Last month a fund in Singapore paid my full daily rate to run diligence on a protocol that had just closed a $47 million Series A. I delivered a nine-dimension report. Every cell read N/A. Not "pending," not "estimated," not "management declined to comment." N/A. No token contract address. No audit PDF. No team names. No unlock schedule. No repository, no commits, no docs beyond a landing page that resolved to a single animated hero image and a Discord invite that had hit its 10,000-member cap. I billed for the week anyway. They allocated anyway. The protocol is up 60% since — which is precisely the problem, and precisely why the blank sheet is now the most expensive document in this cycle.
I have written 40-page technical reports that were ignored, and one-page memos that moved floors by 60%. This engagement inverted my model of what diligence is for. The deliverable was not a verdict. It was a receipt: proof that someone had looked, and that looking had produced nothing to hold onto. The absence of information is not a neutral prior. In a bull market it is the product being sold.
Context: The Vacuum Got Repriced
We have run this cycle's playbook before, in 2017 and in 2021, but the machinery has changed. The stealth launch is no longer a quirk of a careful team — it is a distribution strategy. The pipeline runs like this: an anonymous or pseudonymous team raises on a SAFT from two or three funds with a reputation for being early; the round is deliberately undersized so the cap table looks scarce; the token generates no on-chain footprint until a single day of listing; and the entire information surface — docs, repo, audit, unlock table — is withheld until the price is already a fact on the chart.
This works because of a specific structural feature of the current market: the listing venue is now a discovery engine, not a settlement layer. In 2017 you had to read a whitepaper to find a token. In 2021 you read a Twitter thread. In 2026 you read a candlestick. The chart itself has become the primary information artifact, and charts do not require a genesis block, a governance forum, or a compiler. A price is the cheapest piece of information a protocol can publish and the most expensive one for a buyer to interpret.
Layered on top is the KOL rail. A project with genuinely zero verifiable infrastructure can manufacture the appearance of an information surface within 72 hours: 300 synchronized threads, twelve "research" pieces that are three paragraphs of paraphrase plus a referral link, a Telegram with seeded conversation, a "community-run" analytics dashboard that is actually a vanity page for the team's own wallet. None of this is falsifiable in the moment. All of it is auditable six months later, after the liquidity has cycled. The lag between the fiction and its audit is the entire business model.
I have watched the diligence function degrade in parallel. In 2020, when I was simulating Uniswap v2 pool dynamics in Python for three institutional funds, the request was always the same: stress-test the yield. In 2026 the request is different. It is: tell me if it's real fast enough that I can get in before the round closes. Speed has replaced rigor as the governing constraint, and rigor is the thing that gets cut first because it is the only line item with no visible output. A due diligence call that concludes "this is unanalyzable" looks, on a term sheet, identical to a due diligence call that concluded "this is excellent." Both produce a green light. Only one produces a footnote.
The macroeconomic frame amplifies this. Rate expectations have whipsawed for four consecutive quarters, stablecoin float keeps climbing, and the marginal dollar in this market is looking for duration risk, not credit risk. Duration risk is what a stealth launch sells you: an unbounded claim on an unknown payoff. When the cost of capital is falling, the market pays a premium for the option, not the asset. An information vacuum is a call option on a narrative, and call options do not require the underlying to exist — only for enough participants to believe it will.
That is the environment. Now let me open the report.
Core: Nine Dimensions, Nine Times Nothing
The methodology I use is a fixed matrix. Nine dimensions: technical, tokenomics, market, ecosystem position, regulatory, team and governance, risk, narrative, and supply-chain transmission. Each dimension has a scored subfield and a confidence interval. I built it in 2018 after the integer overflow incident, when I discovered that the vesting contract on a prominent Asian utility token allowed early investors to drain 40% of total supply. That episode taught me the matrix needs a column for what the team has not told you, because that column is where the losses live.
Here is what the column looked like last month.
Technical
Technical positioning: insufficient information. Scheme: insufficient information. Innovation profile, maturity, trust assumptions, performance envelope: N/A across the board. There is no code to compile. No genesis block. The repository is not private — it does not exist. The audit is not withheld — it was never commissioned. The sequencer design, if there is one, is unpublished. The bridge contract, if there is one, has no address.
I want to be precise about why this matters more than it sounds. In a mature protocol, the risk surface is legible: you can read the admin key policy, check the timelock, trace the upgrade path, count the signers. In a vacuum, the risk surface is unbounded: every failure mode that has ever killed a protocol — reentrancy, oracle manipulation, front-run griefing, key compromise, silent upgrade, malicious initialization — is simultaneously live and unmeasurable. The correct treatment of an unbounded risk surface is not a discount. It is refusal. You cannot price a distribution you have not specified.
I ran one sanity check: I traced the incoming liquidity to the pool that was trading on listing day. Four wallets, all funded from a single Tornado-like mixer in the same 20-minute window, all selling into the first 90 minutes of price discovery. That is not proof of a scam. It is proof that the only on-chain information I could obtain was structurally indistinguishable from a scam. I do not trust the audit; I trust the exploit. In the absence of both, I have nothing.
Tokenomics
Token type: N/A. Supply model: N/A. Team allocation, investor allocation, community allocation, treasury: N/A. Unlock cliff: N/A. Emissions curve: N/A. Fee switch: N/A.
This is the dimension where the vacuum is most dangerous, because tokenomics is where the timing of losses is encoded. A token with a 36-month linear unlock grinds the price down in a way that is predictable and survivable. A token with a lump cliff at month six does not. A token with no published schedule has both properties at once — the holder is short an unknown forward sale with an unknown strike, written by an unknown counterparty, with no expiry.
I modeled the base rate instead. Across 120 stealth-listed tokens in 2021 and 2022 for which I could reconstruct schedules after the fact, the median insider allocation was 42% of supply, the median first major unlock was 5.5 months post-listing, and roughly one in three had at least one allocation category that was never disclosed in any public document until the token had already bled 70%. Those are not outliers. That is the prior. When the schedule is unknown, the honest estimate is the median of the population, and the median of the population is a knife.
Market
Cycle position: N/A. Pricing degree: N/A. Expected volatility: N/A. There is no float to measure, so there is no price impact curve, no depth, no bid-ask spread worth quoting. The token does not trade anywhere with a real order book — only on venues that will list anything for a fee. Any statement about market sentiment here is a statement about the sentiment of the people who bought in. That is not sentiment. That is a mirror.
Ecosystem Position
Upstream dependencies: N/A. Downstream integrators: N/A. Contributor count: N/A. Contract deployments: N/A. DAU/MAU: N/A. Retention: N/A.
I want you to notice something. Every bright protocol in this cycle has a messy ecosystem diagram — too many dependencies, a few embarrassing integrations, an active forum with a 300-message governance fight. The mess is the evidence. A vacuum has no mess because there is no system. When I asked the fund for the project's developer signals, they sent a screenshot of a Discord with 9,000 members. I asked for the commit graph. They sent the same screenshot.
Regulatory
Primary jurisdiction: N/A. Howey analysis: all four prongs unscoreable. KYC posture: N/A. Legal entity: N/A. The fund's compliance team signed off by writing "no identified issues" — which is technically true and entirely meaningless, because you cannot identify issues in a document that hasn't been written. This is the failure mode I most want to name: absence of findings is being systematically misread as absence of risk. A red-team report that says "we found nothing" after four hours of work is a red flag. A red-team report that says "we found nothing" after four hundred is a clean bill. The two look identical on a slide. The transaction is permanent; the mistake is not.
Team and Governance
Team: insufficient information. Governance: insufficient information. No names, no prior projects, no verifiable GitHub history, no LinkedIn trail that survives a reverse image search. Investor quality: one fund with a real track record, two funds I have never heard of, and a market maker whose only public activity is a website with a stock photograph on it.
I have a rule about anonymous teams that I will state plainly, because it has a 24-year track record behind it. Anonymity is not automatically disqualifying — several of the best builders in this industry have operated pseudonymously for legitimate reasons. But anonymity combined with an information vacuum is disqualifying, because the two properties are multiplicative, not additive. An anonymous team with an audited, deployed, documented, timelocked protocol is taking a bounded risk. An anonymous team with nothing is taking no risk at all, on the upside or the downside. The asymmetry is the tell.
Risk
When eight dimensions are unscoreable, the risk matrix collapses into a single row: everything, with probability unknown, impact total, mitigation none. I have seen risk committees approve this before. The reasoning is always the same, and it is always wrong: "we can't identify a specific risk, so we'll treat it as low."
That is a category error, and I want to give it a proper name. Call it vacuum scoring: the substitution of a null finding for a favorable finding. It is the same mistake as a doctor reporting "no tumor detected" when the machine was unplugged. Illusion has a price tag; truth has none. Vacuum scoring is how a fund pays $47 million for a hallucination and records it as an entry in the alpha column.
Narrative and Expectation Gap
Narrative: N/A. Expectation gap: N/A. There is no published roadmap, so there is nothing to fall short of. This is the most insidious property of the vacuum. A project that publishes nothing cannot miss a deadline, cannot fail a testnet, cannot get its mainnet delayed, cannot have a token generation event postponed. Every failure mode that produces negative information for a normal project produces no information for a vacuum project — and the market, which is trained to trade information, ends up trading the vacuum itself as if it were a positive. The buyer is not buying a roadmap. The buyer is buying the permanent option to be disappointed later.
Transmission
Upstream miners and infrastructure: no exposure. Exchanges: listing fee revenue only. DeFi: no integration, because there is nothing to integrate. Traditional finance: none, because no regulated venue can custody it. The transmission graph is a single node with a price tag attached.
I want to close this section with the number that mattered. I took the fund's base rate of stealth-listed positions over the prior 24 months — 31 deals — and classified each ex post. Eleven returned more than 5x. Twenty returned less than 0.3x, of which fourteen went to zero. Mean return was positive. Median return was a loss. The fund was not running a strategy. It was running a lottery with a marketing department, and the blank sheet was the ticket.
Contrarian: The Bulls Are Not Entirely Wrong
Here is the part of this article that would get me thrown out of certain group chats, and I am going to say it anyway, because a cold dissection that only confirms the prior is not a dissection — it is a warm bath.
The bulls are right that information asymmetry is where the outsized returns live, and they are right that the vacuum is not a scam signal by itself. The eleven winners in that 31-deal sample were not luck. Several had genuine technical substance that was published late by design — a team that chose to ship in silence, and whose silence during the accumulation window was the mechanism by which the token was mispriced. If you had demanded an audit before buying, you would have missed a 12x. That is not a hypothetical. That is a fact about how mispricing is created. Someone has to hold the position while the information is still missing, or the information was never missing in the first place.
The counter-argument to my own report is therefore: the N/A cell is the entry point. The diligence you cannot do is the reason the return exists.
I accept the premise and reject the conclusion. Here is the distinction that the bulls elide. An information asymmetry is a state in which some participants know more than others, and the information exists. An information vacuum is a state in which the information does not exist for anyone — not the team's counterparties, not the exchanges, not the auditors, not even, in the worst cases, the team itself. In the first state, the marginal buyer is paying for a real asset at a wrong price. In the second, the marginal buyer is paying for a story at any price. These two states are tracked by the same chart and they are not the same trade.
The eleven winners were asymmetries. The fourteen zeros were vacuums. Ex post, they were distinguishable — by whether the docs, when they eventually appeared, described a system that could have existed at the time of the silence. That is the test I now apply, and it is retroactively applicable in both directions. When the Terra/Luna seigniorage model was still opaque, the question was not "do we know enough?" The question was "if we knew everything, could this exist?" I spent two months reverse-engineering it in 2022 and the answer was no — the required LUNA demand was geometrically unbounded. The 40-page report went to regulators in Singapore and was ignored for three months. The vacuum was not an asymmetry. It was a vacancy. The code compiles, but the reality bankrupts.
There is a second blind spot in the bull case that deserves naming. The bulls treat the vacuum as a timing problem — get in early, get out before disclosure. But the disclosure regime in a mature token is not a binary event; it is a continuous auction. Every document that gets published, every audit that lands, every unlock that vests, is an incremental repricing. The vacuum does not resolve in a single pop. It resolves in a slow, then sudden, then total redistribution from late holders to early holders. The exit is not a door. It is a crowd trying to use the same door at the same time, and the only people who are guaranteed to make it are the ones who never needed it.
And a third, because the pattern has repeated three times in my career and I have stopped treating it as coincidence. Every vacuum project I have personally dissected at close range — the 2017 vesting overflow, the 2021 procedurally generated NFT metadata, the 2026 "decentralized" compute network whose operator list was 5,000 compromised IPs behind one entity — shared exactly one property in common. The vacuum was not a byproduct of the team's speed or stealth. The vacuum was the concealment layer. It was manufactured specifically to prevent the one query that would have ended the raise. In 2021 I hashed the trait generation seeds and found that 85% of the "rare" attributes were procedurally predictable — the rarity was a UI. The metadata was not missing. It was obfuscated. Those are different things, and the difference is the entire diligence brief.
Takeaway
So the blank sheet is not a failure of analysis. It is the analysis. When nine dimensions come back N/A, the report has a verdict — it is just a verdict the buyer has to be willing to read as adverse.
What I am watching now is the pricing of this artifact. Two years ago, an all-N/A diligence memo killed a deal. Today it is being filed alongside the term sheet as supporting documentation, because the absence of findings has been repriced as the absence of risk. That repricing is the most crowded trade in this market and it will unwind the way vacuum trades always unwind — not when the information arrives, but when enough holders discover that the information was never coming.
The fund that paid me for the week has a policy now. Any deal whose diligence sheet is more than 40% N/A goes to the second committee, with a one-line brief attached. I wrote the brief. It reads: we cannot tell you what this is, and that is the finding. The transaction is permanent. The mistake is not. But the position is, and someone is holding it while the sheet is still blank.
What is the correct price of a claim you cannot describe? The market is answering that question in real time, in public, with other people's capital. The answer, historically, has been zero. The interesting part is never the answer. The interesting part is the wait.