For platforms · The week on chain · 11 to 17 October 2026

The week on chain.

What every readable collectibles platform's own public record shows for one week: packs, sell-backs, time to sell-back, spend, buy-backs and what left the vault, read from public chains with no keys. Exact where every event was read; an estimate with its 95% interval where we sampled.

Chain week
11 to 17 October 2026
Read
21 and 22 October 2026
Followed to
21 October 2026
Platforms measured
7

Before the table

How we know what we know.

Every figure here comes from the public record, read by us. On these platforms, each pack a buyer opens, each card the operator buys back, each transfer and each card burned for shipping is written to a public blockchain as a dated transaction anyone can read. We read them through public nodes, with no keys, no accounts and no access to any company's systems. We never read a company's own servers; where a platform keeps its records there, we say so and leave the figure blank.

A figure marked exact means we read every event in the week. A figure marked estimate means we drew a random sample of the week's transactions and scaled it, and we show the range it could fall in. Where the record cannot tell two things apart, such as a card sold back and a card shipped, we say that too.

The chain proves the token: when it was made, who held it, when it was sold back, when it was burned. It does not prove the card on the shelf. Every count is rerun the same way each week with the same scripts, and a correction, when one is needed, is published with its date.

The table

Every measured platform, side by side.

Each cell is marked. Exact means every event in the week was read. Estimate means a seeded random sample of the week’s transactions, scaled to the week, and the cell carries its 95% interval. A sample figure without an interval is not shown; the cell says why. Packs bought in the week are followed to 21 October 2026 for their sell-backs.

  1. Collector Crypt

    Solana

    Exact counts from memos; dollars from a seeded random sample of 1,000 of the week's packs

    Packs opened or bought

    192,742 packs

    exact

    Every storefront that settles through the platform's gacha wallet

    How paid

    USDC from wallets; card-rail top-ups are not separable per pack (0 of the 1,000 sampled opens)

    from a sample

    Sold straight back

    188,387 packs

    exact

    Followed to 21 October 2026 23:20 UTC

    Share sold straight back

    97.7%

    exact

    Median time to sell-back

    7 s

    exact

    90th percentile time to sell-back

    50 s

    exact

    Pack spend

    $42,738,611

    95%: $34,858,439 to $50,618,784

    estimate, 95% interval

    A few $2,500 and $5,000 packs carry much of the spend, which is why the interval is wide

    Buy-backs

    188,387 packs

    exact

    Buy-back value

    $35,144,590

    95%: $28,116,065 to $42,173,115

    estimate, 95% interval

    Buy-back cash in the same 1,000-pack sample, scaled to the week

    Spend returned through buy-backs

    82.2%

    95%: 72.4% to 92.1%

    estimate, 95% interval

    Ratio estimate from the same 1,000-pack sample

    Left the vault

    1,613 cards

    exact

    Burned for shipment in 488 transactions, 0.8% of packs; $23,734 in fees

    Average spend per pack

    $222

    95%: $181 to $263

    estimate, 95% interval

    Cert record

    Not part of the weekly read; the cert record is kept by the daily cert check

  2. Collector Crypt

    Robinhood Chain

    Exact, from contract events

    Packs opened or bought

    757 packs

    exact

    How paid

    USDG from wallets

    exact

    Sold straight back

    576 packs

    exact

    Of 756 packs with a card minted

    Share sold straight back

    76.2%

    exact

    Median time to sell-back

    34 s

    exact

    Block differences times the window's mean block time

    90th percentile time to sell-back

    118 s

    exact

    Pack spend

    $173,930

    exact

    Buy-backs

    576 packs

    exact

    Buy-back value

    $81,659

    exact

    Spend returned through buy-backs

    46.9%

    exact

    Left the vault

    10 tokens

    exact

    Tokens listed in redemption requests

    Average spend per pack

    $230

    exact

    Cert record

    Not part of the weekly read; the cert record is kept by the daily cert check

  3. Courtyard

    Polygon

    Exact, from contract events

    Packs opened or bought

    269,548 packs

    exact

    160,590 of them paid from a wallet; the sold-back, spend and buy-back rows cover that wallet-paid set

    How paid

    59.6% wallet USDC; 25.2% card checkout with a USDC leg; 15.2% card only

    exact

    Sold straight back

    146,867 packs

    exact

    Wallet-paid packs, followed to 21 October 2026 22:37 UTC

    Share sold straight back

    91.4%

    exact

    Of wallet-paid packs

    Median time to sell-back

    21 s

    exact

    90th percentile time to sell-back

    3.9 h

    exact

    Pack spend

    $8,040,293

    exact

    Wallet-paid packs; a further $3,711,072 of USDC legs sits under card checkouts, and card-only spend is not visible on chain

    Buy-backs

    146,867 packs

    exact

    Matched to the week's wallet-paid packs; all buy-backs in the window by any path: 255,198

    Buy-back value

    $5,931,563

    exact

    All paths in the window: $12,911,602

    Spend returned through buy-backs

    73.8%

    exact

    Wallet-paid cohort

    Left the vault

    Not confirmed on chain; an upper bound of 26,679 burns by a role wallet

    exact

    Average spend per pack

    $50

    exact

    Wallet-paid; $55 excluding $0 packs

    Cert record

    Not readable: token metadata sits on the platform's own server, whose terms bar automated reading

  4. Phygitals

    Solana

    Sample estimates: random samples of 750 and 300 of the week's transactions, scaled to the week's counts

    Packs opened or bought

    47,108 payment transactions

    95%: 42,874 to 51,342

    estimate, 95% interval

    Payments into the platform, not packs; a payment can cover more than one pack

    How paid

    USDC from wallets

    from a sample

    Sold straight back

    74,305 sell-back transactions

    95%: 70,070 to 78,539

    estimate, 95% interval

    Share sold straight back

    A ratio of estimates from two different samples (sell-backs on the USDC account, awards on the settlement wallet), with no interval, so not shown

    Median time to sell-back

    22 s

    95%: 19 s to 23 s

    estimate, 95% interval

    From the card's award to its sell-back; order-statistic interval on 459 sampled sell-backs

    90th percentile time to sell-back

    126 s

    95%: 77 s to 4 min

    estimate, 95% interval

    From the card's award to its sell-back; order-statistic interval

    Pack spend

    $4,775,468

    95%: $1,973,070 to $7,577,866

    estimate, 95% interval

    A $5,000 pack in a sample of hundreds moves the estimate a great deal, which is why the interval is wide

    Buy-backs

    74,305 sell-back transactions

    95%: 70,070 to 78,539

    estimate, 95% interval

    Buy-back value

    $2,336,057

    95%: $1,728,496 to $2,943,617

    estimate, 95% interval

    Spend returned through buy-backs

    48.9%

    95%: 17.1% to 80.7%

    estimate, 95% interval

    Ratio estimate over the USDC-account sample, with its residual standard error

    Left the vault

    Not separable: the sample held no burns

    Average spend per pack

    $101

    95%: $42 to $160

    estimate, 95% interval

    Per payment transaction, not per pack

    Cert record

    Not readable: the on-chain asset holds a name and a URI only, and the metadata host's terms bar bots

  5. Gacha Pull

    Abstract

    Exact, from contract events

    Packs opened or bought

    13,382 packs

    exact

    How paid

    USDC.e from wallets

    exact

    Sold straight back

    12,632 packs

    exact

    Share sold straight back

    94.4%

    exact

    Median time to sell-back

    34 s

    exact

    Block differences times the window's mean block time of 1.49 seconds

    90th percentile time to sell-back

    5 min

    exact

    Pack spend

    $404,004

    exact

    Buy-backs

    12,632 packs

    exact

    Buy-back value

    $325,553

    exact

    Split evenly across the cards in one buy-back event

    Spend returned through buy-backs

    80.6%

    exact

    Left the vault

    Upper bound of 9 burns outside sell-backs

    exact

    Average spend per pack

    $30

    exact

    Cert record

    Not sampled: token metadata sits on the platform's own server

  6. Beezie

    Solana

    Sample estimates: a random sample of 200 of the week's 14,679 pool transactions, scaled to the week

    Packs opened or bought

    8,000 play payments

    95%: 6,987 to 9,013

    estimate, 95% interval

    A play payment can cover several plays, so pulls are not counted

    How paid

    USDC from wallets

    from a sample

    Sold straight back

    6,532 sell-back transactions

    95%: 5,521 to 7,543

    estimate, 95% interval

    2.4 cards per sell-back on average

    Share sold straight back

    Plays are not counted, so no share of pulls can be read

    Median time to sell-back

    Not measured this week

    90th percentile time to sell-back

    Not measured this week

    Pack spend

    $1,657,920

    95%: $1,077,235 to $2,238,605

    estimate, 95% interval

    Buy-backs

    6,532 sell-back transactions

    95%: 5,521 to 7,543

    estimate, 95% interval

    Buy-back value

    $1,256,100

    95%: $812,165 to $1,700,036

    estimate, 95% interval

    Spend returned through buy-backs

    75.8%

    95%: 35.2% to 116.3%

    estimate, 95% interval

    Ratio estimate over the same sample, with its residual standard error

    Left the vault

    Not measured this week

    Average spend per pack

    $207

    95%: $139 to $275

    estimate, 95% interval

    Per play payment

    Cert record

    The metadata host refused the read

  7. Raflux

    Solana

    Sample estimates: a random sample of 100 of the week's 6,364 pool transactions, scaled to the week

    Packs opened or bought

    1,655 spin payments

    95%: 1,108 to 2,202

    estimate, 95% interval

    How paid

    USDC from wallets

    from a sample

    Sold straight back

    Not separable: a token is minted only if the buyer keeps the item, so cash-outs and sell-backs look the same on chain

    Share sold straight back

    Not separable from cash-outs

    Median time to sell-back

    Not measured this week

    90th percentile time to sell-back

    Not measured this week

    Pack spend

    $87,187

    95%: $53,850 to $120,524

    estimate, 95% interval

    Buy-backs

    1,400 payouts

    95%: 883 to 1,917

    estimate, 95% interval

    Buy-back value

    $70,477

    95%: $34,368 to $106,587

    estimate, 95% interval

    Spend returned through buy-backs

    80.8%

    95%: 24.5% to 137.2%

    estimate, 95% interval

    Payouts over spend; ratio estimate over the same sample, with its residual standard error

    Left the vault

    Not measured this week

    Average spend per pack

    $53

    95%: $42 to $63

    estimate, 95% interval

    Per spin payment

    Cert record

    The card collection was not identified

Where adding up is valid

Packs, counted exactly

476,429

exact

Collector Crypt on Solana, Collector Crypt on Robinhood Chain, Courtyard on Polygon, Gacha Pull on Abstract, every pack read. Phygitals, Beezie and Raflux count payments, not packs, so they are not added.

Spend, exact part

$12,329,299

exact

collector-crypt-robinhood, courtyard (wallet-paid), courtyard (card-checkout USDC legs), gacha-pull. The sample estimates for Collector Crypt on Solana, Phygitals on Solana, Beezie on Solana, Raflux on Solana are not added to it, and Courtyard's card-only spend is not visible on chain.

Spend visible on chain, exact part plus sample estimates

$61,588,484

95%: $53,204,639 to $69,972,329

estimate, 95% interval

The exact part plus the sample estimates, their variances added as independent. Courtyard's card-only spend is not visible on chain and is not included.

Not measured this week

Named, with the reason.

A platform is left out when its public endpoint does not serve the week without a key, when it refused the read, or when its contract carried too little in the week to measure. The reason is written down rather than a zero.

  • Collector Crypt on BaseBase history is not served by public endpoints without a key.
  • Beezie on BaseBase history not public without a key; Base side winding down.
  • Beezie on Flow EVMNo mints or burns in the census week.
  • rip.fun, DYLIBase history not public without a key.
  • Shiny on Base, Abstract, SolanaBase history not public; no order events on Abstract in the census week.
  • RenaissPublic BNB endpoints refused historical log reads.
  • Ready Cards19 treasury transactions in the census week.
  • BAXUSSpirits; no pack or buy-back flow.
  • ArianeePassports without custody; no pack or buy-back flow.
  • OpenVinoWine; no pack or buy-back flow.

Method

Read only, public endpoints, no keys.

All reads were made on 21 and 22 October 2026 (UTC), read only, through public endpoints with no keys, no accounts, no paid services and no company APIs. The chain week is 11 to 17 October 2026, 00:00 UTC on the first day to 00:00 UTC after the last. Packs bought in the week were followed forward to 21 October 2026 for their sell-backs.

Only the platforms’ own programs, contracts and treasury wallets are used. Wallet addresses of buyers and sellers were used in memory to join the legs of a pack and are not written anywhere. No individual wallet is shown or stored.

Endpoints

  • https://api.mainnet-beta.solana.com (archive, about one call a second)
  • https://api.mainnet.abs.xyz (Abstract)
  • https://polygon-bor-rpc.publicnode.com
  • https://rpc-mainnet.matic.quiknode.pro
  • https://rpc.mainnet.chain.robinhood.com (Robinhood Chain)
  • https://solana-rpc.publicnode.com (recent history)

How each figure is read, per platform, what the chain proves and does not, and which platforms publish certs: the method page. The monthly edition is the Platform Record.

Reading signed by BSJ Ventures, 23 October 2026.

Brich House records; it never buys, sells or holds, and never signs a verdict. BSJ Ventures, its parent, read the chains and signed the reading. Corrections are welcome at hello@brich.house.

The signed reading, PDF

Download

Each file carries the licence line: public data read by Brich House, the arithmetic our own, no licensed data source included, reuse with attribution to brich.house, not a verdict, not a forecast, not investment advice. The same data answers by API with a key: the data and API page.

This page describes the public record of each platform for the chain week 11 to 17 October 2026 (UTC), read on 21 and 22 October 2026. It is not a verdict on any platform or item, not a forecast and not investment advice. Platform and chain names belong to their owners, who do not endorse this page.