2022-23 Recon

• G-001 FOTL hobby SKU release date conflict: dacardworld.com's FOTL box listing states "Release Date: May 24th, 2023" (same day as the general hobby release), while steelcitycollectibles.com's FOTL box listing states the box "released 5/16/23" (~8 days earlier). Both are working retailer sources; not resolved to a single date. General hobby release date (2023-05-24) used in release.csv is unaffected, since 5 independent sources agree on it.
• G-002 Card count disagreement: cardsmithsbreaks.com states 40 cards; sportscardchecklist.com states "39-40 Cards" (itself internally ambiguous). 40 used in sets.csv as the unambiguous figure from a full checklist source.
• G-003 No blog.paniniamerica.net ("The Knight's Lance") QC Gallery post or preview article found for 2022-23 Panini Recon Basketball despite explicit search (site: search and targeted queries for "dropping"/"arrives"/QC Gallery all returned only 2020-21 Recon, 2022-23 Contenders/Spectra, and 2023-24 Recon posts). Panini's own blog could not be used to corroborate or contradict the release date or checklist for this specific season.
• G-004 Base card checklist for numbers 121-200 is drawn from a single fully-quoted source (cardsmithsbreaks.com); tcdb.com corroborates cards 1-120 (identical player order/names where both were fetched) and independently confirms card #101 (Jerami Grant) and the total range boundaries (#200 Markelle Fultz / #201 Paolo Banchero split), but its AI-summarized fetch did not return a verbatim list for 121-200 to fully re-verify against.
• G-005 No numeric pull-rate odds (e.g. "1:X packs", "1 per case") found for any base parallel or insert/autograph set beyond print-run counts and general box-average guarantees ("2 autographs, 20 inserts or parallels per box on average"), which the scarcity spec explicitly excludes as odds. Searched dacardworld.com, steelcitycollectibles.com, blowoutcards.com, and blog.paniniamerica.net. Only Holo Dots carries any scarcity-relevant descriptive language ("super short-printed"); see scarcity_parallels.csv.
• G-006 Five plain 1..N runs are written on catalogue evidence only, with no uncontaminated marketplace seller title naming a player at a specific printed number: All Systems Go (20), Rookie Recon (25), Rookie Review (15), World Travelers (15) and Base Rookies (201-250). Each is corroborated first-and-last row number-for-number by three sources that agree exactly -- https://www.tcdb.com/Checklist.cfm/sid/358174/2022-23-Panini-Recon , https://www.cardboardconnection.com/2022-23-panini-recon-basketball-nba-cards and https://www.checklistcenter.com/2022-23-panini-recon-basketball-card-checklist/ -- and both aggregators passed calibration on the plain inserts of this release while failing on its coded autographs, but tcdb, cardboardconnection and checklistcenter are not demonstrably independent of one another on the plain sets, so a 1..N renumbering cannot be positively ruled out for these five. The other six plain sets (Fleet Feet, Future Legends, Rock the Rim, Sky's the Limit, Top of the Charts, Vector) each carry at least one clean seller/auction confirmation, cited in their rows' notes. For Base Rookies specifically, every eBay page found that names a card in the 201-250 range (e.g. https://www.ebay.de/itm/227073539716 for #237 Jalen Williams and https://www.ebay.com/p/6061300820 for #201 Paolo Banchero) carries the tcdb flag string "(RC)" on the page and is therefore circular.
• G-007 Three autograph sets' letter codes rest on tcdb alone: Claim to Fame Signatures (CFS-, 30 cards), Rookie Jersey Autographs (RJA-, 31) and Sophomore Acetate Autographs (SAA-, 40). Each code was read from that set's own tcdb base checklist and re-read identically on its Platinum parallel page, but both pages are the same source, so this is two opinions from one catalogue, not corroboration. Every marketplace candidate found for these three either carried a tcdb flag string on the page -- e.g. https://www.ebay.com/itm/135134880144 ("#GS-JSK /125 (AU)") and https://www.ebay.com/itm/336696342647 ("#RRS-KVH Red /99 (AU, RC)", whose search snippet had looked like a genuine seller title with an inventory SKU "16qs") -- or was a cross-product hit for a different Panini release that happens to share the RJA- prefix (2022-23 Select, Origins and One and One all use RJA-). comc.com card pages, whose URL path segment is the printed number, return HTTP 403 on this line. The other six autograph sets (CES-, GS-, RRS-, DNS-, and the plain prefixes reported in the notes) do have clean independent confirmation from goldin.co lot titles.
• G-008 Card-number conflict on one card, unresolved. tcdb renders Josh Giddey as DSG-JGD on both the base page https://www.tcdb.com/Checklist.cfm/sid/368588/2022-23-Panini-Recon and the Platinum parallel https://www.tcdb.com/Checklist.cfm/sid/368592/2022-23-Panini-Recon , while an eBay listing titled "2022-23 Recon Destined for Greatness Auto Purple #DSG-JGO Josh Giddey 09/75" at https://www.ebay.com/itm/304954504782 renders DSG-JGO. That page carries a tcdb "(AU)" flag string in its related-search module, which makes it circular evidence rather than an independent seller transcription, so it does not override tcdb -- but it is not dismissible either. DSG-JGD is written; DSG-JGO is recorded here and not written. A clean auction-house lot for this card would settle it.
• G-009 One card withheld for want of a printed number: Daishen Nix (Houston Rockets). tcdb stores the literal string "TBD" in the card-number column for this row on both the base page https://www.tcdb.com/Checklist.cfm/sid/368642/2022-23-Panini-Recon and the Platinum parallel https://www.tcdb.com/Checklist.cfm/sid/368646/2022-23-Panini-Recon , so the parallel-page test returns no answer. https://www.tcdb.com/Errors.cfm/sid/368642/2022-23-Panini-Recon---True-Potential-Signatures does return an entry for this card -- "Daishen Nix AU, SN149, EXCH" -- which identifies it as an exchange card but supplies no printed number. 39 of the set's 40 cards are written; sets.csv's card_count of 40 is therefore correct and the shortfall is this one row.
• G-010 Contradiction found and resolved in favour of sets.csv, recorded for audit. tcdb's base checklist https://www.tcdb.com/Checklist.cfm/sid/368635/2022-23-Panini-Recon renders 39 rows and its header states 39, but all three of its parallel tiers render 40 -- Gold https://www.tcdb.com/Checklist.cfm/sid/368636/2022-23-Panini-Recon , Green https://www.tcdb.com/Checklist.cfm/sid/368637/2022-23-Panini-Recon and Platinum https://www.tcdb.com/Checklist.cfm/sid/368638/2022-23-Panini-Recon -- and the extra card is SAA-KAI Kai Jones. sets.csv records 40 from cardsmithsbreaks.com, and its note does not cite tcdb, so the base-vs-parallel tiebreaker is armed for this set and sides with the parallels: the base page is short, this is recovery rather than parallel scope, and SAA-KAI has been written citing the parallel pages. checklistcenter.com also states 40. This supersedes nothing in G-002 but does resolve its 39-vs-40 ambiguity at 40 on a third independent source.

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