2016-17 Court Kings

• G-001 ROOKIE FLAG IS UNKNOWN ON 199 OF 200 BASE ROWS, AND THAT DIVERGES FROM THE SIBLING SEASON. tcdb renders the token "RC" on exactly ONE row of this 200-card set - card 118, DeAndre Bembry, rendered "118, DeAndre' Bembry, Atlanta Hawks, RC, Rookie I" on https://www.tcdb.com/Checklist.cfm/sid/136900/2016-17-Panini-Court-Kings?PageIndex=2 - and on no other row, including the other 99 rows of the four Rookie tiers and the three further DeAndre Bembry cards at 143, 168 and 193. Confirmed on a second differently-varied endpoint: https://www.tcdb.com/Gallery.cfm/sid/136900/2016-17-Panini-Court-Kings?PageIndex=12 renders tiles 111-120 with "Rookie I" on all ten and "RC" on 118 alone. So the renderer CAN emit RC and emits it once, which means its silence on the other 199 rows is not a statement that those cards are not rookie cards, and EXTRACTION_SPEC 2.4 permits yes/no only where a source explicitly shows an RC designation. rookie_flag is therefore yes on card 118 and UNKNOWN on the other 199. courtkings-bkb-2017-18 wrote rookie_flag=yes on all 132 of its rookie-tier rows and no on its 100 veteran rows, because on that parent tcdb renders "RC" on every rookie-tier row; the evidence differs, the convention differs, and a consumer comparing the two seasons will see a hole here where there is data there. No second source was found that prints RC designations for this release.
• G-002 BASE PRINT RUN CONVENTION DIVERGES FROM SIX OF THE EIGHT BUILT COURT KINGS SEASONS, DELIBERATELY AND FLAGGED. This release records base_print_run=unnumbered with base_print_run_basis=assumed, matching courtkings-bkb-2017-18 and diverging from courtkings-bkb-2018-19, 2019-20, 2020-21, 2021-22, 2022-23, 2023-24 and 2024-25, all seven of which record UNKNOWN/UNKNOWN. The evidence here is the EXTRACTION_SPEC 2.1 worked example exactly: the parent renders no serial token on any of its 200 rows (Checklist.cfm pages 1 and 2, Gallery.cfm pages 8, 9, 10, 12, 18, 19, 20) while sid 141377 renders "SN25" on all 100 of its rows and sid 141376 renders "SN1" on all 100 of its rows. The spec calls that shape assumed, not sourced, and that is what is written. Note caveats CAV-024, which records that two Court Kings seasons were demoted FROM unnumbered TO UNKNOWN for comparability - so the divergence is a live convention question across this line, not a defect of this release.
• G-003 THE TIER FILE AND THE RELEASE ROW DISAGREE BY DESIGN, AND ONE TIER NAME IS WEAKER THAN THE OTHER FOUR. release.csv records base_print_run=unnumbered/assumed while all five rows of base_card_tiers.csv record print_run=UNKNOWN and serial_numbered=UNKNOWN, because base_card_tiers.csv has no basis column: writing unnumbered there would assert an inference as a sourced fact with no way to label it, and COURT_KINGS_TIER_ADDENDUM requires a short verbatim snippet proving a tier print run, which no source provides for any of the five. Consistent with caveats CAV-024. SEPARATELY AND MORE IMPORTANT: the four rookie tier names are printed by tcdb inline on every row ("Rookie I" ... "Rookie IV") and are fully sourced, but tcdb prints NO label on any row of 1-100. The name "Base Set" for that tier rests on cardboardconnection prose alone - "Base set is joined by multiple tiers of rookie cards, each with a different design." (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball) - normalised in capitalisation to match this release's sets.csv base row. Its BOUNDARY is independently evidenced: both tcdb base parallels state "Total Cards: 100" and render cards 1-100 only. A reviewer who judges that prose insufficient should delete the whole tier file rather than rename this row, because the five tiers tile 1-200 exactly and removing one breaks the partition. Note also that the built Court Kings seasons do not agree with each other on this name: 2017-18 and 2018-19 write "Base Set", 2019-20 writes "Base Set", 2020-21 through 2023-24 write "Base", and 2024-25 writes "Base (Veterans)".
• G-004 SEVEN TIERS HAVE A PER-CARD VARIABLE PRINT RUN, WHICH THE SCHEMA CANNOT EXPRESS. Censuses taken row by row over every rendered row, never sampled. FIVE are parallel rows, written print_run=UNKNOWN / serial_numbered=UNKNOWN even though every row is PROVABLY serial numbered: Art Nouveau Jumbo Nameplate (sid 141406, 38 rows, distinct tokens MEM plus SN4, SN5, SN6, SN7, SN8, SN9, SN10, SN11, SN14 - nine distinct serial values); Artistic Endeavors Prime (141411, 27 rows, MEM plus SN25, SN17, SN15, SN3, SN2); Vintage Materials Prime (141397, 16 rows, MEM plus SN25, SN10, SN6, SN5, SN4); Sketches and Swatches Prime (141400, 31 rows, AU, MEM plus SN25, SN10, SN8, SN5); Expressionists Memorabilia Prime (141423, 12 rows, MEM plus SN25, SN5); and 2 on 2 Quad Memorabilia Prime (141417, 8 rows, MEM plus SN25, SN10) - six parallel rows in total. TWO MORE ARE WHOLE SETS, and sets.csv has no print-run column at all, so the finding lives only here and in their notes: Sketches and Swatches (141398) renders ten distinct serial tokens across its rows - SN16, SN26, SN35, SN55, SN60, SN65, SN75, SN125, SN149, SN199; Vintage Materials (141395) renders SN149 on 17 rows, SN75 on 1 and SN49 on 2; Artistic Endeavors (141410) renders SN149, SN85 and SN49; and Le Cinque Piu Belle (141380) renders a different token on every one of its five rows - SN2, SN23, SN3, SN10, SN41 - which cardboardconnection independently corroborates as #/2, #/23, #/3, #/10 and #/41. THE STRUCTURAL RESULT IS WORTH RECORDING: the standing generalisation that variability concentrates in the base tiers of AUTOGRAPH sets does not hold on this release. Its four pure autograph sets - Heir Apparent (all SN150), Fresh Paint (no serial token), Fresh Paint Variation (all SN200), Fresh Paint Duals (all SN50) - are FLAT, while the variability sits in MEMORABILIA tiers and in the Prime patch parallels. The one variable autograph set, Sketches and Swatches, is also a relic set. Nothing at the 1/1 end varies: all eight Masterpiece tiers are uniform SN1. See caveats CAV-041 for the expressiveness gap; none of these is base-scope, so all are at least legal as UNKNOWN rows.
• G-005 PORTRAITS COUNT DISAGREEMENT RESOLVED AS A UNIT QUESTION, RECORDED RATHER THAN SMOOTHED. cardboardconnection states "100 cards. Serial Numbered #/175." for a single Portraits entry (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball). tcdb gives TWO sids: Portraits states "Total Cards: 70" and renders 70 rows numbered 1-70 all SN175, and Rookie Portraits states "Total Cards: 30" and renders 30 rows numbered 1-30 all SN175. 70 + 30 = 100, so the two sources decompose the same number consistently and the disagreement is about the UNIT, not about coverage. The per-sid figures are written. THE SAME SHAPE WAS RECORDED ON courtkings-bkb-2017-18 as its G-005, with the identical 70/30 split and the identical cardboardconnection figure of 100 - so this is a stable property of how the two sources file this set, across two seasons. CONSEQUENCE FOR THE PARALLEL ROWS: cardboardconnection's "PARALLEL CARDS: Ruby #/75, Sapphire #/25, Emerald #/10, Masterpiece 1/1." sits under that 100-card heading, so it is treated as covering both tcdb sets, which is why the four Rookie Portraits parallel rows carry source_confidence=multi.
• G-006 THE TWO SOURCES NAME THE SAME EXPRESSIONISTS MEMORABILIA TIER DIFFERENTLY. tcdb files sid 141423 as "2016-17 Panini Court Kings - Expressionists Memorabilia Prime" and renders MEM plus SN25 on eleven rows and MEM plus SN5 on card 12. cardboardconnection states "PARALLEL CARDS: Sapphire #/25, Masterpiece 1/1 (*Card #12 - #/5)." (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball) - naming the /25 tier Sapphire, with the same values on the same card. There is no Expressionists Memorabilia Sapphire sid on tcdb and no Expressionists Memorabilia Prime entry on cardboardconnection, so these are one tier under two names rather than two tiers. tcdb's name is written because tcdb holds the sid and the per-card rows. A resolver keyed on cardboardconnection will not find this row.
• G-007 THE TWO SOURCES DISAGREE ON THE VALUES OF 2 ON 2 QUAD MEMORABILIA PRIME, AND NEITHER IS WRITTEN. tcdb sid 141417 states "Total Cards: 9", renders 8 rows, and its distinct token set is MEM plus SN25 and SN10, with card 7 the SN10 row. cardboardconnection states "PARALLEL CARDS: Prime #/10 (*Cards #2, 3, 10 - #/5), Masterpiece 1/1." (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball) - a /10 tier with three /5 exceptions on different cards. The two accounts agree only that the tier is variable. print_run and serial_numbered are UNKNOWN and value_contested is yes. One rendered row of the nine was not returned by the fetch layer, so the tcdb token set is over 8 of 9 rows; that does not change the verdict, since two distinct tokens already establish variability.
• G-008 CARDBOARDCONNECTION APPEARS TO ATTACH A #/5 EXCEPTION TO THE WRONG EXPRESSIONISTS MEMORABILIA TIER. Its line reads "PARALLEL CARDS: Sapphire #/25, Masterpiece 1/1 (*Card #12 - #/5)." (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball). tcdb renders MEM plus SN1 on all 13 rows of Expressionists Memorabilia Masterpiece (sid 141422) and renders MEM plus SN5 on card 12 of Expressionists Memorabilia Prime (sid 141423). Both cannot be right about card 12. The per-row enumeration is preferred, Masterpiece is written as 1 with source_confidence=single and value_contested=yes, and the disagreement is recorded here rather than averaged away. This is unresolved: nobody has looked at a card in hand or at a marketplace listing.
• G-009 SET-WIDE PRINT RUNS CANNOT BE HELD BY sets.csv AND ARE RECORDED IN WORDS ONLY. sets.csv has no print_run column and EXTRACTION_SPEC 2.2 forbids putting the figure in card_count. Every row of the following sets carries one uniform serial token on tcdb, enumerated over all rendered rows: 2 on 2 Quad Memorabilia SN99 (9 rows), Art Nouveau Jumbo MEM plus SN99 (38 rows), Expressionists Memorabilia MEM plus SN149 (14 rows), Fresh Paint Duals AU plus SN50 (10 rows), Fresh Paint Variation AU plus SN200 (40 rows), Heir Apparent AU plus SN150 (10 rows), Performance Art Jerseys MEM plus SN249 (15 rows), Portraits SN175 (70 rows) and Rookie Portraits SN175 (30 rows). cardboardconnection independently states #/99, #/99, #/149, #/50, #/200, #/150, #/249 and #/175 for these (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball), and tcdb's own Inserts.cfm Notes column states "Autographed and serial numbered to 50.", "Autographed and serial numbered to 200." and "Serial numbered to 175" (https://www.tcdb.com/Inserts.cfm/sid/136900/2016-17-Panini-Court-Kings). Note also the sets that carry NO serial token on any row: AKA, Arc-eologists, Aurora, 5x7 Box Toppers Panoramics, 5x7 Box Toppers Rookie Royalty (no token of any kind), Art Nouveau (MEM only), Fresh Paint (AU only) and 5x7 Box Toppers Autographs (AU only). Those absences are NOT written as unnumbered anywhere, because sets.csv has no field for it and because an absent token is not a statement.
• G-010 NO VARIATION LAYER WAS RULED OUT, AND Errors.cfm CANNOT RULE ONE OUT - WITH A LOCAL DEMONSTRATION. https://www.tcdb.com/Errors.cfm/sid/136900/2016-17-Panini-Court-Kings returns "No errors or variations have been added for this set yet." Per caveats CAV-020 that sentence is returned identically on sets that provably DO carry errata, so it discriminates nothing and is quoted here as a row, not as evidence of absence. THIS RELEASE DEMONSTRATES THE POINT FROM INSIDE ITSELF: the Maestros subset, sid 141427, renders a "UER" annotation on card 1 (Ish Smith, "Photo actually Tony Wroten"), so errata exist in this product while the parent's Errors.cfm reports none. What survives independently: the base run is contiguous 1-200 with no unissued number and no b-suffixed row seen, established by a full enumeration across Checklist.cfm pages 1 and 2 and Gallery.cfm pages 8, 9, 10, 12, 18, 19 and 20, with the 171-179 overlap between the two endpoints agreeing row for row. What does not survive: any claim that no variation set exists.
• G-011 CHANNEL IS UNKNOWN ON ALL 26 PARALLEL ROWS AND channel_basis IS EMPTY ON ALL 26, WHICH IS THE SCHEMA SAYING THERE IS NO CHANNEL FOR A BASIS TO DESCRIBE. No source read states a trade channel, box SKU or region for any tier of this release. cardboardconnection gives only a product configuration, "Product Configuration: 10cards per pack, 1 pack per box, 16 boxes per case" (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball), which is a box format and not a channel; breakninja independently prints the same configuration as "16 Boxes per Case - 1 Pack Per Box - 10 Cards Per Pack" (https://breakninja.com/16-17-Panini-Court-Kings-Basketball-Checklist.html) and likewise states no channel. tcdb carries no channel note on any of the 48 sids. channel_confidence is left empty on every row, per the brief's reading of caveats CAV-038.
• G-012 TWO SOURCED PACK-ODDS STATEMENTS ABOUT BASE TIERS CANNOT BE STORED ANYWHERE. cardboardconnection states "Rookies III - 1:4 boxes" and "Rookies IV - 1 per case" (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball). Those are odds for two tiers of the BASE RUN, and base_card_tiers.csv has no stated_odds column while sets.csv holds exactly one base row for the whole 200-card set. The figures are therefore recorded only here. They also carry structural information the rest of the build could not source: the four rookie tiers are of increasing scarcity, which is the Court Kings stratification described in caveats CAV-020, and nothing in tcdb states it. Note cardboardconnection gives no odds line for Rookies I or Rookies II.
• G-013 SKETCHES AND SWATCHES IS NUMBERED 1-40 WITH SEVEN NUMBERS UNUSED, AND THAT RECONCILES ITS HEADER EXACTLY. tcdb states "Total Cards: 33". ViewSet.cfm rendered only 23 rows while spanning card numbers 1-40 - the caveats CAV-046 shape, where a row count disagreeing with a number span is the tell that interior rows were dropped. https://www.tcdb.com/Checklist.cfm/sid/141398/2016-17-Panini-Court-Kings-Sketches-and-Swatches enumerated the card numbers as 1-40 less 2, 6, 7, 12, 14, 23 and 24, and 40 - 7 = 33. The seven absences are NOT recorded as unissued numbers: per caveats CAV-044 a hole in one rendering is not evidence until a parallel page of the same product has been opened, and although the Masterpiece parallel (sid 141399) enumerates the same 33 numbers, both readings came from the same source. The card-level question belongs to the chase pass. card_count=33 rests on tcdb's header plus cardboardconnection's independent "33 cards." (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball).
• G-014 ONLY ONE EDITORIAL LINEAGE WAS AVAILABLE, AND THE RELEASE DATE RESTS ON TWO SOURCES THAT MAY BE ONE. release_date=2016-12-02 rests on tcdb's parent page, which renders an unqualified "Release Date: December 2, 2016" with the phrase "Estimated Release Date" nowhere on the page (https://www.tcdb.com/ViewSet.cfm/sid/136900/2016-17-Panini-Court-Kings), and on cardboardconnection's product page, which prints "Release Date: 12/2/2016" (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball). Those agree exactly, but per the 2016-17 scope memo the discriminator for independence is publication distance and neither publication distance was tested here, so this is not proven corroboration. No third lineage was reachable: beckett.com is confirmed unreachable per caveats CAV-033 (a Beckett news article on this exact product surfaced in search and was not opened); a web search for a checklistcenter.com page for this product surfaced none, and per caveats CAV-037 no URL was constructed; breakninja carries the product but states no release date, no base set size and no print runs. EVERY editorial claim in this release therefore rests on cardboardconnection alone.
• G-015 SIX PARALLEL TIERS HAVE A SMALLER ROSTER THAN THE SET THEY PARALLEL, AND FOUR OF THE SIX ARE CORROBORATED EXACTLY. Vintage Materials is 20 cards while its Masterpiece states 18 and its Prime states 16; Artistic Endeavors is 30 while its Prime states 27; Sketches and Swatches is 33 while its Prime states 31; Expressionists Memorabilia is 14 while its Masterpiece and Prime each state 13. cardboardconnection independently accounts for four of these by naming the missing cards - "Prime #/25 or less (*No card #15, 17, 19, 20), Masterpiece 1/1 (*No card #17, 20)." for Vintage Materials, "Prime #/25 or less (*No card #11, 26)" for Sketches and Swatches and "Prime #/25 or less (*No card #14, 18, 22)" for Artistic Endeavors (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball) - and 20-4=16, 20-2=18, 33-2=31 and 30-3=27 all reconcile exactly. The Expressionists 14-to-13 shortfall is NOT accounted for by any source. No parallel roster is written anywhere in this release, since parallels.csv has no card_count column; this is recorded so the chase pass does not read the shortfalls as coverage holes.
• G-016 INSERT AND AUTOGRAPH CARD-LEVEL CHECKLISTS ARE NOT WRITTEN. Per EXTRACTION_SPEC, insert_cards.csv is a separate pass under CHASE_CHECKLIST_SPEC and no insert_cards.csv exists in this directory. The 22 non-base sets in sets.csv have sourced names and counts but no card lists. Useful starting facts, all established in this pass: the sid block is 141376-141427 with BOTH boundaries proved foreign (141375 is "1994-95 Wheeling Thunderbirds (ECHL)", 141428 is "1992-93 Tulsa Oilers (CHL)"); Inserts.cfm listed exactly 48 rows under two headings, "Parallel Sets (2)" and "Insert Sets (46)", with no truncation and no refusal (https://www.tcdb.com/Inserts.cfm/sid/136900/2016-17-Panini-Court-Kings); the four numeric gaps inside the block were each probed and are all foreign hockey sets (141378, 141389 and 141401 are Roox Quad City Mallards (UHL) sets and 141413 is "2001-02 Quad City Mallards (UHL)"); so 48 listed rows map to 48 real sids with no interpolation and no omission. Letter-coded sets that will need care: Fresh Paint ("FP-AJH" ... "FP-WB4"), Fresh Paint Variation ("FP-AJH2" ... "FP-WB42") and Heir Apparent ("HA-BH" ... "HA-WB"). 5x7 Box Toppers Autographs uses plain integers running 2 to 24 for 22 cards.
• G-017 FRESH PAINT VARIATION IS FILED AS A SET AND THAT IS CONTESTABLE. tcdb gives it its own sid (141425) under the Insert Sets heading, its own "Total Cards: 40", and card codes that are the Fresh Paint code plus a trailing 2 - "FP-AJH2", "FP-JAM2", "FP-WB42" against Fresh Paint's "FP-AJH", "FP-JAM", "FP-WB4". cardboardconnection lists it as its own entry, "Fresh Paint Variation SP (40 cards, #/200)" (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball). It is written as an auto set on that basis. THE RISK: per CHASE_CHECKLIST_SPEC the trailing character on a tcdb variation row can be tcdb's own disambiguator rather than a printed character, in which case these 40 cards share their printed codes with Fresh Paint and the object is better modelled as a serial-numbered parallel of it. Nobody has looked at a card. Note Fresh Paint itself carries AU and no serial token on all 40 rows while this set carries AU and SN200 on all 40, which is consistent with either model.
• G-018 THE TIER NAMES ARE SINGULAR HERE AND THE SOURCES DISAGREE WITH EACH OTHER. tcdb renders the subset label inline on every row from 101 up as "Rookie I", "Rookie II", "Rookie III" and "Rookie IV", read off https://www.tcdb.com/Checklist.cfm/sid/136900/2016-17-Panini-Court-Kings?PageIndex=2 and confirmed on Gallery.cfm pages 12, 18, 19 and 20. cardboardconnection writes the PLURAL in its odds lines: "Rookies III - 1:4 boxes" and "Rookies IV - 1 per case" (https://www.cardboardconnection.com/2016-17-panini-court-kings-basketball). base_card_tiers.csv and base_cards.card_tier both follow tcdb, which prints a per-card label rather than prose, and they match each other exactly as caveats CAV-043 requires. The built Court Kings seasons do not agree either: courtkings-bkb-2017-18 writes the singular and 2018-19 through 2024-25 write the plural, so a consumer grouping tiers by name across the line will split them.
• G-019 TRUNCATION AND ENDPOINT BEHAVIOUR MEASURED ON THIS RELEASE, RECORDED FOR THE NEXT BUILD. Checklist.cfm page 1 of the parent returned 76 rows spanning card numbers 1-76 - contiguous, so a clean tail cut rather than the interior-drop shape of caveats CAV-046. Page 2 returned 79 rows spanning 101-179, also contiguous. THE ?sort=0 REMEDY MADE PAGE 2 WORSE, NOT BETTER: the same page under ?sort=0 returned 75 rows spanning 101-175, five fewer, which reproduces the impeccable-bkb-2016-17 finding recorded in caveats CAV-046 rather than the optic-bkb-2016-17 one. Gallery.cfm topped up both pages cleanly at 10 tiles per page and its 171-179 overlap with Checklist.cfm page 2 agreed row for row. Inserts.cfm did NOT truncate and did NOT refuse on this parent, with both sections returned complete and each section's row count matching the count in its own heading. ViewSet.cfm on the parent capped at 10 rows of 200, as documented. ViewSet.cfm on subset sids came back SHORT of its own stated total on six of the 48 sids - 141377 (81 of 100), 141376 (75 of 100), 141398 (23 of 33), 141399 (24 of 33), 141417 (8 of 9) and 141416 (8 of 9) - and a second endpoint recovered the full set every time it was tried.

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