2016-17 Impeccable

• G-001 THE TASK BRIEF PREDICTED base_set_size=158 WITH TWO UNISSUED NUMBERS INSIDE A TIERED BASE RUN, AND THE MEASUREMENT REFUTES IT. Recorded because it is a second data point on the open flawless-bkb-2018-19 schema question and the answer turns out to be the easy one. What is actually true: the base run is cards 1-100, contiguous, with NO unissued number and a UNIFORM print run of 99, so base_print_run is a single integer and no base_card_tiers.csv is needed or written. The deferred check_tier_partition() in migrations/009_triggers_and_views.sql is never engaged by this release and nothing had to be forced. tcdb's parent Notes line "Distributed in one series (#1-160). #103 and #137 do not exist." is quoted correctly by Phase 0 but the 1-160 series is not one base run: Checklist.cfm?PageIndex=2 renders all 58 rows of 101-160 and every one carries AU, so 101-135 ("AU, RC", 34 cards) and 136-160 ("AU, MEM, RC", 24 cards) are autograph blocks that EXTRACTION_SPEC 5.3 excludes from base_set_size. BOTH HOLES SIT INSIDE THOSE BLOCKS, NOT IN THE BASE RUN - 103 in the first, 137 in the second. 100 + 34 + 24 = 158 reconciles tcdb's header exactly, and cardboardconnection reaches the same decomposition independently. CONVENTION CHECK PER EXTRACTION_SPEC SECTION 6: impeccable-bkb-2017-18 applies the identical exclusion (base_set_size=100 against a parent Total Cards of 159, with 101-120 and 121-160 filed as separate autograph sets), so the two adjacent seasons of this line agree and there is no divergence to repair here. https://www.tcdb.com/Checklist.cfm/sid/156489/2016-17-Panini-Impeccable?PageIndex=2 ; https://www.cardboardconnection.com/2016-17-panini-impeccable-basketball
• G-002 THE 101-135 ROOKIE AUTOGRAPH BLOCK HAS NO SINGLE PRINT RUN AND sets.csv CANNOT HOLD WHAT IT DOES HAVE. sets.csv has no print_run column (EXTRACTION_SPEC 2.2), so the figures survive only in that row's notes and here. The distinct serial-token set was enumerated over all 34 rows rather than sampled: SN75 on cards 102 Brandon Ingram, 106 Kris Dunn, 107 Buddy Hield and 110 Jamal Murray, and SN99 on the other thirty. Confirmed row for row on a second endpoint for cards 101-111, where Gallery.cfm?PageIndex=11 renders the identical SN75/SN99 split. The four SN75 cards are the four highest-profile rookies in the block, which is a coherent reason for the split rather than a transcription artifact. WHAT WAS REFUSED: no modal token was written; "99" is not this block's print run. https://www.tcdb.com/Checklist.cfm/sid/156489/2016-17-Panini-Impeccable?PageIndex=2 ; https://www.tcdb.com/Gallery.cfm/sid/156489/2016-17-Panini-Impeccable?PageIndex=11
• G-003 THREE SETS ARE NUMBERED TO A PER-CARD VALUE, WHICH IS THE CAV-023 SCHEME AND WHICH NO COLUMN IN THIS SCHEMA CAN CARRY. Each distinct token set was enumerated over every row of the set, not sampled. Impeccable Stats Autographs, 19 rows, FOURTEEN distinct tokens: SN81, SN73, SN64, SN63, SN62, SN61, SN60, SN25, SN24, SN23, SN13, SN12, SN11, SN7. Impeccable Jersey Numbers Autographs, 20 rows, SIXTEEN distinct tokens: SN91, SN35, SN33, SN32, SN30, SN27, SN24, SN23, SN22, SN21, SN20, SN14, SN11, SN4, SN2, SN1 - each card evidently numbered to that player's own jersey number, which is the Jersey Number scheme CAV-023 names. Impeccable Season Autographs, 10 rows, three distinct tokens: SN21, SN20, SN19. A three-row sample of any of these would have produced a clean sourced-looking integer and been wrong; no modal token was written for any of them. https://www.tcdb.com/Checklist.cfm/sid/158688/2016-17-Panini-Impeccable-Impeccable-Stats-Autographs?PageIndex=1 ; https://www.tcdb.com/Checklist.cfm/sid/158690/2016-17-Panini-Impeccable-Impeccable-Jersey-Numbers-Autographs?PageIndex=1 ; https://www.tcdb.com/Checklist.cfm/sid/158689/2016-17-Panini-Impeccable-Impeccable-Season-Autographs?PageIndex=1
• G-004 TWO MORE SETS WITH NO SINGLE PRINT RUN, AND ONE OF THE TWO SWEEPS IS NOT PROVABLY COMPLETE. Indelible Ink (sid 158683, 50 rows) carries the distinct set AU+SN75 and AU+SN99. Elegance Veteran Jersey Autographs (sid 158691, 29 rows) carries AU+MEM+SN99 and AU+MEM+SN75. THE HONEST QUALIFICATION ON THE SECOND ONE: that response's own per-token counts summed to 23 against its own 29 rendered rows, so six rows are unaccounted for and a third token could exist on a row the fetch dropped. The two tokens seen are evidence and the conclusion that the set is variable is safe in the direction it matters; the completeness of the token SET is not established, and one differently-varied re-fetch would close it. This is the same shape the 2016-17 scope memo records for the Prestige Distinctive Ink probe. https://www.tcdb.com/Checklist.cfm/sid/158683/2016-17-Panini-Impeccable-Indelible-Ink?PageIndex=1 ; https://www.tcdb.com/Checklist.cfm/sid/158691/2016-17-Panini-Impeccable-Elegance-Veteran-Jersey-Autographs?PageIndex=1
• G-005 CARD-COUNT DISAGREEMENT RESOLVED AGAINST tcdb's OWN BASE PAGE, AND RECORDED RATHER THAN SMOOTHED. tcdb sid 158696 states "Total Cards: 6" and renders 6 rows. TWO INDEPENDENT LEGS SAY 10: cardboardconnection states "Elegance Retired Jersey Autographs Set Checklist: 10 cards", and tcdb's own Silver parallel of the same set (sid 158700) states "Total Cards: 10" and renders ten rows - Anfernee Hardaway, Allen Iverson, Alonzo Mourning, Chauncey Billups, David Robinson, George Gervin, Kenny Smith, Kurt Thomas, Ray Allen, Rashard Lewis. This is the CHASE_CHECKLIST_SPEC 1c-ter shape where the base page is short rather than the parallel being out of scope, and it is settled here by an editorial source that is independent of tcdb rather than by taste. 10 is written. Note the tier topology is NOT nested: Holo Gold and Platinum each render 9 and Holo Silver renders 5, so no page is a superset of any other and the true roster could in principle exceed 10. A chase pass should build this set from the Silver page, not the base page. https://www.tcdb.com/Checklist.cfm/sid/158696/2016-17-Panini-Impeccable-Elegance-Retired-Jersey-Autographs?PageIndex=1 ; https://www.tcdb.com/Checklist.cfm/sid/158700/2016-17-Panini-Impeccable-Elegance-Retired-Jersey-Autographs-Silver?PageIndex=1 ; https://www.cardboardconnection.com/2016-17-panini-impeccable-basketball
• G-006 ONE PARALLEL TIER IS PROVABLY SERIAL NUMBERED WITH A PER-CARD VARIABLE RUN, WHICH THE SCHEMA CANNOT EXPRESS, SO THE ROW CARRIES UNKNOWN/UNKNOWN AND THE HELD FACT LIVES HERE. tcdb sid 158697 states "Total Cards: 9", rendered all nine rows in one fetch, and the distinct flag-token set enumerated over all nine is "AU, MEM, SN10" on eight rows and "AU, MEM, SN8" on one, Chauncey Billups. Per EXTRACTION_SPEC 2.3 the forced print_run/serial_numbered pairing collapses this to UNKNOWN/UNKNOWN, which a consumer cannot distinguish from "nothing was found" - it is not that. The tier is set-scope, so UNKNOWN is legal; had it been base-scope, CAV-041 says there would have been no legal row at all. THE COST HERE IS SMALL AND THE SHAPE IS THE EXPENSIVE ONE: 1 of 20 parallel tiers in this release, 5 percent, against 40.5 percent measured on 2017-18 Immaculate and 40 percent on 2017-18 National Treasures. This release is a strong counter-example to the premium-line variability prior at the PARALLEL level, while confirming it at the autograph-set level, where 5 of 9 sets are variable. https://www.tcdb.com/Checklist.cfm/sid/158697/2016-17-Panini-Impeccable-Elegance-Retired-Jersey-Autographs-Holo-Gold?PageIndex=1
• G-007 CHANNEL IS WRITTEN AS hobby ON PRESENCE EVIDENCE ONLY, NOT EXCLUSIVITY. The only fetched source that names a trade channel for this product is a retailer product page whose own title reads "2016/17 Panini Impeccable Basketball Hobby Box", with a configuration line "3 Boxes Per Case, 1 Pack Per Box, 8 Cards Per Pack" that matches cardboardconnection's "Product Configuration: 8 cards per pack, 1 pack per box, 3 boxes per case" exactly. cardboardconnection itself contains no sentence using the words hobby, retail, exclusive or distributor - asked directly, the answer was that none is present - and tcdb states no channel. Per CAV-010 a single named channel value asserts EXCLUSIVITY, which this evidence does not support, so channel_basis is sourced-presence and channel_confidence is single. The value is applied to every tier because this is a single-SKU super-premium product with one box configuration and no retail version was found by any source; that reasoning is stated rather than hidden. DIVERGENCE FROM THE SIBLING SEASON, RECORDED NOT COPIED: impeccable-bkb-2017-18 writes channel_confidence=multi on all 22 of its parallel rows off a single steelcitycollectibles product title, which looks like an over-claim under the vocabulary that channel_confidence counts INDEPENDENT SOURCES; this release writes single. Nothing in the other release was touched. https://www.dacardworld.com/sports-cards/2016-17-panini-impeccable-basketball-hobby-box ; https://www.cardboardconnection.com/2016-17-panini-impeccable-basketball
• G-008 THIS RELEASE RESTS ON EXACTLY TWO LINEAGES AND ONE OF THEM IS tcdb. checklistcenter carries NO page for this release: a site-scoped search returned its 2019-20, 2020-21, 2021-22, 2022-23 and 2023-24 Impeccable pages and nothing for 2016-17 or 2017-18. Per CAV-037 no URL was constructed and no 404 was treated as evidence; the absence here comes from a search result, which is the only admissible route. beckett.com was not read, per CAV-033 and the task brief, because it 302s to a maintenance page. The practical consequence: every value in this release is corroborated by at most tcdb plus cardboardconnection. Where the two agree - base set 100 cards at /99, the 34 and 24 block sizes, and the Silver/Holo Silver/Holo Gold/Platinum ladder - they are genuinely independent legs, because cardboardconnection carries the block names and counts that tcdb does not name at all (both blocks are sid-less inside the parent checklist) and tcdb carries the row-level serial tokens cardboardconnection has no equivalent of. Those rows are marked multi. Every set-scope parallel print run rests on tcdb alone and is marked single. https://www.checklistcenter.com/category/basketball-checklists/
• G-009 NO CLAIM IS MADE THAT THIS RELEASE HAS NO VARIATION SET. Errors.cfm on the parent sid 156489 returned the sentence "No errors or variations have been added for this set yet." Per CAV-020 that string is returned identically on sets that provably DO have errata and on sets that appear not to, so it discriminates nothing and nothing is concluded from it. WHAT SURVIVES WITHOUT IT, because it rests on independent contiguous enumeration rather than on this endpoint: base_set_size=100, the contiguity of 1-100 with no unissued number, the two unissued numbers at 103 and 137 inside the autograph blocks, and the finding that nothing is numbered past 160. WHAT DOES NOT SURVIVE: any statement about b-suffixed or photo-variation rows, which a base-run enumeration is not designed to surface and which Gallery.cfm strips in any case. https://www.tcdb.com/Errors.cfm/sid/156489/2016-17-Panini-Impeccable
• G-010 rookie_flag IS UNKNOWN ON EVERY BASE CARD. tcdb rendered a flag column on every page fetched and it carried serial tokens throughout, and it carries RC on the 101-160 autograph rows of the same parent - so the renderer demonstrably emits RC for this product - yet no base row 1-100 carries one on either Checklist.cfm or the ten Gallery.cfm pages. That is a positive discriminative signal that tcdb does not designate these as rookie cards, but it is still an absence rather than a source stating "no", and the base run does include at least one 2016-17 rookie (card 67 Ben Simmons). UNKNOWN is written rather than "no", per CAV-003 and CLAUDE.md's rule that UNKNOWN means unresolved. cardboardconnection prints no RC designations either. https://www.tcdb.com/Checklist.cfm/sid/156489/2016-17-Panini-Impeccable?PageIndex=1 ; https://www.cardboardconnection.com/2016-17-panini-impeccable-basketball
• G-011 THE Silver TIER COLLAPSES TO parallel_family=other. parallel_family was derived, as required, from the single ordered FAMILIES list in pipeline/fix_schema.py lifted by AST so the repo keeps one definition; nothing was added to that file and no family was minted, per CAV-039. The list has Platinum and it has Holo, so Holo Silver, Holo Gold and Platinum derive correctly; it has no Silver pattern, so all five Silver rows derive to other. Reported so the question can be settled once for the Impeccable line rather than per release - the line's ladder is Silver / Holo Silver / Holo Gold / Platinum in both 2016-17 and 2017-18. Note CAV-021 already records an unresolved question about what parallel_family should hold, and CAV-039 records 268 hand-written values across ten releases including five Impeccable seasons, so a fix here must be decided with those. No corpus-wide re-derivation should be run before that decision. https://www.tcdb.com/Inserts.cfm/sid/156489/2016-17-Panini-Impeccable
• G-012 EIGHT PARALLEL TIERS RENDER FEWER CARDS THAN THEIR PARENT SET, recorded because it looks exactly like a coverage hole and is not one, and because parallels.csv has no card_count column to hold it. Indelible Ink is 50 and its Silver renders 33. Elegance Veteran Jersey Autographs is 29 and its Holo Gold renders 28, Platinum 28, Silver 22 and Holo Silver 16. Elegance Retired Jersey Autographs is 10 and its Holo Gold renders 9, Platinum 9 and Holo Silver 5. In every case the tier's own stated Total Cards and its rendered row count agree with each other, so these are genuinely shorter rosters rather than truncated fetches. The three product-wide tiers do NOT have this shape: Holo Silver, Holo Gold and Platinum each state Total Cards 158 and render the full 1-100 base run plus all 58 rows of 101-160. https://www.tcdb.com/Inserts.cfm/sid/156489/2016-17-Panini-Impeccable
• G-013 NO MARKETPLACE EVIDENCE WAS OBTAINED FOR ANY VALUE IN THIS RELEASE. EXTRACTION_SPEC 3.4 requires marketplace corroboration before writing plain integers onto an autograph set. That specific failure is not live here because no card_number was written onto any autograph set - this build produced no insert_cards.csv, and base_cards.csv holds only the base run, whose numbers are plain 1-100 on a non-autograph set. What is uncorroborated by a card in hand: the four print runs 99, 25, 10 and 49, each of which rests on tcdb's flag column plus cardboardconnection. One goldin or eBay free-text title showing a 2016-17 Impeccable base card serialled to 99, or a Holo Silver to 25, would close it. The 2016-17 rookie class in this product (Simmons, Ingram, Hield, Murray, Brown) is exactly the raw material the scope memo says exists and nobody has tested for this season. https://www.cardboardconnection.com/2016-17-panini-impeccable-basketball
• G-014 NO PRIOR-SEASON SUBSET APPEARS UNDER THIS PARENT, AND THAT IS EVIDENCED RATHER THAN ASSUMED; THE MIRROR DIRECTION WAS NOT CHECKED. Inserts.cfm on sid 156489 lists exactly 24 rows under two headings, "Insert Sets (20)" and "Parallel Sets (4)", and every one of the 24 belongs to 2016-17 Panini Impeccable - none carries another season's prefix or another product's name, unlike the 68 foreign rows the 2016-17 scope memo measured on the National Treasures parent. So EXTRACTION_SPEC 5.5 had nothing to file here. The enumeration is bounded on both sides: sid 158679 is "2002-03 Cartes, Timbres et Monnaies Sainte-Foy Drummondville Voltigeurs (QMJHL)" and sid 158705 is "1997 Scoreboard Mickey Mantle - Promos", both foreign, and the one numeric gap inside the block, sid 158681, was probed and is "1999-00 Basingstoke Bison (BNL)" - also foreign - so 25 sids minus 1 foreign equals the 24 listed rows exactly and nothing is hiding. WHAT WAS NOT DONE: the EXTRACTION_SPEC 5.5 mirror. The 2017-18 Impeccable parent (sid 179535) was not checked for a "2016-17 Panini Impeccable Update" block, so a card that every page here says is unissued could in principle sit there. This applies specifically to 103 and 137. https://www.tcdb.com/Inserts.cfm/sid/156489/2016-17-Panini-Impeccable
• G-015 THREE NAMES DIFFER FROM THE ADJACENT SEASON OF THE SAME LINE AND NONE WAS NORMALISED, per the never-rename rule. (1) tcdb prints "Impeccable Jersey Numbers Autographs" here, plural Numbers, where impeccable-bkb-2017-18 carries "Impeccable Jersey Number Autographs", singular. (2) tcdb prints "Indelible Ink" where cardboardconnection prints "Indelible Ink Autographs Set Checklist"; this release follows tcdb, which renders the cards, and 2017-18 does the same. (3) The 2016-17 base-run parallel ladder is Holo Silver /25, Holo Gold /10, Platinum 1/1 with a partial Silver /49 over 101-160 only; 2017-18 is Holo Silver /49, Holo Gold /10, Platinum 1/1 with no Silver tier at base scope. Holo Silver MOVES from /25 to /49 between the two seasons, so nothing here was inherited and a resolver keyed on the tier name across seasons will get the print run wrong. Separately, name_ambiguous is written "no" for Holo Gold in this release because no parallel named plain Gold exists in it, where the 2017-18 sibling writes "yes" with ambiguous_with=Gold against a set named Gold NBA Logo rather than a parallel; CAV-012 scopes the flag to sibling PARALLEL names, so this release follows CAV-012. https://www.tcdb.com/Inserts.cfm/sid/156489/2016-17-Panini-Impeccable
• G-016 ADVANCE WARNING FOR THE CHASE PASS, not a defect in this release's files. No insert_cards.csv was produced - card-level insert and autograph checklists are a separate pass. What was observed in passing and is worth not rediscovering: Indelible Ink (sid 158683) renders LETTER-CODED card numbers of the form II-AD, II-CB, II-DB, and two codes appeared twice in one rendering (II-DC and II-JB), which per CHASE_CHECKLIST_SPEC may be a genuine initials collision rather than an error and must be adjudicated, not dropped. By contrast the two autograph blocks in the parent checklist are numbered as PLAIN INTEGERS 101-135 and 136-160, which is the base-block scheme and not Panini's usual letter coding, so marketplace corroboration is required before those integers are written as card numbers. cardboardconnection was not tested for renumbering on any coded set in this release, so it carries no calibration verdict here. https://www.tcdb.com/Checklist.cfm/sid/158683/2016-17-Panini-Impeccable-Indelible-Ink?PageIndex=1
• G-017 THREE FETCH-LAYER DEFECTS FIRED IN THIS BUILD AND ALL THREE WERE CAUGHT BY THE DOCUMENTED CHECKS. (1) A SELF-COUNT DISAGREED WITH ITS OWN LIST, TWICE. Checklist.cfm?PageIndex=2 on the parent asserted "Total rows listed: 60" over a list of 58; Checklist.cfm?PageIndex=2 on the Platinum sid asserted 38 rows over a list of 58. My own count of the enumerated rows was used both times and the header reconciliation 100+34+24=158 confirms 58 is right. (2) A Gallery.cfm PAGE SILENTLY DROPPED A TILE. Gallery.cfm?PageIndex=10 on Silver NBA Logo sid 158680 returned nine tiles with card 99 absent; varying the URL with &sort=0 returned all ten including 99 Carmelo Anthony. Had the first reading been trusted, an unissued number would have been recorded on a set that has none. (3) ?sort=0 DID NOT BEAT TRUNCATION HERE, against the 2016-17 scope memo's 2-for-2 record: on Holo Silver sid 158702, Checklist.cfm?PageIndex=1 returned 75 rows under the default sort and the same page under ?sort=0 returned fewer while still reporting the same 1-75 span, so it was used as a differently-varied confirmation rather than as a truncation fix and Gallery.cfm supplied the tail. Recorded so the sort=0 remedy is not treated as reliable. https://www.tcdb.com/Gallery.cfm/sid/158680/2016-17-Panini-Impeccable-Silver-NBA-Logo?PageIndex=10&sort=0
• G-018 THE SUSPICIOUS-UNIFORMITY CHECK FIRED AND WAS INVESTIGATED RATHER THAN IGNORED. All eight Impeccable seasons already in data/ carry base_set_size=100 with base_print_run=99 and base_print_run_basis=sourced, and this release makes a ninth with identical values. CLAUDE.md fabrication check 5 names identical structure across releases as a template-fill smell, so it was tested rather than accepted. WHAT MAKES IT A REAL FINDING RATHER THAN AN INHERITANCE: every value here was derived from 2016-17 evidence only - cardboardconnection's "100 cards. Serial numbered #/99." on the 2016-17 product page, and the tcdb flag token "SN99" enumerated over all 100 base rows on two endpoints - and NO 2017-18 value was carried across. The structures around the base run actively DIVERGE between the two adjacent seasons, which a template fill would not produce: 2016-17 Holo Silver is /25 where 2017-18 Holo Silver is /49; 2016-17 carries a partial Silver /49 tier over 101-160 that 2017-18 has no equivalent of; the two autograph blocks are 101-135 and 136-160 here against 101-120 and 121-160 there, with 34 and 24 cards against 20 and 39; and 2016-17 has two unissued numbers where 2017-18 has one, at a different number. The observation to carry forward is a genuine one about the LINE, stated as a consistency finding and not as a source: Impeccable has kept a 100-card base run serialled to 99 in every season this catalogue has measured, 2016-17 through 2023-24. https://www.cardboardconnection.com/2016-17-panini-impeccable-basketball

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