2018-19 Prizm

• G-001 Channel unresolved, channel=UNKNOWN with an empty channel_basis, for 15 base-set parallels: Silver, Hyper, Ruby Wave, Red White and Blue, Red /299, Blue /199, Purple Ice /149, Blue Ice /99, Purple /75, Orange /49, Purple Pulsar /35, Mojo /25, Gold /10, Black Gold /5, Black 1/1. cardboardconnection lists these tiers with a print run but no format annotation, while annotating eight sibling tiers with one, and tcdb states no channel for any of them. Sources: https://www.cardboardconnection.com/2018-19-panini-prizm-basketball-nba-cards ; https://www.tcdb.com/Inserts.cfm/sid/188541/2018-19-Panini-Prizm
• G-002 The channel enum cannot express two box SKUs that cardboardconnection states for this release. Red White and Blue is rendered "Red/White/Blue (Multi-Pack)" and Purple Pulsar is rendered "Purple Pulsar #/35 (Gravity Feed)". channel_taxonomy.csv already records cello as a box SKU with no enum value, and gravity feed is not in the file at all. Both were written as channel=UNKNOWN with the SKU in channel_sku rather than forced to retail, because retail would be our inference and not the source's statement. DIVERGENCE: prizm-bkb-2019-20 writes Red White and Blue Prizm as channel=retail with channel_basis=unsupported, and prizm-bkb-2020-21, 2021-22 and 2022-23 do the same. Same evidence shape, opposite value. Source: https://www.cardboardconnection.com/2018-19-panini-prizm-basketball-nba-cards
• G-003 DIVERGENCE on Silver Prizm's channel. This build writes channel=UNKNOWN with an empty channel_basis because no 2018-19 source states a format for the tier. prizm-bkb-2019-20, prizm-bkb-2020-21, prizm-bkb-2021-22 and prizm-bkb-2022-23 all write channel=multi with channel_basis=unsupported for the same tier of the same line. Four seasons say multi and this one says UNKNOWN, and nothing on either row tells a consumer which convention it is under. Recorded rather than smoothed over; a repair pass should settle the line.
• G-004 Channel axis conflict on the five Choice tiers. cardboardconnection heads the group "CHOICE PRIZMS (International)", which asserts a REGION, while the tier name asserts a BOX SKU. channel_taxonomy.csv records choice as a box-sku and international as a region, they are orthogonal axes, and the column holds one value. channel=choice with channel_basis=from-name was written and the international annotation is preserved in visual_notes. The regional claim is unrecorded in structured form. Source: https://www.cardboardconnection.com/2018-19-panini-prizm-basketball-nba-cards
• G-005 Sensational Swatches Prime carries the tcdb Notes annotation "Serial numbered to 10, 5 or 3". The tier is provably serial numbered with a per-card variable run, but the schema forces print_run and serial_numbered to move together and has no way to say serial_numbered=yes with print_run=UNKNOWN, so both are UNKNOWN. The fact held and lost is: this tier IS numbered, to one of 10, 5 or 3 depending on the card. This is the same expressiveness gap the extraction spec records as recurring in four consecutive seasons of another line. Source: https://www.tcdb.com/Inserts.cfm/sid/188541/2018-19-Panini-Prizm
• G-006 Base-set scope confirmed by direct probe for only 3 of the 37 base parallels. Prizms Red, Prizms Silver and Prizms White Sparkle each render "Total Cards: 300", matching the base run exactly. The other 34 are taken as base-wide from their placement under the parent's Parallel Sets heading and from cardboardconnection listing the same 37 names under its Prizm parallel headings. No tier was found whose lowest rendered card number is above 1, and the release has no numbered block past 300, so a block parallel is not possible here. Probes: https://www.tcdb.com/ViewSet.cfm/sid/191002/2018-19-Panini-Prizm-Prizms-Red ; https://www.tcdb.com/ViewSet.cfm/sid/190991/2018-19-Panini-Prizm-Prizms-Silver ; https://www.tcdb.com/ViewSet.cfm/sid/204721/2018-19-Panini-Prizm-Prizms-White-Sparkle
• G-007 The line's naming convention masks a genuine cross-scope name collision, so scope_ambiguous understates it. Base-set tiers are written with the suffix Prizm, for example Silver Prizm, while set-scope tiers are written bare, for example Silver scoped to Dominance. The literal strings therefore never collide and scope_ambiguous computes to no on all 37 base rows, yet both sources call both tiers "Silver" and a consumer matching on what sources print WILL collide them. The convention is inherited from prizm-bkb-2019-20 through 2022-23 and was kept so the line stays joinable across seasons. Fixing it is a catalogue-wide decision, not a per-release one.
• G-008 Emergent and Freshman Phenoms render the SAME 25-player rookie roster in exactly reverse order. Emergent runs Deandre Ayton, Marvin Bagley III, Luka Doncic ... Anfernee Simons, Moritz Wagner; Freshman Phenoms runs Moritz Wagner, Anfernee Simons ... Marvin Bagley III, Deandre Ayton. Two distinct sids, two distinct names, and cardboardconnection independently lists both sets at 25 cards, so both sets are real and the copy check is not a deletion signal. What is NOT established is whether the rosters are genuinely identical or whether one fetch rendered a differently sorted view. The card-level checklist pass must re-derive both independently rather than copying one from the other. Sources: https://www.tcdb.com/ViewSet.cfm/sid/191171/2018-19-Panini-Prizm-Emergent ; https://www.tcdb.com/ViewSet.cfm/sid/191162/2018-19-Panini-Prizm-Freshman-Phenoms
• G-009 Set name spelling divergence. tcdb renders "Fastbreak Autographs" and "Fastbreak Autographs Rookies" as one word; cardboardconnection renders "Fast Break Autographs"; prizm-bkb-2019-20 carries "Fast Break Autographs" and "Fast Break Rookie Autographs". tcdb's spelling is written here because tcdb is the enumerating source for the set inventory, but the sibling season is not joinable on set_name without an alias. Sources: https://www.tcdb.com/Inserts.cfm/sid/188541/2018-19-Panini-Prizm ; https://www.cardboardconnection.com/2018-19-panini-prizm-basketball-nba-cards
• G-010 Sensational Swatches is a memorabilia set - every rendered row carries the literal token MEM and no row carries AU - and is filed as set_type=insert because the enum has no relic value and an unsigned relic set is not auto. Recorded so a consumer is not misled into treating it as a plain insert. Source: https://www.tcdb.com/ViewSet.cfm/sid/191393/2018-19-Panini-Prizm-Sensational-Swatches
• G-011 Luck of the Lottery row 15 of 15 is rendered "Class of 2018" rather than a player name. The count of 15 is written as tcdb and cardboardconnection both state it, but 15 distinct cards is not 15 distinct players. Source: https://www.tcdb.com/ViewSet.cfm/sid/191147/2018-19-Panini-Prizm-Luck-of-the-Lottery
• G-012 Card counts for six of the seventeen sets rest on a single source, the tcdb Total Cards line, with no second source stating a figure: Fastbreak Autographs 60, Fastbreak Autographs Rookies 40, Rookie Signatures 37, Sensational Signatures 95, Sensational Swatches 100, Signatures 55. cardboardconnection names four of these sets but states no count for any of them. In each case the rendered rows were not counted independently, so the Total Cards header is unchecked against the body - the extraction spec warns that the header counts different things on different lines.
• G-013 Card counts for the 63 set-scope parallel tiers were not sourced. Only the 16 parent sets were fetched for a Total Cards line; Inserts.cfm carries no card-count column, so each tier would have cost its own fetch. The tiers' existence and print runs ARE sourced; their sizes are not.
• G-014 No source found states a licensing grantor for this product. license_status=licensed (NBA), licensed=yes and league=NBA are written to match the catalogue-wide value and are supported only by the product being an NBA-team, NBA-player checklist of 300 current and retired NBA players on NBA franchises. This is the same absence recorded catalogue-wide, not a gap specific to 2018-19. Sources: https://www.tcdb.com/Checklist.cfm/sid/188541/2018-19-Panini-Prizm ; https://www.cardboardconnection.com/2018-19-panini-prizm-basketball-nba-cards
• G-015 base_print_run=unnumbered carries base_print_run_basis=assumed, not sourced. The evidence is strong indirect evidence and not a statement: the tcdb renderer prints the literal token SN299 on every row of Prizms Red and prints no serial token on any of the 300 base rows across three Checklist pages and thirty Gallery pages. No source says in words that the base cards are unnumbered. Sources: https://www.tcdb.com/ViewSet.cfm/sid/191002/2018-19-Panini-Prizm-Prizms-Red ; https://www.tcdb.com/Checklist.cfm/sid/188541/2018-19-Panini-Prizm
• G-016 DIVERGENCE in print-run coverage against the sibling season. All 80 set-scope parallel rows in prizm-bkb-2019-20 carry print_run=UNKNOWN. This build writes real integers on 32 of its 63 set-scope rows, unnumbered on 30 and UNKNOWN on 1, because the tcdb Inserts.cfm listing renders a per-set "Serial numbered to N" annotation that the 2019-20 build did not use. The 2019-20 rows are probably recoverable the same way and are worth a repair pass; they are not evidence that the figures do not exist. Source: https://www.tcdb.com/Inserts.cfm/sid/188541/2018-19-Panini-Prizm
• G-017 rookie_flag rests on a single structural source. 53 of the 300 base rows carry a tcdb RC designation and 247 do not; no second source was consulted for rookie status. The no values are written rather than UNKNOWN because the same renderer marks RC on sibling rows of the same page, which is a positive discriminative test, but the whole column is one source deep. Sources: https://www.tcdb.com/Checklist.cfm/sid/188541/2018-19-Panini-Prizm ; https://www.tcdb.com/Gallery.cfm/sid/188541/2018-19-Panini-Prizm
• G-018 Base checklist assembled from two endpoints with a measured delivery truncation, so the joins matter. Checklist.cfm?PageIndex=1 delivered through row 78 of 100, PageIndex=2 through row 71 of 100 and PageIndex=3 through row 72 of 100; the tails came from Gallery.cfm pages 8, 9, 10, 18, 19, 20, 28, 29 and 30. The overlapping rows 71-78, 171 and 271-272 were rendered by both endpoints and agree on number, player and team, which is the only cross-endpoint check available. Rows outside those overlaps rest on one rendering each, EXCEPT rows 291-300, which were re-rendered from a different sid entirely - Gallery.cfm on the Prizms Silver parallel sid 190991 page 30 - and agree with the parent on number, player and team for all ten. That cross-sid probe also showed tcdb rendering NO RC token on card 299 Elie Okobo while the parent renders one, so the RC flag is a property of the parent rendering and a parallel sid must not be used to source rookie_flag. Sources: https://www.tcdb.com/Checklist.cfm/sid/188541/2018-19-Panini-Prizm ; https://www.tcdb.com/Gallery.cfm/sid/188541/2018-19-Panini-Prizm ; https://www.tcdb.com/Gallery.cfm/sid/190991/2018-19-Panini-Prizm-Prizms-Silver?PageIndex=30
• G-019 No cross-filed or hidden Prizm set was found, and the search was not exhaustive. Checks run: ViewAll for basketball year 2018 lists exactly two Prizm-named products, the parent 188541 and Mosaic Prizm 204836; ViewAll for basketball year 2019 lists no set labelled 2018-19, so there is no next-season Update block holding 2018-19 Prizm cards; Checklist.cfm?PageIndex=4 renders no rows and Gallery page 30 is the last, so nothing is numbered past 300. What was NOT done is a probe of every sid gap inside the parallel block: 190991-191030 is interleaved with unrelated products, and of the six gaps only 190992, 190999 and 191017 were probed. 190992 and 190999 are Magic the Gathering sets and 191017 does not exist. Sources: https://www.tcdb.com/ViewAll.cfm/sp/Basketball/year/2018 ; https://www.tcdb.com/ViewAll.cfm/sp/Basketball/year/2019
• G-020 Release date 2018-11-21 is two-sourced and neither source labels it an estimate: tcdb renders "Release Date: November 21, 2018" and cardboardconnection renders "11/21/2018". Beckett was not consulted - the extraction brief records it as unreachable for six consecutive passes - so the third structural source used for the 2019-20 sibling is absent here and the date rests on two sources rather than three. Sources: https://www.tcdb.com/ViewSet.cfm/sid/188541/2018-19-Panini-Prizm ; https://www.cardboardconnection.com/2018-19-panini-prizm-basketball-nba-cards
• G-021 Possible single-opinion corroboration. cardboardconnection and tcdb agree 37 of 37 on the base parallel roster and 25 of 25 on the stated print runs, which is the strongest agreement in this build and carries most of parallels.csv. The two are structurally independent - one is an editorial product page, the other a user-maintained database - but no third source was reached to test whether either copies the other. checklistcenter, checklistinsider and Beckett were not consulted for this release.
• G-022 Brand identity checked and clean, recorded because the check is required. The product's official name is "2018-19 Panini Prizm Basketball" per cardboardconnection; tcdb renders the set title as "2018-19 Panini Prizm", dropping the sport word as it does for every basketball product. reference/brands.csv carries prizm to "Panini Prizm Basketball", which is the season-stripped form of the cardboardconnection name, so product_name was written byte for byte from brands.csv and nothing was minted. Distinct from the unlicensed prizm-bkb-2025-26 and from the collegiate prizmdp line; the 300-card checklist is entirely NBA franchises. Also distinct from Prizm Mosaic, which tcdb files separately at sid 204836. Sources: https://www.cardboardconnection.com/2018-19-panini-prizm-basketball-nba-cards ; https://www.tcdb.com/ViewSet.cfm/sid/188541/2018-19-Panini-Prizm ; https://www.tcdb.com/ViewAll.cfm/sp/Basketball/year/2018

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