Scoring Our Own Test

21 October 2026. Figures and platform data are as of the dates stated in the text.

Interest disclosure: T-OpenAI and T-Kalshi are tokenized loan participation rights providing economic exposure linked to those companies' pre-IPO valuations, each issued through a dedicated issuing subsidiary. T-SpaceX, issued through its own dedicated issuing subsidiary, provides economic exposure linked to SpaceX; following the SpaceX listing, the investment underlying T-SpaceX is being divested. Redemption opens only once a Redemption Start Date is announced, and as of 21 October 2026 none has been. The argument this piece scores is one Tessera was built on, and Tessera has a commercial interest in it being right. Tessera also appears as a quote-asset category on StonkFun, which this piece discusses; Tessera has no commercial, integration or incentive arrangement with StonkFun or Raydium. Tessera has no arrangement with Kraken, Backed, Binance, Ondo, Superstate or Robinhood. Tessera's founder, and entities he controls, hold related exposure. Read what follows with that in mind.

Tessera's piece Something That Moves, first published on 28 September (republished here on 10 October), argued that "retail adoption of tokenized assets is arriving through convenience rather than through rights-first design." It ended with the evidence that would show it wrong. One of its two tests was this: "if RWA.xyz's holder count falls back as launchpad activity quiets, then it was a trade rather than adoption."

Three weeks of data later, the same sources give a first answer. It is not the one we would like to claim, and it says as much about the test as about the argument.

The Launchpad Went Quiet

StonkFun, the Solana launchpad that lets coins trade against tokenized stocks, was the part of the story we had not expected. DefiLlama counts every buy and sell on its launchpad's bonding curves, whatever they are priced in, so its figures cover more than the tokenized-stock pairs. Something That Moves called the same series all of StonkFun's trading; we have not established whether it counts trading in a coin after the coin leaves its curve.

DefiLlama's daily figures trace the decline:

• Peak: $63.0 million on 21 September, a week before the test was set.
• When the test was set: $24.6 million on 28 September.
• After it: the series did not quiet down straight away. It rose to $36.0 million on 29 September, and it has stayed below its level of 26 September, the last day the earlier piece reported, only since 7 October.
• Latest: $1.7 million on 20 October, about 93% below 28 September and 97% below the peak.
• By week: the seven days to 20 October came to $11.6 million, against $37.8 million in the seven days before and $148.9 million in the seven before that (our arithmetic from DefiLlama's daily figures).

That half of the test has been met, but for two weeks rather than the whole period.

The Holder Count Has Not Fallen Back

RWA.xyz counted 4,375,492 addresses holding tokenized stocks in its data of 20 October, against 4,010,763 on 28 September: 364,729 more, up 9.1% (our arithmetic). These are addresses rather than people, and an address can be a pool, a contract or a wallet an app created for its user. Every platform page cited on 28 September shows more holders now except one, on counts that cover all of each platform's products:

• bStocks: 1,548,922, up from 1,402,291.
• xStocks: 761,818, up from 666,831.
• Superstate Opening Bell: 83, down from 84.
• Robinhood and Ondo: 1,629,534, up from 1,525,585, and 557,640, up from 505,751. As on 28 September, this piece does not lean on them: Robinhood's count spans two products RWA.xyz classes differently, and Ondo's spans all its products, most of them Treasury funds.

Not every platform in the count is one StonkFun lists as a pairing. Of those listed here, xStocks, which StonkFun does list, has grown fastest since 28 September, by 14%, and it is also the only one whose count fell in the past week: from 842,888 on 14 October, down about 10% (our arithmetic). Nothing public says how many of its holders came through StonkFun.

So far, on the test as written, the argument survives. But that comparison starts before the quiet did. The fairer one starts inside it: from 14 October to RWA.xyz's data of 20 October, holders went from 4,387,867 to 4,375,492, down 0.3% (our arithmetic). That is less than the 2% set below as the smallest change this piece will count, so it does not count as a falling-back. But the count is no longer rising.

What the Test Could Not Tell Us

The argument survives a test that could hardly have failed this way, and that should be said.

A holder count is a net number. It can rise while every address that arrived in September leaves, if more arrive behind them, and it cannot tell a holder who is waiting from a balance nobody will touch again. RWA.xyz's thirty-day growth in holders fell from 67% on 28 September to 19% on 14 October, but that was mostly August's surge leaving the window. Per day, the count rose at about the same pace, roughly 23,000 addresses, in the fortnight before the test and in the sixteen days after it (our arithmetic from RWA.xyz's figures and rounded percentages). Then it stopped. It was 4,418,469 on 16 October and 4,375,492 four days later, down about 1%. That is inside the band set below for no change, so this piece does not read much into it. It is, though, what the start of a falling-back would look like, and it came in the second week of the quiet.

The launchpad is also small against what it was scored on. DefiLlama's all-time StonkFun total, about $0.9 billion across every quote asset, is less than a tenth of one month's tokenized-stock transfer volume on RWA.xyz. The two measures differ, but the gap is telling. Holders rose at much the same daily pace through the launchpad's busiest weeks and the first week of its quiet, and stopped rising only in the second. A launchpad this size could hardly move the count much either way, so the stall should not be read as its doing either. That is the main reason the test was easy to pass.

On RWA.xyz's tokenized-stocks summary, the page the 28 September holder figure came from, holders rose 15% in the thirty days to 20 October while monthly active addresses fell 53%, to 1,643,275, and monthly transfer volume fell 74%, to $10.28 billion. More addresses are holding, and far fewer are moving. Those thirty-day changes reach back before the test, and for transfer volume most of the fall came before it: Something That Moves reported it, from the 28 September reading, already "down 71% in thirty days, to $12.29 billion, while holders rose 67%." Since then it has fallen about 16% (our arithmetic, on thirty-day windows that overlap). And as the same piece said, "Transfer volume is not trading volume, and more holders with less money moving is consistent with small positions, but it is not proof of them." Small positions would explain less money moving. They would not explain fewer addresses moving. Active addresses have turned up in the past week, from 1,257,641 on 14 October to 1,643,275, up 31% (our arithmetic) on a rolling thirty-day count. A week does not undo a fall of more than half, but it is the first reading we have that points the other way.

RWA.xyz's platform pages do not show a fall on the same scale. On their figures for the tokenized-stock class, transfer volume on bStocks, xStocks, Robinhood and Ondo together came to about $38 billion in the data of 20 October, up about 4% in thirty days, and their active addresses together fell about 18% (our arithmetic; adding them up double-counts any address active on more than one, and it includes Robinhood and Ondo, which this piece does not lean on for holders). Those changes also reach back before the test. Four platforms alone come to more than three times the summary's total, and when we checked on 16 and 21 October, the bStocks page's own headline volume did not match its stock-class figure, though bStocks lists only stocks. We cannot reconcile the two views, so this piece does not lean on the platform pages for activity. The same doubt applies, more loosely, to their holder figures; the summary's are used because that is the page the test's figure came from.

On that view, then, over thirty days holding rose and movement fell, and in the past week holding stalled and movement picked up.

What Tessera Said Adoption Was

Tessera's piece Count Holders, Not Dollars, first published on 12 August (republished here on 5 September), said that demand-side adoption "shows up in holder count, and only in holder count." Something That Moves said: "So the case rests on holders rather than on volume." On that measure the count held, though it has stopped rising. We now think holder count alone is not enough, which is why the next test reports active addresses alongside it.

Something That Moves also gave its reading "that many holders with small positions value a token they can keep in the wallet they already use, and move at three in the morning, more in daily practice than rights they may rarely exercise." It offered StonkFun as "one of the places holders have taken it ...". With the summary's active addresses down by more than half over thirty days, the evidence that holders move these tokens is weaker than the evidence that they hold them, and that part of the case has less behind it than it did.

If these addresses are mostly retail, which we believe they are, the number holding did not fall in the first week the launchpad was quiet, and dipped only slightly in the second. Whether the same addresses stayed, this count cannot say. That fits the convenience argument better than the simplest version of "it was a trade", in which people arrive and then go. A trade that left balances behind would look much like this too.

The Other Side of the Ledger

Superstate's Opening Bell, the registered shares on-chain that Something That Moves used as the rights-first case, had 83 holders in RWA.xyz's data of 20 October and $49.8 million of value distributed through it, across three assets. As then, the comparison is not like-for-like.

The first of the two tests, whether rights-first venues draw holders at scale once they are live under the SEC's exemption, has not started. Tessera's piece Nine Said No, first published on 19 October (republished here on 20 October), was about the first venue we have found to have named, in a notice of 4 October, the stocks it intended to trade under that exemption. As of 21 October it is not yet trading: its page still says "OKXICE TSV is coming".

The Next Test

On the one test that can be scored yet, the argument survives. On the stronger question, whether these holders amount to adoption or to balances left behind, it is unproven. The test we set was easy to pass, so here is a harder one, fixed before the next look. On 21 November Tessera will score it again, on RWA.xyz's tokenized-stocks summary page only, against the readings in this piece, read on 21 October from RWA.xyz's data of 20 October: 4,375,492 holders and 1,643,275 monthly active addresses.

• Holders below that reading: it was a trade rather than adoption.
• Holders up, active addresses down: it fails on the stronger question, with balances left behind rather than use.
• Both up: it passes on both.
• No change: a change of less than 2% in holders, or less than 10% in monthly active addresses, will be reported as no change. Monthly active addresses is a rolling thirty-day count that moved about 10% in two days in October, and 31% in six, so a single reading is a noisy test.

This piece scores a test of an argument Tessera was built on, so Tessera is the least neutral party to mark it. That is why the figures are RWA.xyz's and DefiLlama's, and why this piece has said where they let the argument off lightly.

T-Tokens are not in RWA.xyz's tokenized-stock figures (checked 21 October). They are tokenized loan participation rights providing economic exposure, in the category next to tokenized stocks, so this is evidence about the category next to Tessera's rather than about Tessera's own. DefiLlama's StonkFun count covers every quote-asset category on the launchpad's menu and does not split them, so it is not a tokenized-stock figure, and this piece makes no claim about what it contains. The convenience argument is an argument for permissionless access, and "permissionless" describes the technology, not who may lawfully use it. Under Tessera's terms, T-Tokens are not available to persons in the US or other restricted territories, on any venue.

Sources: RWA.xyz tokenized stocks summary and platform pages for bStocks, xStocks, Robinhood, Ondo and Superstate Opening Bell, app.rwa.xyz, read 21 October 2026, data timestamped 20 October 2026 at 06:36 and 06:39 (RWA.xyz gives no time zone; the "UTC" given in Something That Moves was an assumption), including each platform's figures for the tokenized-stock class; the same pages read 14 October 2026 (data timestamped 14 October at 08:31 and 08:34) and 16 October 2026; the same pages as read on 28 September 2026 for Something That Moves (Opening Bell data of 27 September). DefiLlama StonkFun daily volume series and methodology, api.llama.fi, read 21 October 2026. StonkFun homepage, stonkfun.xyz, read 21 October 2026. OKXICE TSV page, okx.com/en-us/okxicetsv, read 21 October 2026. Tessera terms, tessera.pe/terms, read 21 October 2026. Tessera, "Count Holders, Not Dollars" (first published 12 August 2026, republished at blog.tessera.pe on 5 September 2026), "Something That Moves" (first published 28 September 2026, republished at blog.tessera.pe on 10 October 2026) and "Nine Said No" (first published 19 October 2026, republished at blog.tessera.pe on 20 October 2026), quoted as published at blog.tessera.pe.

This is market commentary, not investment advice. It is not a recommendation to acquire, hold or redeem any Tessera product, to use any platform mentioned, or to take or avoid exposure to any asset mentioned.

T-Tokens are tokenized loan participation rights, not equity. High risk. DYOR. Not financial advice. Not available in the US or other restricted territories. tessera.pe/terms