Required · Core evidence assessment
Is the central claim supported?
Six weighted tests: the problem, the mechanism, execution, outcomes, governance, and whether any of it can actually be verified. Scored out of five.
Methodology
How reviews are scored.
Two standards are in play: the fixed 35-point V1 rubric, and V2, which grades the evidence behind a claim and publishes no number when the evidence isn't there. Which one applies to a given review is stated on the review itself.
Start here
Two standards, and which one you are looking at.
Every review published so far carries a score out of 35 from the V1 rubric, which grades seven mixed dimensions in one composite number. V2 is the approved standard for new reviews: it scores the evidence behind the central claim out of five, keeps any token or share in a separate record, and publishes nothing when the evidence is too thin. Nothing has been published under it yet.
The two are not comparable, so no score is ever converted, averaged, or re-ranked across them.
Methodology 2.0 · two records, two questions
Core judges the claim. Asset judges one instrument.
Core is required for a scored V2 review. Asset only appears when there is a specific token or share to examine. The two results are never averaged together.
Optional · one named instrument
What does holding it actually get you?
A token, share, security or NFT gets its own separate record covering rights, exposure and how you lose money. It never forecasts price.
Evidence architecture
How much evidence there is, and which way it points.
Those are two different things. Strong evidence can produce a low score when it cuts against the claim. How much of it exists decides whether a score may be published at all.
E0
No admissible evidence
No admissible evidence tests the material claim. A subject assertion without inspectable support may establish what was claimed, but not whether it is true.
E1
Limited evidence
Evidence is limited to subject assertions, a single weak or stale source, a narrow observation, or materially incomplete data.
E2
Adequate evidence
Current primary or observable evidence tests the claim and has some triangulation, but material gaps or limits remain.
E3
Strong evidence
Multiple current, independent, authoritative, or reproducible sources test the claim and its countercase with no material coverage gap.
Core coverage floor
Insufficient when
- — Any of the six Core dimensions is E0.
- — Fewer than four of the six Core dimensions are E2 or E3.
Strong only when
- — All six dimensions are E2 or E3.
- — At least three dimensions are E3.
- — The central claim and every critical risk claim have qualified evidence.
- — The evidence is current enough for the claim's operating tempo.
Asset coverage floor
Insufficient when
- — Any Asset dimension is E0.
- — Fewer than two of the three Asset dimensions are E2 or E3.
Strong only when
- — All three dimensions are at least E2.
- — At least two dimensions are E3.
- — The instrument's rights, backing, issuance, and critical loss conditions are independently or authoritatively verified.
Insufficient evidence is not a bad score. It means there was not enough to score at all, so no number, verdict or label gets published.
Core dimensions
Six weighted tests of the central claim.
The verdict comes from the exact weighted result. Rounding only affects the number shown on the page.
Core 01
Problem and thesis
Is the problem material, and is the project's central thesis specific, coherent, and testable?
Inspect all five score anchors
- 1The stated problem or central premise is contradicted, immaterial, or internally incoherent.
- 2The problem is plausible, but the thesis is vague, self-asserted, or dependent on unsupported assumptions.
- 3The problem is real and the thesis is testable, but material assumptions remain unresolved.
- 4The problem is material, the thesis is clear and falsifiable, and evidence beyond the subject supports it.
- 5The problem and thesis are independently established across relevant contexts and survive serious counterevidence.
Core 02
Product and mechanism
Can the current product, protocol, or mechanism plausibly produce the promised result?
Inspect all five score anchors
- 1A critical design flaw or observed behaviour prevents the mechanism from producing the claimed result.
- 2The mechanism is conceptual, materially incomplete, or too opaque to connect its inputs to the promised outcome.
- 3A live or credible mechanism exists, but important dependencies, assumptions, or failure modes remain unresolved.
- 4The live design is coherent, its critical path is documented, and its material assumptions have been tested.
- 5The critical path, dependencies, and failure modes have survived meaningful independent or adversarial testing under relevant conditions.
Core 03
Execution and reliability
Has the subject shipped, operated, and responded to failure reliably enough for the assessed claim?
Inspect all five score anchors
- 1Repeated or material execution failure directly undermines the claim and remains unresolved or misrepresented.
- 2Evidence is dominated by plans, prototypes, irregular delivery, or unverified operating claims.
- 3The product is live and delivery is credible, but operating history, reliability, or incident evidence remains limited or mixed.
- 4Delivery is sustained, operation is reliable for the relevant maturity stage, and incidents are addressed transparently and effectively.
- 5Reliable operation has been demonstrated repeatedly at relevant scale or stress, including credible failure response and recovery.
Core 04
Demonstrated outcomes
Are the claimed benefits visible, durable, retained, and credibly attributable to the project rather than incentives or noise?
Inspect all five score anchors
- 1Observed results contradict the claimed benefit, show material harm, or demonstrate systematic extraction.
- 2Activity exists, but the promised outcome is absent, transient, incentive-dominated, or not credibly attributable.
- 3Early or mixed outcomes exist, with material retention, attribution, quality, or concentration questions.
- 4Repeated outcomes, retained use, paying demand, or repeat customers are credibly attributable to the product.
- 5Durable outcomes are independently evidenced across relevant cohorts, cycles, or operating conditions.
Core 05
Control and accountability
Who can change, stop, override, or extract from the system, and are those powers visible and accountable?
Inspect all five score anchors
- 1Critical powers or rights are hidden, materially misrepresented, or associated with unresolved abuse or denial of user rights.
- 2Control is concentrated or unclear, accountability is weak, and credible oversight, appeal, redress, or exit is missing.
- 3Operators and powers are disclosed, but material control, rights, oversight, or redress risks remain.
- 4Controls are proportionate and disclosed, while checks, oversight, timelocks, appeal, or exit mechanisms operate in practice.
- 5Powers are limited and auditable, independent oversight is meaningful, and redress or exit has been tested under pressure.
Core 06
Transparency and verifiability
Can material claims, methods, metrics, rights, limitations, and conflicts be checked by someone other than the project?
Inspect all five score anchors
- 1Material claims are misleading or contradicted, or the evidence required to test them is inaccessible.
- 2Evidence is mostly self-reported and important methods, data, rights, limitations, or conflicts are missing.
- 3Primary documentation and observable evidence cover the main claim, but material verification gaps remain.
- 4Current methods, rights, metrics, and limitations are accessible and independently triangulatable.
- 5Material results are reproducible, current, and substantially complete, with key limitations disclosed and serious challenge possible.
Asset dimensions
Three equal tests of one exact instrument.
Two instruments means two separate records. They are never blended into one.
Asset 01
Holder rights and value linkage
What does the holder own or receive, and how does value or utility reach the instrument?
Inspect all five score anchors
- 1The claimed exposure or rights do not exist, are materially contradicted, or cannot function as represented.
- 2Rights or value linkage are indirect, discretionary, undocumented, or dependent on unaccountable intervention.
- 3Rights and linkage are documented and plausible, but operation, enforcement, or durability remains unproven.
- 4Rights or value flows are mechanical, enforceable, transparent, and operating as documented.
- 5Rights and value linkage are independently verified, durable, and proven under relevant stress or adverse conditions.
Asset 02
Market structure and incentives
Are supply, distribution, liquidity, incentives, and manipulation risks structurally defensible?
Inspect all five score anchors
- 1Structure is opaque, extractive, readily manipulated, or dominated by concealed adverse incentives.
- 2Ownership, liquidity, issuance, unlocks, or incentives are highly concentrated, reflexive, fragile, or poorly disclosed.
- 3Material structural risks are disclosed and the instrument is usable, but concentration, liquidity, or incentive risks remain.
- 4Supply, distribution, liquidity, and incentives are transparent, sufficiently diverse, and reasonably aligned.
- 5Structure is demonstrably resilient under stress, with no dominant extraction path and independently verified market integrity.
Asset 03
Loss conditions and protections
Are custody, legal, technical, redemption, and counterparty loss conditions visible and credibly mitigated?
Inspect all five score anchors
- 1Material loss conditions or rights are concealed, misrepresented, or shown to fail when exercised.
- 2Major legal, custody, solvency, redemption, counterparty, or technical risks remain unresolved with weak recourse.
- 3Material risks are documented and partial safeguards exist, but protections or recovery remain conditional.
- 4Protections are enforceable or technically credible, disclosures are complete, and operations or recovery have been tested.
- 5Legal and technical protections are robust, independently verified, and transparent under material failure or market stress.
Result language
Evidence verdicts, not investment bands.
Core and Asset use different labels. The exact result picks the label; the rounded value shown on the page never moves a boundary.
Core evidence verdictUnrounded result
Evidence cuts against the claim
1.0 ≤ x < 1.8
Not established
1.8 ≤ x < 2.6
Mixed evidence
2.6 ≤ x < 3.4
Supported with reservations
3.4 ≤ x < 4.2
Strongly supported
4.2 ≤ x ≤ 5.0
Asset structure labelUnrounded result
Structurally impaired
1.0 ≤ x < 1.8
Material structural concerns
1.8 ≤ x < 2.6
Conditional
2.6 ≤ x < 3.4
Structure broadly supports the stated exposure
3.4 ≤ x < 4.2
Structure strongly supports the stated exposure
4.2 ≤ x ≤ 5.0
Early / provisional
Maturity is visible.
Early projects still have to clear the evidence floor. Their Core verdict cannot go above Mixed evidence, and every record names a date or event that triggers another look.
Lower-only gates
Red flags can only drag a result down.
An unresolved critical flag on the project caps Core at Not established. One on the instrument caps Asset at Material structural concerns. The underlying number stays visible either way.
Critical record
Six named red-flag classes.
Deception, security, solvency, hidden control, unverifiable backing and manipulated headline metrics. Each stays dated and sourced, including how it was resolved.
Publication eligibility
The score is the last step, not the premise.
A V2 review needs a claim that could be proven wrong, an observable subject, four to eight material claims, independent evidence on outcomes and risk, a countercase, and a documented hunt for evidence against the thesis.
Length contract
- Target
- 1600–2600
- Warning
- 2800+
- Override required
- above 3200
- Hard ceiling
- 4000
When scoring stops
If coverage comes back Insufficient, editorial integrity is blocked, or the piece cannot meet the evidence requirements, it gets published as a Breakdown or Analysis instead. That call is made by the editor, not automatically.
Immutable history
Fresh research creates a fresh record.
Corrections, re-scores on the same method, and moves between methodologies each link back to a fixed snapshot. Earlier records stay readable.
correction
A factual, sourcing, disclosure, or copy correction
New assessment: Only when the judgement, dimension inputs, coverage, gate, or verdict changes
reassessment
Fresh evidence produces a replacement assessment under the same methodology version
New assessment: Yes
methodology change
Fresh research produces an assessment under a new methodology version
New assessment: Yes; never represented as a numeric before/after conversion
Limits and accountability
None of this predicts price, recommends a trade, endorses an instrument, or replaces your own diligence. Core weighs the evidence for a claim; Asset describes how one instrument is put together.
No project pays for a review, a score, or a reassessment. Disclosures, sources, evidence cutoffs, gates and corrections all stay on the public record.
Methodology 1.0
The fixed 35-point rubric.
Every V1 score on the site comes from this rubric. It stays public so those scores can still be checked, and it is never converted into a V2 number.
Public scoring standard
35 pts
01
02
03
04
05
06
07
- Dimensions
- 7
- Points each
- 5
- Total
- 35
A review is a fixed rubric applied to public evidence, token-thesis linkage, team credibility, traction, timing, risk/reward, and transparency. Not a vibe check.
Version boundary
Comparable inside V1. Not comparable with V2.
V1 totals, bands, dimension order and public rendering are frozen. A later V2 reassessment creates a new dated record; the V1 score stays visible as history rather than being overwritten.
V1 dimension grid
Seven fixed dimensions, five points each.
The labels, order, questions and anchors below are exactly those used by every published V1 scorecard.
01
Underlying thesis strength
Is the problem real, and is this a credible way to attack it?
- Scores low when
- The thesis only works inside its own pitch deck.
- Scores high when
- The thesis survives contact with sceptical first principles.
02
Token-thesis linkage
Does the token capture the value the thesis creates, or just sit next to it?
- Scores low when
- The token is a fundraising instrument with a narrative attached.
- Scores high when
- Value accrual is mechanical, documented, and already in force.
03
Founder/team credibility
Has this team shipped before, and do their public claims check out?
- Scores low when
- Anonymous or unverifiable, with claims that don't reconcile.
- Scores high when
- Named, accountable, with a verifiable record in the domain.
04
Community and traction
Is usage organic and retained, or rented through incentives?
- Scores low when
- Activity is mercenary and disappears when emissions stop.
- Scores high when
- Retention holds without payment; users would miss the product.
05
Narrative timing
Is the market ready for this story, or is the project early, late, or off-cycle?
- Scores low when
- Fighting the cycle, or arriving after the narrative is exhausted.
- Scores high when
- Positioned ahead of a narrative with visible catalysts.
06
Risk/reward
At current valuation, what do you risk against what you could plausibly make?
- Scores low when
- Downside is structural; upside requires everything to go right.
- Scores high when
- Asymmetry favours the holder even under conservative assumptions.
07
Transparency and disclosure
Can the project's own numbers and claims be independently verified?
- Scores low when
- Key facts are missing, obfuscated, or contradicted by on-chain data.
- Scores high when
- Docs, treasury, vesting, and metrics are public and reconcile.
V1 score anchors
What a V1 1 to 5 means.
Five points is the most granularity the frozen V1 rubric claims.
1/5
Claim contradicted by evidence, or no evidence exists.
2/5
Claim unverified; what evidence exists cuts against it.
3/5
Mixed: real signal, real unresolved problems.
4/5
Claim substantially supported by verifiable evidence.
5/5
Claim independently verified; would survive a hostile audit.
V1 verdict bands
V1 verdicts stay pinned to /35 totals.
These band labels read like investment calls. V2 deliberately does not; it talks about evidence instead.
BandScoreMeaning
Avoid
Below 40% (0–13)
The evidence cuts against the pitch. The burden of proof is on the project, and it hasn't met it.
Watch
40–59% (14–20)
Real signal, real unresolved problems. Worth tracking; not worth underwriting until named risks resolve.
Speculative
60–79% (21–27)
The thesis holds but key claims remain unproven. Constructive, with explicitly sized uncertainty.
Conviction
80%+ (28–35)
Claims verified, incentives aligned, risks named and survivable. The rarest band by design.
See the rubric in action
Every scored review links back to this standard. The scorecard is useful only if readers can inspect the same rules every time.