An Exchange Tokenizes Crypto Treasury Companies: Four Layers of the Same Exposure, and the Coldest Group on the Book

TL;DR

An exchange issues equity tokens for companies whose main asset is cryptocurrency, and the largest of them holds the exchange's own coin. Yet this group is not hot — it is the coldest group by turnover in this series so far.

• BNCB alone is 76.3% of this group's market cap ($37.38M against $48.98M for all five) and 70.4% of its volume. It listed 2026-09-14, has been live only 47 days, and is already the 10th largest instrument on the book — by far the youngest name in the top 15.
• It is 68.3% of its own listing month. September 2026 brought 17 instruments totalling $54.71M; BNCB by itself is $37.38M.
• Four layers of exposure stack on one token. The snapshot gives BNCB the CMC slug bnb-standard-corporation-tokenized-bstocks: the underlying company is a BNB treasury, the issuer is Binance's own BTech Holdings, the only venue is Binance, and the only supported chain is BNB Chain. These four layers are not inferred — they are four fields in the data.
• And the group is cold. Group turnover is 0.4682x against 0.6700x for the other 95 — the rest of the book turns 1.43x faster than this group. Both large members sit at the back of the turnover ranking: BNCB 78th, BMNRB 88th.
• Counter-example one: BNCB is not the worst performer. Its 30-day −14.61% beats the book median of −16.71% and is the second best of these five. If stacked exposure were the dominant risk here, BNCB should have been the worst.
• Counter-example two: these five are not one factor. Over 7 days CYPHB is +7.21% while the other four fell 4.86% to 17.17%; the 30-day spread runs −11.91% to −39.32%, a gap of 27.4 percentage points.
• Four of the five carry no underlying symbol in the snapshot, two are unclassified by kind, and two have derived company names. Reference-data quality in this group is visibly below the book average.

1. What these five are

bStocks is Binance's own on-chain equity issuer (entity BTech Holdings Limited, ADGM / FSRA listed certificates, single chain BNB Chain, Binance only). CoinMarketCap reports 105 instruments; this piece uses a snapshot covering 100 of them, timestamped 2026-10-31T06:06:52Z.

What these five share: the underlying company's main asset is cryptocurrency rather than an operating business. Public markets usually label such companies treasury vehicles or coin holding shells. This site groups them because in this channel they form a particular structure — section 6 takes it apart.

Group market cap is $48.98M, 3.72% of the book; volume is $22.93M, 2.63%. The volume share is below the market-cap share.

The data quality here needs stating up front, because it is worse than in other groups:

• Four of the five carry no underlying symbol field in the snapshot (BNCB, CYPHB, USDEB, FWDIB). Of 100 instruments on the book, 17 lack a symbol — this group holds 4 of them.
• Two are unclassified by kind (CYPHB, USDEB). The book has 8 unclassified instruments; this group holds 2.
• Two have company names flagged as derived in the metadata (nameSource is derived for CYPHB and USDEB). The book has 6 such instruments.
• All five listed after 2026-08-04, so not one of them belongs to the two large June and July cohorts.

This is not a data error, it is evidence of a lag: the exchange listed these faster than the reference data could classify them. A sector tracking companies whose main asset is cryptocurrency happens to be exactly where the public-market classification system falls furthest behind.

2. One instrument is three quarters of the group

BNCB is 76.3% of this group's market cap. Its $37.38M ranks 10th on the book, and it has been live only 47 days — every other name in the top 15 came from June or July, making it the only September listing that gets anywhere near. The five flagship instruments in the first piece in this series all listed in June or July.

Grouped by listing month, the point is sharper:

September 2026 brought 17 instruments totalling $54.71M, and BNCB alone is 68.3% of that. Two thirds of the value listed that month sits in this one token.

The other four live on a different scale entirely: BMNRB $9.68M ranks 24th, CYPHB $911.4K ranks 73rd, USDEB $518.7K ranks 89th, FWDIB $501.2K ranks 92nd. The last three are all below the book median market cap of $2.39M, and together they come to $1.93M, under 5.2% of BNCB.

3. The coldest group on the book

The most counterintuitive thing here: a crypto exchange tokenized crypto treasury companies, and the resulting tokens trade less than the book average.

The other 95 instruments turn 1.43x faster than this group. The two large members drag hardest:

The two instruments carrying 96.1% of the group's market cap rank 78th and 88th on turnover. The three ahead of them add up to $1.93M in market cap, and their high turnover is mostly a denominator effect — the previous piece quantified that once and this one does not repeat it, beyond noting that USDEB's 2.681x sits on $1.39M of absolute volume, so "the float turns 2.7 times a day" and "the float is tiny" are two readings of the same fact.

There is one more layer: all five listed after August, and instruments listed after August turn faster than the book overall. The September cohort turns 1.075x and the October cohort 1.143x, both well above the book's 0.6625x. BNCB's 0.432x is 40% of its own listing month's level. It is not just colder than the book, it is colder than the things listed alongside it.

4. Counter-examples: these five are not one factor, and BNCB is not the worst performer

This group carries a structural concern (section 6), and the two conclusions most easily drawn from it both fail.

Counter-example one: BNCB beat the book median and ranks second best within its own group.

If four layers of the same exposure were the dominant risk in this group, BNCB should have fallen hardest. It did not. It is second best of the five and ahead of the book median — 82 of the 100 instruments are negative over 30 days, and BNCB is lighter than most of them. The structural concern is real and nameable, but this snapshot does not show it converting into worse price outcomes. Both statements have to be made together.

Counter-example two: these five are not five expressions of one factor.

Over 7 days CYPHB gained 7.21% while the other four lost 4.86% to 17.17%. The 30-day spread runs −11.91% to −39.32%, a gap of 27.4 percentage points. Five proxies for one crypto-asset exposure would not separate like this. "Holds crypto on the balance sheet" is an accounting feature, not a shared price factor.

Counter-example three: these five are not the most violent either. BMNRB listed 2026-08-04 alongside nine others; ranked by range width:

BMNRB is a bitcoin mining and treasury company, and among these ten its range width ranks only 6th — narrower than a rare-earth miner (USARB 2.313x), a satellite communications company (ASTSB 2.025x) and an optical components maker (COHRB 1.952x).

More to the point, IRENB is also a crypto mining company, listed the same day over the same 88-day window, with a range width of 2.133x — 27% wider than BMNRB — and a 30-day decline of −42.42%, 17.4 percentage points deeper than BMNRB's −25.07%. So "crypto-linked means more volatile" fails inside the crypto subset itself: two crypto companies that listed on the same day differ by 27% on width.

USDEB's cohort says the same thing. Four instruments listed 2026-10-07:

SECZB's underlying is Securitize, a tokenization infrastructure company — also crypto-linked — and its 30-day −38.92% is nearly identical to USDEB's −39.32%, yet its range width of 2.092x is 22% wider than USDEB's 1.711x. Over the same 24-day window, two crypto-linked instruments fell by almost the same amount along paths differing in width by a fifth.

5. Putting this group back in the distribution

First, this group is one of the main clusters of unclassified instruments on the book. Eight of the 100 have kind unclassified:

Two of the eight are in this group. What these eight share is not an industry but the fact that they are all things the public-market classification system has not yet digested: unlisted or newly listed AI compute companies, a quantum computing company, tokenization infrastructure, coin holding shells. Among the 46 instruments live 100 days or more, only 2 are unclassified, with a median width of 2.389x — the second widest of any kind, behind leveraged ETFs at 2.892x.

Second, this group fell harder than the book this month. The book's 30-day median is −16.71% with 82 instruments negative. All five here are negative and three are deeper than the median. Weighting the five by market cap gives BNCB's −14.61% a 76.3% weight, so the weighted result is better than the equal-weighted one — a measurement difference worth noting: the equal-weighted average is −21.95% while the median is −18.83%.

Third, the kind composition. The book's 100 instruments break down as 67 common stock, 8 unclassified, 6 ADR, 6 ETF, 10 leveraged ETF, 2 inverse ETF and 1 pre-IPO. This group is three common stocks plus two unclassified — no leverage, no inverse, no ETF at all. Everything distinctive about this group comes from the nature of the underlying companies and from when they listed, not from product structure.

6. Structure: four layers of the same exposure

All of this comes from this site's issuer record and the snapshot metadata, the same data behind /stocks/bstocks:

• Legal wrapper: tracker certificate, ADGM / FSRA listed certificates, issuing entity BTech Holdings Limited. The holder receives a certificate tracking the underlying price, not a share registration.
• Voting rights: none.
• Dividends: passed through, but after 30% US withholding, then reinvested via a multiplier.
• Redemption: supported, KYC required, $5 minimum.
• Proof of reserve: attestation, not live on-chain proof.
• US persons: blocked.
• Chain: BNB Chain only. BSC deposits only; sending on the wrong network loses the assets.
• Issuance model: exchange self-issued; the only venue is Binance, operated by the issuer.
• Grade on this site: B.

Put those fields next to BNCB and a structure appears that is worth writing out.

In the snapshot BNCB's CMC slug is bnb-standard-corporation-tokenized-bstocks, while its underlying.name field reads "CEA Industries Inc." Two fields naming the same entity two ways, and the slug side says BNB explicitly. That is the basis for grouping it as a treasury company here.

So someone buying BNCB on Binance holds all four of the following at once:

Layer 1  Underlying company   main asset is BNB
Layer 2  Legal wrapper        tracker certificate issued by BTech Holdings Limited (Binance owned)
Layer 3  Settlement network   BNB Chain (the only supported chain)
Layer 4  Trading venue        Binance (the only venue, operated by the issuer)

All four layers are written in the data, not inferred. issuanceModel is exchange-self, venues is a single { Binance, operatedByIssuer: true }, chains is a single BNB Chain, and the nature of the underlying company comes from that slug.

What is worth spelling out is the concrete consequence of this structure, rather than attaching a label to it:

The redemption path never leaves this set. Redemption converts the token back into the certificate (KYC, $5 minimum), and the certificate's issuer and the trading venue belong to the same group. If the underlying exposure goes wrong, the redemption exit sits inside the same structure as the exposure that went wrong. That is the difference between having a redemption channel and having an independent one.

Proof of reserve is an attestation, not live on-chain proof. This site records proofOfReserve.kind as attestation. For comparison, xStocks uses Chainlink proof of reserve, updated daily or whenever reserves move more than 10%. In a structure where issuance, settlement and trading all belong to one company, the third-party-verifiable link happens to be the weakest one.

Voting rights are none, so holders have no channel at all on the underlying company's treasury decisions. For a company whose main asset is cryptocurrency, the most consequential decisions are exactly those about buying, selling and staking those coins — and that layer is entirely closed to token holders.

Finally, to connect back to section 4: all of the above is structural description, not a price forecast. In this snapshot BNCB's 30-day decline of −14.61% beats the book median of −16.71%, and the stacked exposure has so far not shown up as a worse outcome. Structural risk is about what happens if something goes wrong, not a claim that something is going wrong now.

7. This snapshot was taken on a Saturday

The timestamp 2026-10-31T06:06:52Z is Saturday morning UTC. The US cash market closed Friday at 20:00 UTC and does not reopen until Monday — for more than two days after this snapshot, these token prices are the only prices these names have.

That carries a particular meaning for this group: the underlying companies' main assets keep trading over the weekend. Crypto markets never close, so a coin holding company's net assets continue moving on Saturday and Sunday while its stock does not. The token-side quote is, for those two days, the only place that can reflect both "the crypto asset is moving" and "the stock cannot trade" — but this snapshot is a single moment and cannot see that process.

The same issuer has already provided a quantified version of this: one stock, four books laid AAPL's four tokenized versions side by side on a single Saturday and concluded that a quote during non-trading hours is a market maker's opinion, not a market price.

The 24-hour changes in this piece (USDEB −7.28%, FWDIB −4.47%) cover Friday 06:06 UTC to Saturday 06:06 UTC, which includes a full Friday US trading session, so they are not purely closed-market quotes.

8. What this method cannot compute

First, it cannot tell you how much crypto these companies hold. The snapshot has only token prices and market caps — no balance sheet for any underlying company. So this piece cannot say how much BNB corresponds to BNCB, nor compute a discount to net asset value. The metric most often discussed for these companies in public markets, the ratio of share price to crypto net asset value per share, is entirely unsupported by this data.

Second, the "crypto treasury" classification comes from this site, not from the issuer. What these five share is this site's own grouping judgement; the snapshot supplies only company name, kind and the CMC slug. BNCB's BNB nature has data support — that slug — while the other four have no crypto-asset category in the data: the snapshot does not say BMNRB holds bitcoin, or which stablecoin USDEB relates to. Readers who need that layer have to go to each company's public disclosures.

Third, it cannot compute premium or discount. The snapshot carries only the token's own price, with no reference price for the underlying stock. More fundamentally, the bStocks dividend mechanism is "30% US withholding, then reinvested via a multiplier" — without knowing the current multiplier, the token price and the underlying share price are not comparable in the first place. So no premium figure appears anywhere in this piece.

Fourth, it cannot compute shares per token. The snapshot has no conversion ratio. BNCB has 7,364,066 tokens circulating at $5.08, and without the ratio those two numbers cannot be converted into any share count.

Fifth, the range-width windows are badly unequal, especially here. The five are 47 / 88 / 39 / 24 / 32 days, and not one of them has been live 100 days. So none of the five enters the comparison across the 46 instruments live 100 days or more, and their widths (1.635x / 1.680x / 1.835x / 1.711x / 1.419x) are not comparable with each other either — BMNRB at 88 days has three times as long as USDEB to reach an extreme. The two same-day cohort tables in section 4 work precisely because cohort windows are identical.

Sixth, market cap is the CMC measure (price times circulating supply), and the book total of $1,318.38M is the sum of these 100 rows; the bStocks line on this site's /stocks page shows an on-chain size of $914M — two numbers from different sources at different moments that will not tie. This piece uses the former throughout.

Seventh, 100 of 105. CMC reports 105 instruments and the snapshot in hand holds 100; the missing 5 appear in no statistic here.

Eighth, turnover is 24h volume divided by market cap. A small denominator inflates the ratio, so USDEB's 2.681x (10th on the book) and FWDIB's 2.620x (12th) are first of all the result of $518.7K and $501.2K market caps, and must not be read as these two being the most liquid in the group.

Conclusion. This is a set of equity tokens an exchange issued for companies whose main asset is cryptocurrency, and the largest of them tracks a company holding that exchange's own coin. Four layers of exposure stack on one token — the underlying company holds BNB, the certificate is issued by a Binance-owned entity, the only settlement network is BNB Chain, and the only venue is Binance — and all four are written in four fields of the data. The concrete consequences: the redemption exit sits inside the same structure as the exposure it would exit, and the third-party-verifiable link, proof of reserve, happens to be an attestation rather than live on-chain proof.

But the numbers do not cooperate with any dramatized version of that story. This is the coldest group by turnover in this series so far: 0.4682x against 0.6700x for the other 95, with BNCB and BMNRB — which carry 96.1% of the group's market cap — ranking 78th and 88th on turnover. And BNCB's 30-day decline of −14.61% beats the book median of −16.71%, second best within its own group.

Three counter-examples belong together. These five are not one factor: over 7 days CYPHB is +7.21% while the other four lost 4.86% to 17.17%, and the 30-day spread is 27.4 percentage points. They are not the most violent either: BMNRB ranks only 6th of ten on width in its own cohort, and IRENB, another crypto mining company that listed the same day, is 27% wider and 17.4 percentage points deeper. The structural risk is nameable, but it has not yet appeared in the prices — both sentences are required, and dropping either one distorts the picture.

Data from CoinMarketCap, timestamped 2026-10-31T06:06:52Z, covering 100 of the 105 instruments bStocks has reported. Instrument details and quotes at /stocks/bstocks. This is research notes, not investment advice. DYOR.