Section 2.3 records that the forced pairing of print_run and serial_numbered collapses this state to UNKNOWN/UNKNOWN, which is indistinguishable from having found nothing. Recorded here because the CSV cannot carry it. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/210862/2018-19-Panini-Contenders-Optic-NBA-Ink ; https://www.tcdb.com/ViewSet.cfm/sid/210858/2018-19-Panini-Contenders-Optic-Legendary-Contenders-Autographs ; https://www.checklistcenter.com/2018-19-panini-contenders-optic-basketball/
• G-011 THE INSERTS.CFM NOTES COLUMN ON THIS PARENT IS A FIFTH SHAPE, AND THE SPEC'S FOUR-SHAPE CORRECTION DOES NOT COVER IT. Section 3.2's dated correction lists four behaviours - Prizm (print runs with a split, the only shape where blank means unnumbered), Obsidian (blank on all rows beside provably numbered tiers), Prizm Mosaic (no Notes column at all) and Contenders / Hoops Winter (image-completion percentages). This parent is a fifth: EVERY one of its 85 rows carries an image-completion percentage, AND a small minority of rows additionally carry a real serial statement. Measured on one fetch of Inserts.cfm covering all 75 insert rows and all 10 parallel rows with no truncation: "Front Row Seat Gold (sid: 210876) - Notes: '6.7% of images added' and 'Serial numbered to 10'" and "Front Row Seat Gold Vinyl (sid: 210877) - Notes: '0.0% of images added' and 'Serial numbered to 1'", while the other 83 rows carry a percentage and nothing else. So the column is neither pure percentages nor a Prizm-style split: there are NO blank cells at all, so the blank-annotation test cannot be run in either direction, and a non-blank cell is not a serial statement because 83 of 85 non-blank cells are not one. CONSEQUENCE, MEASURED NOT ASSUMED: the two rows that DO carry a serial statement prove the annotation is real, and the absence of one on the other 83 proves nothing - Class Acts Gold (sid 210871) carries only a percentage and renders no serial token on any probed row, yet checklistcenter states "Gold #/10" for it and the visually identical Front Row Seat Gold renders "SN10". Reading the silence as unnumbered would have been FALSE on Class Acts Gold, which is the Obsidian failure in a new costume. The fallback named in section 3.2 - a within-sid serial-token probe per tier - was used throughout instead. A human should add this fifth shape to the section 3.2 table. Evidence: https://www.tcdb.com/Inserts.cfm/sid/206591/2018-19-Panini-Contenders-Optic ; https://www.tcdb.com/ViewSet.cfm/sid/210876/2018-19-Panini-Contenders-Optic-Front-Row-Seat-Gold ; https://www.tcdb.com/ViewSet.cfm/sid/210871/2018-19-Panini-Contenders-Optic-Class-Acts-Gold ; https://www.checklistcenter.com/2018-19-panini-contenders-optic-basketball/
• G-012 TCDB'S ROW-LEVEL SERIAL ANNOTATION IS INCOMPLETE FOR INSERT-SCOPE TIERS ON THIS PARENT, WHICH LIMITS WHAT THE NO-TOKEN TEST CAN LICENSE. Class Acts Gold (sid 210871) renders no token on rows 1 to 5 while checklistcenter states "Gold #/10" for that set and Front Row Seat Gold (sid 210876) renders "SN10" on rows 1 to 5 of the same parent. Two tiers of the same name and the same nominal run, one annotated and one not. Every Gold row in parallels.csv therefore carries print_run=10 on checklistcenter's statement, with tcdb corroborating at row level only where it annotates (Front Row Seat, NBA Ink, Sophomore Contenders, and the Veteran and block ladders). The two Cracked Ice tiers per insert set are written unnumbered NOT on tcdb silence alone but on a same-set contrast - Front Row Seat Gold renders "SN10" while Front Row Seat Blue Cracked Ice (210878) and Front Row Seat Red Cracked Ice (210879) render no token on the identical first five rows - reinforced by checklistcenter listing Cracked Ice on a separate line with no figure in a ladder where every numbered tier carries one. That is weaker than the within-sid test used for the base-scope Red tier and it is recorded as such. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/210871/2018-19-Panini-Contenders-Optic-Class-Acts-Gold ; https://www.tcdb.com/ViewSet.cfm/sid/210876/2018-19-Panini-Contenders-Optic-Front-Row-Seat-Gold ; https://www.tcdb.com/ViewSet.cfm/sid/210878/2018-19-Panini-Contenders-Optic-Front-Row-Seat-Blue-Cracked-Ice ; https://www.tcdb.com/ViewSet.cfm/sid/210879/2018-19-Panini-Contenders-Optic-Front-Row-Seat-Red-Cracked-Ice ; https://www.tcdb.com/ViewSet.cfm/sid/210873/2018-19-Panini-Contenders-Optic-Class-Acts-Blue-Cracked-Ice ; https://www.checklistcenter.com/2018-19-panini-contenders-optic-basketball/
• G-013 ONE TCDB SID CARRIES THREE SCOPES WITH THREE DIFFERENT PRINT RUNS, AND THE SCHEMA SURVIVES IT ONLY BECAUSE SCOPE IS A COLUMN. Nine of the ten base parallel sids span the 100-card base run AND both halves of the 101-130 autograph block, with a different figure in each. Measured on Red (sid 210837, stated "Total Cards: 146"): rows 1-100 render no token at all, rows 101-130 plain render "AU, SN149" and rows 101-130 variation render "AU, SN99, VAR". Measured on Blue (210838, "156"): no-token base, "AU, SN99" plain block, "AU, SN49, VAR". Three parallels.csv rows are written per such sid, under three scopes. Silver (210836, "100") is the ONE base parallel that does not span the block, which is itself the fourth independent line of evidence that 101-130 is not part of the base run. Anyone reading a sid as one tier with one run would get every one of these wrong. Arithmetic checked by hand rather than taken from the page: Red 100 + 23x2 = 146 and the 23 numbers were counted from the rendered list, Blue 100 + 28x2 = 156 and the 28 numbers were counted from the rendered list, Orange and Gold and Gold Vinyl and three of four plates 100 + 30x2 = 160. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/210837/2018-19-Panini-Contenders-Optic-Red ; https://www.tcdb.com/Checklist.cfm/sid/210837/2018-19-Panini-Contenders-Optic-Red?PageIndex=2 ; https://www.tcdb.com/ViewSet.cfm/sid/210838/2018-19-Panini-Contenders-Optic-Blue ; https://www.tcdb.com/Checklist.cfm/sid/210838/2018-19-Panini-Contenders-Optic-Blue?PageIndex=2 ; https://www.tcdb.com/ViewSet.cfm/sid/210839/2018-19-Panini-Contenders-Optic-Orange ; https://www.tcdb.com/ViewSet.cfm/sid/210836/2018-19-Panini-Contenders-Optic-Silver
• G-014 THE ROOKIE TICKET PARALLEL ROSTERS ARE SHORT OF THE 30-CARD BLOCK, AND THE TWO SOURCES CORROBORATE EACH OTHER AT ROW LEVEL. tcdb's Red sid renders only 23 of the 30 block numbers and its Blue sid only 28. The seven numbers absent from Red are 101 Collin Sexton, 102 Bruce Brown, 105 Jalen Brunson, 107 Josh Okogie, 115 Jaren Jackson Jr., 116 Wendell Carter Jr. and 117 Kevin Huerter, counted off the rendered list by hand; checklistcenter annotates its Rookie Ticket Auto Set ladder "Red #/149 (*No Sexton, Brown, Brunson, Okogie, Jackson, Carter, Huerter)" - seven names, the same seven. The two numbers absent from Blue are 101 Sexton and 115 Jackson; checklistcenter annotates "Blue #/99 (*No Sexton, Jackson)". Seven of seven and two of two. This is recorded rather than acted on because the schema has no per-tier roster: parallels.csv carries the tier and its run, not which cards it covers, so a consumer computing 30 cards times each tier will over-count Red by 7 and Blue by 2. Evidence: https://www.tcdb.com/Checklist.cfm/sid/210837/2018-19-Panini-Contenders-Optic-Red?PageIndex=2 ; https://www.tcdb.com/Checklist.cfm/sid/210838/2018-19-Panini-Contenders-Optic-Blue?PageIndex=2 ; https://www.checklistcenter.com/2018-19-panini-contenders-optic-basketball/
• G-016 NO UNISSUED NUMBERS IN THIS BASE RUN, AND BOTH OF THE FAMILY'S FALSE-POSITIVE TESTS WERE STILL RUN. Cards 1-100 render contiguously with no skipped number, cross-read two ways: Checklist.cfm?PageIndex=1 renders 1 through 74 and Gallery.cfm pages 8, 9 and 10 render 71-80, 81-90 and 91-100, with cards 71, 72, 73 and 74 appearing in both renderings with identical players and teams (71 Danilo Gallinari, 72 Otto Porter Jr., 73 Donovan Mitchell, 74 Serge Ibaka). THE PARALLEL-SET TEST that reversed the 2021-22 card 111 conclusion was run anyway: Silver (sid 210836) states "Total Cards: 100" and Orange, Gold, Gold Vinyl and three of the four Printing Plates state 160 against a 60-row block, all consistent with an unbroken 100-card run, and the individually probed base rows 1, 2, 3, 10, 25, 40, 55, 70 rendered the expected players on five different parallel sids. THE NEXT-SEASON UPDATE TEST that reversed the 2022-23 card 125 conclusion was also run: Inserts.cfm on the 2019-20 Panini Contenders Optic parent (sid 244800) renders "Insert Sets (80)" and "Parallel Sets (12)" and lists NO set whose name contains "2018-19" or "Update". Note this is an absence result on one page of one source and is therefore weaker than the positive finding the sibling Contenders line got - data/contenders-bkb-2018-19 G-004 DID find a 2018-19 Contenders Update I block filed under the 2019-20 Contenders parent - so the two lines of this family behave differently here and the absence should not be read as proof that no such block exists anywhere. Evidence: https://www.tcdb.com/Checklist.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=1 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=8&sort=0 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=9&sort=0 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=10&sort=0 ; https://www.tcdb.com/ViewSet.cfm/sid/210836/2018-19-Panini-Contenders-Optic-Silver ; https://www.tcdb.com/ViewSet.cfm/sid/210839/2018-19-Panini-Contenders-Optic-Orange ; https://www.tcdb.com/Inserts.cfm/sid/244800/2019-20-Panini-Contenders-Optic
• G-017 ROOKIE FLAG IS A WITHIN-SID POSITIVE DISCRIMINATIVE TEST, NOT A DEFAULT. Every base row is written rookie_flag=no because the same renderer that prints the literal token "RC" on all 60 rows of the 101-130 block - as "(AU, RC)" and "(AU, RC, VAR)" - prints no token on any of the 100 base rows, on Checklist.cfm and on Gallery.cfm alike. That is the strongest form of the test named in section 6, run within one sid. The temporal check agrees: the entire 2018 draft class sits in the block (Ayton 111, Doncic 128, Young 124, Bagley III 110, Jaren Jackson Jr. 115, Mikal Bridges 130 and the rest) while the base run is veterans and earlier draft classes - its most recent players are 2017 draftees in their second season (17 Jayson Tatum, 50 Kyle Kuzma, 73 Donovan Mitchell, 81 De'Aaron Fox, 90 Lauri Markkanen), none of whom could carry an RC designation in 2018-19. It rests on tcdb alone. Evidence: https://www.tcdb.com/Checklist.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=1 ; https://www.tcdb.com/Checklist.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=2 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=11&sort=0
• G-018 SECTION 5.2 BASE-CHECKLIST-SHARING CHECK RUN AND NEGATIVE, AND THE TWO OVERLAP MEASURES ARE REPORTED SEPARATELY ON PURPOSE. Of this release's 100 (card_number, player_name) pairs, exactly ONE coincides positionally with data/contenders-bkb-2018-19/base_cards.csv: card 50, Kyle Kuzma, Los Angeles Lakers, in both. One in a hundred is chance level and is not a shared checklist. The line's positional overlap against Contenders now reads 1/100 for 2018-19, 1/100 for 2019-20, 78/100 for 2020-21, 0/100 for 2021-22, 1/100 for 2022-23 and 0/100 for 2023-24 - 2018-19 sits with the five low seasons and the 2020-21 figure remains the outlier. SEPARATELY, and NOT the same measurement: roster overlap ignoring slot is 72 of 100 player names shared with Contenders 2018-19, which is a shared league-wide player pool in a 30-team league and not a shared checklist. Against data/contendersoptic-bkb-2019-20 the figures are 1/100 positional (card 20, Khris Middleton) and 69/100 by roster; against data/optic-bkb-2019-20 they are 0/200 positional and 86 shared names. Conflating the two measures would read as a section 5.2 collision when it is not one. The two products also differ on every structural axis: separate parent sids (206591 against 188543), separate release dates (2019-07-31 against 2018-12-26), separate box configurations (1 pack of 6 against 12 packs of 10), separate block sizes (30 rookies against 45) and separate set inventories. Evidence: https://www.tcdb.com/ViewAll.cfm/sp/Basketball/year/2018 ; https://www.tcdb.com/ViewSet.cfm/sid/206591/2018-19-Panini-Contenders-Optic ; https://www.tcdb.com/Checklist.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=1 ; https://www.cardboardconnection.com/2018-19-panini-contenders-optic-basketball-cards
• G-019 CONTENDERS OPTIC PREVIEW IS A DIFFERENT OBJECT FROM THIS RELEASE AND THE DISTINCTION IS EASY TO LOSE. "Contenders Optic Preview" is a 100-card INSERT SET INSIDE 2018-19 Panini Contenders - tcdb sid 194856, page title "2018-19 Panini Contenders - Contenders Optic Preview Basketball Checklist", "Total Cards: 100" - and it is already catalogued as an insert row of data/contenders-bkb-2018-19, whose own copy check confirmed it is not that release's base run renamed. THIS release is the standalone product 2018-19 Panini Contenders Optic at its own parent sid 206591 on the 2018 season index. The two are the same SIZE, 100 cards, which is precisely what makes the confusion available, and they are not the same checklist: the Preview renders 1 Mikal Bridges, 2 Jaylen Brown, 3 Chandler Hutchison, 10 Chris Paul, 25 JR Smith, 40 Khris Middleton, 55 Marc Gasol, 70 James Harden against this product's 1 Brandon Ingram, 2 Lonzo Ball, 3 DeMar DeRozan, 10 Kevin Love, 25 Eric Bledsoe, 40 Kristaps Porzingis, 55 Giannis Antetokounmpo, 70 Kyrie Irving. Zero positional coincidences at eight probed slots. Note also that three of the Preview's opening names - Mikal Bridges, Chandler Hutchison and Jaylen Brown - are rookies or sophomores who appear in THIS product only in the 101-130 autograph block or not at all, so the two rosters are built on different principles. NO insert_cards.csv was written for either, and a full row-by-row copy check of the two 100-card lists was NOT run - eight slots is a probe, not a proof. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/194856/2018-19-Panini-Contenders-Contenders-Optic-Preview ; https://www.tcdb.com/Checklist.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=1 ; https://www.tcdb.com/ViewAll.cfm/sp/Basketball/year/2018
• G-020 SCOPE_TYPE IS set AND NOT block ON ALL 75 NON-BASE ROWS, DELIBERATELY. Section 2.3 records that a block row resolves against base_card_tiers, that a release with no tier rows gives every block row scope_set_id NULL, and that parallel_unique_in_scope then keys on the name alone - and that data/contenders-bkb-2018-19 wrote 12 block rows naming two autograph blocks and had all six colliding pairs rejected before being re-scoped to set. This release has no base_card_tiers.csv, and its Rookie Ticket and Rookie Ticket Variation ladders share five names each (Red, Blue, Orange, Gold, Gold Vinyl) plus four plate names, with Red and Blue carrying genuinely different runs across the two (149 against 99, and 99 against 49). Under scope_type=block that is nine colliding pairs and a rejected load. Both blocks ARE rows in sets.csv, so scope_type=set resolves each to a distinct set id, which is both legal and more accurate. Recorded because a checker comparing scope STRINGS would see no difference between the two choices. Evidence: https://www.tcdb.com/Checklist.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=2 ; https://www.tcdb.com/Checklist.cfm/sid/210837/2018-19-Panini-Contenders-Optic-Red?PageIndex=2 ; https://www.tcdb.com/Checklist.cfm/sid/210838/2018-19-Panini-Contenders-Optic-Blue?PageIndex=2
• G-021 PRINTING PLATE YELLOW IS ONE ROW SHORT OF ITS OWN SIBLINGS ON THE SAME PARENT. The base parallel Printing Plate Black, Cyan and Magenta subsets each state "Total Cards: 160" while Printing Plate Yellow states "Total Cards: 159". A one-row shortfall against three identically scoped siblings is incomplete cataloguing on tcdb's side rather than a real difference in coverage; which row is absent was not established, because doing so would cost a full enumeration of a 159-row set for a fact that changes no written value. It does not affect the written print run, which is the literal token "SN1" read off the rows themselves, and it does mean no plate subset can be used as an independent count of the parent's rendered rows. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/210842/2018-19-Panini-Contenders-Optic-Printing-Plate-Black ; https://www.tcdb.com/ViewSet.cfm/sid/210844/2018-19-Panini-Contenders-Optic-Printing-Plate-Cyan ; https://www.tcdb.com/ViewSet.cfm/sid/210845/2018-19-Panini-Contenders-Optic-Printing-Plate-Magenta ; https://www.tcdb.com/ViewSet.cfm/sid/210843/2018-19-Panini-Contenders-Optic-Printing-Plate-Yellow
• G-022 THIRTY-ONE PRINT RUNS REST ON CHECKLISTCENTER ALONE FOR THE FIGURE, WITH TCDB SOURCING ONLY THE TIER'S EXISTENCE - the rows carrying source_confidence=single. Enumerated and counted by hand rather than tallied from anything: the Gold and Gold Vinyl rows of Legendary Contenders Autographs and of Up and Coming Contenders Autographs (4); the Gold Vinyl rows of NBA Ink and of Sophomore Contenders Autographs, whose Gold rows ARE corroborated at row level (2); the Gold and Gold Vinyl rows of eight of the nine Cracked Ice insert sets - Class Acts, Hall of Fame Contenders, Historic MVPs, Historic Rookie of the Year, Lottery Ticket, Playing the Numbers Game, Superstar Die Cuts and Winning Tickets, Front Row Seat being the ninth and corroborated (16); and the Cyan, Magenta and Yellow Printing Plate rows of Rookie Ticket, Rookie Ticket Variation and Veteran Season Ticket Autographs, whose Black plate IS corroborated (9). 4 + 2 + 16 + 9 = 31. In every case tcdb lists the tier as its own sid under the parent's Insert Sets heading, so the tier certainly exists; what tcdb does not render is a serial token on its rows, and gap G-012 shows that silence is not informative on this parent. THE SOURCE WAS CALIBRATED BEFORE BEING TRUSTED, on questions that could be checked - every checklistcenter figure that tcdb also annotates at row level was compared, and every one matched. The list, counted by hand: base ladder Blue 99, Orange 49, Gold 10, Gold Vinyl 1 and Printing Plate Black, Cyan, Magenta and Yellow at 1 (8); Rookie Ticket Red 149, Blue 99, Orange 25, Gold 10, Gold Vinyl 1, Printing Plate Black 1 (6); Rookie Ticket Variation Red 99, Blue 49, Orange 25, Gold 10, Gold Vinyl 1, Printing Plate Black 1 (6); Veteran Season Ticket Autographs Red 49, Blue 35, Orange 25, Gold 10, Gold Vinyl 1, Printing Plate Black 1 (6); Front Row Seat Gold 10, NBA Ink Gold 10, Sophomore Contenders Autographs Gold 10 (3). 8 + 6 + 6 + 6 + 3 = 29 figures, 29 matches, 0 misses. Its roster exclusions match tcdb row for row as well - seven of seven on Rookie Ticket Red, two of two on Rookie Ticket Blue, and the Veteran ladder's stated totals of 28, 29 and 29 against Gold 30 land exactly where its parentheticals say. cardboardconnection independently reproduces the base ladder verbatim. ⚠ SECTION 5.4 CAVEAT STATED AS THE SPEC REQUIRES: cardboardconnection and checklistcenter may be ONE opinion counted twice on the base ladder, since they render it in the same order with the same figures; with Beckett unreachable the usual third structural source is gone. The base ladder is nonetheless independently confirmed by tcdb row-level tokens, so nothing load-bearing rests on that pair alone. Evidence: https://www.checklistcenter.com/2018-19-panini-contenders-optic-basketball/ ; https://www.cardboardconnection.com/2018-19-panini-contenders-optic-basketball-cards ; https://www.tcdb.com/Inserts.cfm/sid/206591/2018-19-Panini-Contenders-Optic ; https://www.tcdb.com/ViewSet.cfm/sid/210876/2018-19-Panini-Contenders-Optic-Front-Row-Seat-Gold
• G-023 THE FETCH LAYER INVENTED GLOSSES FOR FLAG TOKENS ON THIS PASS, AS SECTION 3.3 WARNS, AND NONE REACHED THE DATA. Three instances were caught: a Gallery fetch returned the tokens plus the expansion "Each numbered entry appears twice - once as a base version and once as a variant (VAR)"; a Printing Plate fetch returned "The flag tokens appearing after player names indicate autographed (AU), serial numbered (#SN1), and variant (VAR) designations"; a Gallery fetch of the base run volunteered "indicating missing front/back images in the database". Only the literal tokens actually observed - AU, RC, VAR, EXCH, SN1, SN10, SN25, SN35, SN49, SN99, SN149 - are recorded anywhere in this release, and the word "variation" appears in set names because checklistcenter prints it ("Rookie Ticket Variation Auto Set"), not because a fetch glossed VAR. Evidence: https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=11&sort=0 ; https://www.tcdb.com/Checklist.cfm/sid/210842/2018-19-Panini-Contenders-Optic-Printing-Plate-Black?PageIndex=2 ; https://www.tcdb.com/Gallery.cfm/sid/206591/2018-19-Panini-Contenders-Optic?PageIndex=10&sort=0
• G-024 A METHOD FINDING THAT CUTS AGAINST SECTION 3.2's CACHE-BUSTING ADVICE ON ONE HOST. The spec instructs varying a URL with a harmless extra parameter to bust the 15-minute per-URL error cache, because two identical fetches are one observation. On tcdb that worked exactly as documented and was used throughout (&sort=0 and ?sort=0 on Gallery, Checklist and ViewSet). On cardboardconnection it BROKE THE FETCH: the plain product URL returned content, and the identical URL with ?x=1 appended returned a ROBOTS_DISALLOWED error reading "robots.txt fetch failed: Name or service not known". So the cache-busting parameter is not harmless on every host, and an agent that reaches for it by reflex will read a robots refusal as evidence that a page does not exist. Note also that cardboardconnection WAS reachable on this pass, which contradicts data/contenders-bkb-2018-19 G-014 recording that it "refused this environment with a robots.txt error on every attempt" two days ago, and contradicts the scope memo's section 5 note that its dates could not be re-verified. The host is intermittent, not blocked. Evidence: https://www.cardboardconnection.com/2018-19-panini-contenders-optic-basketball-cards
• G-025 SUSPICIOUS-UNIFORMITY CHECK RUN, AND THE COINCIDENCE IS REPORTED RATHER THAN SMOOTHED. This release writes exactly 85 parallel rows, and data/contendersoptic-bkb-2019-20/parallels.csv also has exactly 85. Section 4.5 calls identical structure across releases a template-fill smell, so the composition was compared rather than the total: the 2019-20 file is 12 base-scope rows including a China-exclusive Gold Wave and Red Wave pair and 73 set-scope rows across 19 sets, while this file is 10 base-scope rows with no Wave tier of any kind and 75 set-scope rows across 20 scopes, including four Contenders Tribute year subsets and a nine-plate ladder that the next season does not have. The tier ladders differ set by set (this season gives Veteran Season Ticket Autographs a nine-row ladder with Red #/49 and Blue #/35; the next season gives Veteran Ticket Red #/25 and Blue #/20). Different composition, coincident total. Recorded so a reviewer who notices the equality does not have to re-derive it. Evidence: https://www.tcdb.com/Inserts.cfm/sid/206591/2018-19-Panini-Contenders-Optic ; https://www.checklistcenter.com/2018-19-panini-contenders-optic-basketball/
• G-026 INSERT AND AUTOGRAPH CARD-LEVEL CHECKLISTS ARE OUT OF SCOPE FOR THIS PASS and no insert_cards.csv was written, per the spec's opening paragraph and CHASE_CHECKLIST_SPEC. Prefix inventory measured here for that pass, so it does not have to be rediscovered: Veteran Season Ticket Autographs VT- ; Legendary Contenders Autographs LC- ; NBA Ink NI- ; Sophomore Contenders Autographs SC- ; Up and Coming Contenders Autographs UC- ; Contenders Tribute Autographs 1981 81- , 1998 98- , 2003 03- , 2007 07-. The nine insert sets and Class Acts use plain 1..N numbering. TWO TRAPS FOR THAT PASS: (1) the Contenders Tribute 1981 set prints "81-MB3" for Marvin Bagley III and the base ladder's siblings in this season print "SS-LW4" for Lonnie Walker IV in other products - Roman-numeral name suffixes encoded as digits, which look like typos and must not be normalized; (2) tcdb renders TWO DIFFERENT PLAYERS under the code "NI-DMC" in NBA Ink, DeMarcus Cousins and DeMarre Carroll, both with "AU" - either tcdb has a duplicate-code error or the printed codes genuinely collide, and an insert_cards pass keyed on the code alone will lose one of them. Evidence: https://www.tcdb.com/ViewSet.cfm/sid/210862/2018-19-Panini-Contenders-Optic-NBA-Ink ; https://www.tcdb.com/ViewSet.cfm/sid/210856/2018-19-Panini-Contenders-Optic-Contenders-Tribute-Autographs-1981 ; https://www.tcdb.com/ViewSet.cfm/sid/210846/2018-19-Panini-Contenders-Optic-Veteran-Season-Ticket-Autographs ; https://www.tcdb.com/ViewSet.cfm/sid/210858/2018-19-Panini-Contenders-Optic-Legendary-Contenders-Autographs ; https://www.tcdb.com/ViewSet.cfm/sid/210867/2018-19-Panini-Contenders-Optic-Up-and-Coming-Contenders-Autographs ; https://www.tcdb.com/ViewSet.cfm/sid/210864/2018-19-Panini-Contenders-Optic-Sophomore-Contenders-Autographs
• G-027 THE 2018-19 SID LAYOUT FOR THIS PRODUCT IS NOT CONTIGUOUS WITH ITS PARENT, AND INTERLEAVING WAS TREATED AS THE DEFAULT AS THE MEMO'S TWICE-CORRECTED SECTION 2 NOW REQUIRES. The parent sid is 206591 while every one of its 85 subsets sits in two distant bands, 210836-210914 and 211140-211145 - roughly four thousand sids away from the parent, and split. No sid range was enumerated by extrapolation in this build and no boundary claim is made: the full subset inventory came from ONE Inserts.cfm fetch on the parent, which returned all 75 insert rows and all 10 parallel rows with no truncation, so no sid hunting was necessary. That is itself a finding against section 3.2, which records that Inserts.cfm "does not paginate, and on a large parent it truncates before the Parallel Sets heading - three consecutive seasons on the Optic parent". This Contenders Optic parent did NOT truncate at 85 rows, which is consistent with the same section's later note that truncation is parent-dependent rather than size-dependent. Evidence: https://www.tcdb.com/Inserts.cfm/sid/206591/2018-19-Panini-Contenders-Optic ; https://www.tcdb.com/ViewSet.cfm/sid/206591/2018-19-Panini-Contenders-Optic

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