This Channel Ranks the Mag7 Backwards, and Its Real Mag7 Exposure Is Not in These Five

TL;DR

In this channel the five Mag7 instruments are ranked the exact reverse of public markets: TSLAB alone is larger than the other four combined. But most of this channel's real Mag7 exposure is not in these five.

• TSLAB's market cap of $23.68M is 12.7% more than AAPLB ($5.45M) plus MSFTB ($4.37M) plus AMZNB ($3.73M) plus METAB ($7.46M) combined. It is 53.0% of the group, 5.42x MSFTB and 4.35x AAPLB.
• The book's 30-day median across 100 instruments is −16.71% with 82 negative — and all five here beat the median, three outright positive.
• Of the 18 instruments on the book that are up, three of the four largest by market cap are in this group: TSLAB $23.68M (1st), AAPLB $5.45M (2nd), MSFTB $4.37M (4th). The other 14 positive instruments are all under $2.50M.
• But 76.9% of this channel's Mag7 exposure is not in these five. The seven Mag7 names on the book come to $193.49M, and GOOGLB $104.99M (3rd on the book) plus NVDAB $43.80M (8th) are 76.9% of it. Neither is in this group.
• QQQB alone is 2.20x all five combined. At $98.22M it ranks 5th on the book and listed the same day as MSFTB and METAB (2026-07-02). So "this channel does not want Apple" is wrong — it just does not want it one name at a time.
• Counter-example: TSLAB is the smallest instrument in its own listing cohort. Six listed on 2026-06-11 and TSLAB ranks 6th of 6, while NVDAB from the same day is 1.85x its size.
• Market cap and turnover run backwards inside the group. The largest, TSLAB, turns 0.548x (67th on the book); the smallest, AMZNB, turns 1.617x (26th). Drop TSLAB and group turnover rises from 0.7812x to 1.0437x.

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.

This group is the five Mag7 instruments in this channel. The other two names are also on the book, but not in this group: GOOGLB sits in group one and NVDAB in the semiconductor group. Section 4 turns that into the central constraint on this piece.

All five are common stock, with no leveraged, inverse or ETF structure. Group market cap is $44.69M, 3.4% of the book; volume is $34.92M, 4.0%. The volume share exceeds the market-cap share — the exact opposite of the crypto treasury group.

2. TSLAB alone is larger than the other four combined

TSLAB is 12.7% larger than the other four combined. Pairwise: it is 5.42x MSFTB, 4.35x AAPLB, 6.34x AMZNB and 3.17x METAB.

That ordering is the reverse of public markets. In public markets Apple and Microsoft have long been the two largest companies in the world; in this channel they are third and fourth of these five, together $9.82M, under 41.5% of TSLAB.

One thing has to be nailed down first: none of this can be read as "token market cap reflects the underlying company's market cap," in any version. The market cap here is the CMC measure of the token (token price times circulating supply) and has no mechanical relationship to the underlying company's — the snapshot carries neither a conversion ratio nor an underlying reference price. What this group compares is how large each name's token float is in this channel, a result of how much the issuer minted and how much buyers hold, not a valuation of the underlying company.

Another thing to watch: market cap and turnover run backwards inside this group.

The largest one turns slowest. Take TSLAB out and the remaining four turn 1.0437x, a third higher than the group's 0.7812x and far above the book's 0.6625x. So everything about this group being hotter than the book comes from the four small members; TSLAB is pulling it down.

3. Three of them rose while 82 instruments on the book fell

The book's 30-day median across 100 instruments is −16.71% with 82 negative — and all five here beat the median. To be precise about it, METAB beats it by only 0.7 percentage points, so it sits essentially on the median; the real distance is in the first three.

Lay the 18 positive instruments out by market cap and this group's position stands out:

Three of the four largest positive instruments are in this group. The one in third place, SOXSB, is the −3x semiconductor inverse ETF from the directional 3X group — it is up 70.82% because semiconductors fell, which is a completely different thing. The remaining 14 positive instruments are all under $2.50M.

That is the one conclusion in this group that holds up on size: in a month when 82% of the book fell, almost the only things that both rose and have any scale are these three Mag7 names. The 18 positive instruments add up to $52.52M in market cap, and TSLAB plus AAPLB plus MSFTB alone are $33.50M, or 63.8% of it.

4. Counter-examples: most of this channel's Mag7 exposure is not in these five

This group's argument — that this channel's preferences are the reverse of public markets — holds among the five members, but widening the view to the whole book cuts it down in three steps.

Counter-example one: QQQB alone is 2.20x all five combined. MSFTB and METAB listed 2026-07-02, a day that brought five instruments:

QQQB is the Nasdaq-100 index ETF, and this single instrument at $98.22M is 2.20x the group's combined $44.69M, ranking 5th on the book; within its own cohort it is 13.2x METAB and 22.5x MSFTB. Add the S&P 500's SPYB ($19.08M, 15th on the book) and the two broad index products come to $117.30M, 2.62x this group.

QQQB is one of the five flagship instruments in group one, which already wrote it up as a pricing anchor; this piece does not repeat that role and uses it only to bound this group's claim: this channel does not refuse these companies, it refuses to take them one name at a time. To be clear about the measure: the snapshot contains no index constituent or weight data, so this piece cannot quantify how much of QQQB is Mag7. All it can say is that QQQ and SPY are broad index products and that their token float in this channel is far larger than any single name's.

Counter-example two: GOOGLB and NVDAB hold 76.9% of this channel's Mag7 exposure, and neither is in this group.

The seven Mag7 names on the book come to $193.49M, 14.68% of the book's market cap; GOOGLB plus NVDAB are $148.79M, or 76.9% of the Mag7, leaving 23.1% for these five. So "the cold half of the Mag7" is accurate on size — but the way it is accurate is not "this channel does not want the Mag7," it is "this channel's Mag7 is concentrated in the GOOGL and NVDA names." GOOGLB by itself is 135% more than all five here combined.

Counter-example three: TSLAB is the smallest instrument in its own listing cohort. Six instruments listed 2026-06-11:

TSLAB ranks last of these six on market cap. NVDAB, listed the same day over the same 142-day window, is 1.85x its size, and SPCXB is 7.35x. So "TSLAB is large" is purely a within-group statement — it dominates this group because the other four are small, not because it is large on this book.

The same holds for AAPLB. AAPLB and AMZNB listed 2026-07-28, a day that brought ten instruments:

AAPLB is not the largest in its own cohort — AMATB is 65.7% larger. AMATB is a semiconductor equipment maker, and its turnover of 0.130x is the lowest of all 100 instruments on the book (100th), so its size is not built out of trading interest, which in turn shows that measuring preference by token float is a coarse step to begin with. SOXSB's $5.35M is also 98.2% of AAPLB — a −3x semiconductor inverse ETF and Apple are nearly the same size in this channel.

The one AAPLB claim in this section that fully survives is about path, not size: AAPLB's range width of 1.189x is the narrowest of these ten, while its position in range of 78.1% and its −3.5% from high are the highest of the ten. Over the same 95-day window it is both the one that travelled least and the one closest to its own high. That one is clean, because cohort windows are identical.

5. Putting this group back in the distribution

First, this group's turnover ranks third among the seven groups published in this series, and above the book.

This group turns 2.08x faster than group one while carrying only 7.1% of its size. Group one is the coldest because it is built out of a few very large, very slow instruments (SPCXB 0.458x, CRCLB 0.326x, QQQB 0.327x) — in other words, size and turnover also run backwards at the group level on this book, so the inversion inside this group (section 2) is not something peculiar to it.

Second, the Mag7 is the calmest stretch of path on this book. Among the 46 instruments live 100 days or more, the 15 narrowest range widths:

Five Mag7 names have been live 100 days or more (NVDAB 142d, GOOGLB 117d, TSLAB 142d, METAB 121d, MSFTB 121d), and all five land in the narrowest 15. AAPLB and AMZNB at 95 days do not enter this comparison by the stated measure. The common-stock median width among these 46 is 1.768x, and TSLAB 1.472x, METAB 1.498x and MSFTB 1.535x are all below it.

There is an inversion worth noting here: TSLAB is the single name this channel holds most of, yet its range is narrower than METAB's and MSFTB's, which it dwarfs in size. In the public-market narrative Tesla is usually cast as the most violent of the Mag7; over this snapshot's 142-day window it is not.

Third, the top two are the two broad index ETFs. This is the same fact as section 4 measured a different way: what this channel holds most of is also what travels flattest — SPYB and QQQB hold the 15th and 5th largest market caps on the book while being the two narrowest of the 46.

Fourth, the size distribution. The book has 7 instruments at $50M or more, 12 at $25M or more, 61 under $5M, 33 under $1M, with a median of $2.39M. Not one of these five reaches the $25M bucket — TSLAB ranks 13th on market cap, just behind those 12 — and the other four sit between $3.73M and $7.46M. All five are above the median of $2.39M while all belonging to the smaller half.

6. Structure

All of this comes from this site's issuer record, 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.

Two of those have concrete meaning for this group.

Voting rights are none, and shareholder votes at these five companies carry real content in public markets. Tesla compensation packages, Meta dual-class share structure, Apple and Amazon shareholder proposals are all matters genuinely voted on each year in public markets. What the token holder receives is price exposure, without that layer. This is not a defect peculiar to this channel — it is the definition of the tracker-certificate wrapper — but in the Mag7 group what is given up is more concrete than it would be in a small-cap name.

Dividends are "30% withholding, then reinvested via a multiplier," and some of these five pay. Apple and Microsoft pay quarterly in public markets. Passing a dividend through, then withholding 30%, then reinvesting at a multiplier this piece does not know, means the token's unit price path and the underlying share's unit path are not the same path — one of the reasons section 8 gives no premium figure.

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 five names have.

This group deserves the reminder especially, because its range positions cluster near the top: AAPLB 78.1%, MSFTB 80.2%, TSLAB 76.4%, sitting only −3.5%, −6.9% and −7.6% from their highs respectively. "Only 3.5% below its all-time high" is a sentence with no cash-market quote behind it when you read it on a weekend.

The same issuer has already provided a quantified version: 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. That piece and this one share AAPLB as an instrument and overlap in window, so its conclusion applies directly to AAPLB's +3.20% in section 3.

The 24-hour changes in this piece (AMZNB +5.82%, TSLAB +1.25%) 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 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, and $345.04 cannot be read alongside any Apple quote.

Second, it cannot compute shares per token. The snapshot has no conversion ratio. TSLAB has 59,212 tokens circulating at $399.97 and AAPLB has 15,781 at $345.04 — without the ratio those numbers cannot be converted into any share count, nor used to compare how much underlying asset each instrument represents. Every comparison in section 2 is a comparison between token floats.

Third, it cannot compute index constituents. Section 4 uses QQQB and SPYB to cut down this group's claim, but the snapshot has no index constituent or weight field. This piece says only that QQQ and SPY are broad index products and that their token float is far larger than any single name's; it does not and cannot quantify how much of those two ETFs is exposure to these five names. That layer requires the funds' own published holdings.

Fourth, the range-width windows are unequal. The five are 142 / 121 / 95 / 121 / 95 days, and AAPLB and AMZNB at 95 days do not enter the comparison across the 46 instruments live 100 days or more. So section 5's width ranking contains only TSLAB, METAB and MSFTB; AAPLB's 1.189x is comparable only inside its own 95-day cohort (the table in section 4) and cannot be set directly against SPYB's 1.074x.

Fifth, 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.

Sixth, 100 of 105. CMC reports 105 instruments and the snapshot in hand holds 100; the missing 5 appear in no statistic here. So the accurate phrasing of "the seven Mag7 names" is "the seven Mag7 names among these 100 rows" — if any of the missing 5 is relevant, the 76.9% in section 4 changes.

Seventh, turnover is 24h volume divided by market cap. A small denominator inflates the ratio. These five span $3.73M to $23.68M in market cap, a factor of 6.3, so part of the "market cap and turnover run backwards" result in section 2 is a denominator effect and must not be read entirely as a difference in trading interest. AMZNB's 1.617x sits on $6.04M of absolute volume against TSLAB's 0.548x on $12.99M — in absolute volume TSLAB is still the largest in the group.

Eighth, one 30-day window is not a conclusion. The "three of them rose" in section 3 is the result of roughly 2026-10-01 to 10-31. The 30-day spread across these five runs +11.78% to −16.00%, a gap of 27.8 percentage points — five names that all belong to the Mag7 separated by nearly 28 points in a single month, which by itself says no common Mag7 factor was operating in this window.

Conclusion. In this channel the five Mag7 instruments are ranked the reverse of public markets: TSLAB at $23.68M is 12.7% larger than AAPLB plus MSFTB plus AMZNB plus METAB combined, and 5.42x MSFTB. And in a month when 82 instruments fell and the median was −16.71%, all five beat the median, three are positive, and three of the four largest of the book's 18 positive instruments are in this group — the other 14 are all under $2.50M. That is the one conclusion here that holds up on size.

But the argument has to stop at the boundary of the five members. 76.9% of this channel's Mag7 exposure sits in GOOGLB ($104.99M, 3rd on the book) and NVDAB ($43.80M, 8th), neither of which is in this group, and GOOGLB by itself is 135% more than all five combined; while QQQB alone at $98.22M is 2.20x the five. So "this channel does not want Apple" is wrong — it just does not want it one name at a time.

Two counter-examples narrow the claim further. TSLAB ranks last on market cap among the six instruments in its own listing cohort, with NVDAB from the same day 1.85x its size, so "TSLAB is large" is only a within-group statement. AAPLB is not the largest in its own ten-instrument cohort either — AMATB is 65.7% larger, and AMATB's turnover is the lowest of all 100 on the book, which shows that token float is a coarse proxy for preference to begin with. What does survive for AAPLB is path: of the ten in its cohort it has the narrowest range (1.189x), the highest position in range (78.1%) and the smallest distance from its high (−3.5%).

One last thing to read together with all of it: the Mag7 is the calmest stretch of path on this book. The 15 narrowest range widths among the 46 instruments live 100 days or more contain every one of the five Mag7 names that reach that window, and the two narrowest of all are SPYB and QQQB. What this channel holds most of is also what travels flattest.

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.