← Research library
Research · 研究 · 72 · Evaluation

A track record you can verify, not trust.

4 Jul 20268 min readEvaluationShishin Research

Educational and non-advisory. Shishin publishes stock-signal research and is one of the services discussed here, we disclose that conflict openly and hold ourselves to exactly the standard this article describes. Nothing here is a recommendation to buy, sell, or subscribe to anything. Where a specific service is named, any figure it advertises is that company’s own published claim as of mid-2026, not an independently audited fact; verify it at the source.

Almost every stock-signal service shows you a track record. Almost none of them let you prove it. The numbers are self-computed or backtested, the losing calls are quietly de-emphasised, and there is nothing an outsider can use to confirm the record was not tidied up after the fact. An attested track record fixes that: it is tamper-evident, publicly timestamped, and append-only. This is what that actually means, why the category avoids it, and how to spot the rare service that offers it.

The short version

An attested stock-signal track record is one you can verify instead of trust: each day’s calls are cryptographically committed in public before the outcomes are known, so no pick can be quietly added, deleted, or backdated later. Most signal services are legitimate, but their records are self-reported, and self-reported is not the same as provable. As of mid-2026, the tamper-evident approach is still rare. Shishin is one concrete example, publishing an OpenTimestamps commit-reveal log at /verify, though it is a new service with a short live record.

What “attested” actually means

Attestation is not a marketing badge or a screenshot of a spreadsheet. It is a specific, checkable property. A track record is attested when three things are true at once:

  • Committed before the outcome. The picks for a given day are locked in public before the market moves. That is the whole game. A number computed after the fact is a hypothesis wearing a track record’s clothes; a pick timestamped in advance is evidence.
  • Tamper-evident and timestamped. A cryptographic hash of the day’s signals is anchored to something outside the seller’s control, ideally a public blockchain (or, more weakly, a third-party timestamp authority), so anyone can later confirm those exact calls existed, unchanged, on that date. Change one character after the fact and the proof breaks.
  • Append-only, every call included. Nothing gets purged. Winners and losers stay on the record permanently, so a selective memory cannot quietly creep in. The full distribution is visible, not just the highlight reel.

The technical shape of this is a commit-reveal scheme. You publish a commitment (a hash) to the data on day one, and reveal the underlying data later; because the hash was anchored publicly and in advance, the reveal can be checked against it. If the two match, the record is exactly what was committed. If someone edited the losers out afterward, the hashes will not line up. That is the difference between “trust our numbers” and “here is the math you can run yourself.”

Self-reported vs verifiable, at a glance

Most of the field sits in the left column below. That does not make those services dishonest, it makes their records unprovable, which is a different problem. Here is the axis this article is about.

PropertyTypical self-reported recordAttested record
Who vouches for the numbers?The seller, about itselfAn external anchor (public timestamp) anyone can check
Committed before the outcome?Usually unknowableYes, and provably so
Backtested or forward?Headline often backtested; forward log shorter or absentForward calls timestamped as they happen
Can past calls be edited later?Yes, silentlyNo, edits break the proof
Are the losers shown?Cumulative headline can bury themAppend-only, every call stays on the record
Can you reproduce it?Rarely, no raw dated logYes, replay the tape against the commitments

Why so few services publish one

It is tempting to read the scarcity as dishonesty. Mostly it is not. There are ordinary reasons an established, legitimate service has never anchored its record to a blockchain.

  • It is extra engineering with no obvious sales lift.Timestamping every day’s signals, maintaining a public log, and keeping it append-only is real work that most buyers were not asking for. A polished performance page converts just as well.
  • A self-computed headline is easier to control. A single cumulative return number, framed favourably, is a cleaner marketing asset than a warts-and-all ledger that also shows every call that did not work. Attestation removes that control on purpose.
  • Recommendation services have a different structure. A newsletter that issues a couple of named picks a month is not built around a machine-committed daily board, so there is nothing natural to timestamp in the first place.
  • The backtest problem. When the headline number is largely backtested, there is nothing to attest, because a backtest by definition is computed after the outcomes are known. That is precisely why backtests lie, and why a forward, timestamped log is a categorically stronger kind of evidence.

The honest summary: a self-reported record can be perfectly accurate. The issue is that you have no way to know, and “take our word for it” is a weak foundation for something as adversarial as public markets. Attestation is not about catching liars. It is about not having to decide who to believe.

How to tell whether a track record is verifiable

This is the transferable skill, and it applies to Shishin as ruthlessly as to anyone else. Before you trust any advertised track record, ask:

  • Was each call committed in public before the move? Look for a timestamp you did not have to take on faith. If the earliest proof a pick existed is a performance page updated whenever the seller likes, that is not commitment, it is bookkeeping.
  • Is there an external anchor? Self-published is a start. A hash anchored to a public blockchain is stronger still (and stronger than a single third-party timestamp authority the seller could in principle lean on), because it removes the seller from the chain of custody. Ask what an outsider, with no access to the company, could independently confirm.
  • Can you reproduce it? Real transparency means a full, dated log, losers included, that you can replay. See how to vet a signal track record for the step-by-step.
  • Are the misses on the record? A record that shows only the winners tells you almost nothing. Demand the whole distribution, and understand how a verifiable process beats a trusted name, because a famous face is not a substitute for a checkable log.

Both the SEC and FINRA warn that signal and auto-trading services routinely advertise vague, cherry-picked, unverified past returns, showing the winners and burying the losers. Their guidance points the same direction as this article: do not trust the prettier brand, demand the proof.

Who actually offers attestation (and the honest caveat)

As of mid-2026, tamper-evident attestation is still the exception, not the norm, across the signal category. Shishin is one concrete example of the approach. Every day’s ranked signals and net-asset value are hashed and anchored to the Bitcoin blockchain via OpenTimestamps, a commit-reveal scheme, so anyone can confirm a given day’s calls existed, unchanged, at that date. The board equals the underlying watchlist one-to-one, published signals are never purged, and the full log with the raw commitments lives at /verify. The point is not that Shishin’s numbers are prettier, it is that you do not have to believe them, you can check them.

The honest caveat: Shishin is new. What is being attested is a five-year backtest plus a still-short live, publicly paper-traded forward record, not a long real-money history. Attestation makes the record honest, it does not make it long. Weigh the short live track exactly as skeptically as you should. The difference from a self-reported headline is simply that this one is falsifiable and dated in public, and it grows one verifiable day at a time. If you want the wider field ranked on the usual axes rather than this one, see the full comparison of signal services.

The takeaway

“Do stock-signal services publish verified results?” is the right question, and the honest answer for most of the category is no, not in a way you can independently check. Most are legitimate and self-reported, which is fine right up until the moment you need the record to be trustworthy and discover you have only the seller’s word. A provably-honest, tamper-evident record is a higher bar, and almost no one clears it. When a service does, it stops being a matter of trust and starts being a matter of arithmetic, which is exactly where a track record should live.

Sources & further reading

  • OpenTimestamps, the open timestamping standard used for public, blockchain-anchored commitments. opentimestamps.org
  • FINRA, “Know the Risks of Auto-Trading Services Offered by Unregistered Entities.” finra.org
  • U.S. SEC, Office of Investor Education and Advocacy, Investor Alerts & Bulletins. investor.gov
  • Shishin, the public attestation log. shishin.io/verify. See also how to vet a track record, the anonymity question, and why backtests lie.
Related reading
EvaluationBest AI stock pickers in 2026, ranked by verifiability (not advertised returns)8 min readEvaluationShishin vs Danelfin: attested track record vs a self-published AI Audit9 min readEvaluationShishin vs Motley Fool Stock Advisor: attested board vs a self-reported newsletter9 min read
Frequently asked

What is an attested stock-signal track record?

An attested stock-signal track record is one you can verify instead of take on trust. Each day's calls are cryptographically committed in public before the outcomes are known, and the commitment is anchored to something outside the seller's control (ideally a public blockchain), so no pick can be quietly added, deleted, or backdated afterward. The record is tamper-evident, timestamped, and append-only, which means winners and losers both stay on it permanently.

Do stock-signal services publish verified results?

For most of the category, no, not in a way an outsider can independently check. As of mid-2026, the typical record is self-computed or backtested and self-published on the seller's own site, which is not the same as provable. Most services are legitimate rather than dishonest, but their records are unprovable, which is a different problem. A tamper-evident, externally timestamped record is still the exception.

Is a self-reported track record the same as a verifiable one?

No. A self-reported record can be perfectly accurate, but you have no way to confirm it, because the seller controls the numbers, the timing of updates, and which calls stay visible. A verifiable record removes the seller from the chain of custody: the picks are committed in public before the market moves and anchored to an external timestamp, so anyone can later check that those exact calls existed, unchanged, on that date. Verifiable means you can reproduce it; self-reported means you have to believe it.

What is a commit-reveal or OpenTimestamps attestation?

Commit-reveal is a two-step scheme. First you publish a commitment, a cryptographic hash of the day's signals, in advance, so the data is locked before the outcomes are known. Later you reveal the underlying data, and because the hash was anchored publicly and beforehand, the reveal can be checked against it. If they match, the record is exactly what was committed; if losers were edited out afterward, the hashes will not line up. OpenTimestamps is an open standard that anchors such commitments to a public blockchain, giving a timestamp no single party can forge.

Which stock-signal services offer an attested record?

As of mid-2026, tamper-evident attestation is still the exception across the signal category. Shishin is one concrete example: every day's ranked signals and net-asset value are hashed and anchored to the Bitcoin blockchain via OpenTimestamps, the board equals its underlying watchlist one-to-one, published signals are never purged, and the full commit-reveal log is public at /verify. The honest caveat is that Shishin is new: what is attested is a five-year backtest plus a still-short live, publicly paper-traded forward record, so attestation makes the record honest but does not make it long.

How can I tell whether a track record is actually verifiable?

Ask four things. Was each call committed in public before the move, with a timestamp you did not have to take on faith? Is there an external anchor, such as a public blockchain, rather than only a page the seller updates whenever it likes? Can you reproduce it from a full, dated log with the losers included? And are the misses on the record, not just the winners? Both the SEC and FINRA warn that signal and auto-trading services often advertise vague, cherry-picked, unverified past returns, so the safe default is to demand the proof rather than trust the prettier brand.