Academic access to Hadal instruments
A named academic pays 50% of the annual list on the instruments whose subject matter is research methodology. Their research group and the students they supervise are covered under that one licence. The conditions are real ones — verified affiliation, non-commercial use, citation, and findings published without any veto of mine — and they are set out in full below, along with what the programme cannot cover and why.
50%
50% of the annual list price
The halving itself is a ratified ruling. The list figures it applies to are still anticipated, so the resulting academic figures move if a list figure does — and nothing on this site is purchasable: there is no checkout, no card capture and no account to create.
- Eligible
- 06
- Excluded
- 14
- Catalogue
- 21
- Term
- 12mo
The licence is held by a named academic
Naming the holder is what makes a flat rate sustainable for a one-person desk, so it goes first rather than into the terms. What protects that capacity is not the rate — it is the route. One person holds the licence and one person is the route to me.
- Who may hold one
- A named academic — faculty, permanent research staff, or a principal investigator — at a recognised institution. Institutions and research organisations hold licences through a named individual in the same way.
- Who it covers
- The holder, their research group, and students under their supervision, for non-commercial research and teaching. Cohort size is stated at application and does not change the rate.
- How support runs
- Support runs through the holder and only the holder. A two-hundred-student cohort is one relationship, which is what makes the rate sustainable for a small team.
A cohort is one relationship, not a headcount. Cohort size is stated when the licence is issued and does not change the rate. Students work under the holder’s licence and their questions go to the holder, who teaches them — which is the arrangement a department already runs. Because that is a support-model rule rather than a pricing rule, it survives any future change to the rate.
One flat rate, and the arithmetic in the open
There is one academic rate — 50% of the annual list price — and it applies to every eligible instrument in the same way. No tiers, no negotiated academic band, no application to be assessed on. The arithmetic is published so you can reproduce it the moment the list figure it applies to is ratified:
ANNUAL LISTmultiplied by0.5equalsACADEMIC ANNUAL
Each eligible instrument publishes its own anticipated list indication on its own page, and those figures are reproduced beside each instrument below. The academic annual figure is not printed yet, and I would rather say why than leave a blank: the rate applies to an annual list, no annual figure has been ratified for any instrument, and whether an academic licence is issued per seat or as one holder licence covering a group is not settled either. Both are ratifications I have not yet made, and any figure printed ahead of them would be chosen rather than derived. Until they land, the figure reads NOT YET PUBLISHED and the multiplier above tells you exactly what it will be when it does.
The rate replaces the bands; it never compounds with them.The academic rate applies instead of the seat and term bands, never on top of them. Stacked, the three would land below the cost of serving the account, and a rate I cannot honour is worse than one I never offered. The bands themselves — how seats and term length move a commercial licence — are set out in how a licence is shaped.
Standing policy
- No promise of profit, ever. Hadal makes no performance claims and carries no implied edge. Past measurements describe instrument behaviour — never future returns.
- You own risk management. Hadal cannot control it and does not insure it. Good tools do not fix bad discipline — and this site says so.
- Analytical tools, for discretionary use. Nothing here is investment advice or a recommendation to trade. Every decision, and every outcome, is yours.
The same statement stands in the footer of every page.Terms Privacy
What the rate covers, and the line that draws it
An instrument qualifies when its subject matter is research methodology. That is a principle rather than a shortlist, and it is why the six instruments below are the ones on it. Each carries the reason it qualifies, in full, so you can judge whether the line was drawn honestly.
- P2Backtest Honesty
Overfit Auditor
Mathematical bounds on how hard you tortured the data.
Backtest overfitting is an open research literature, not a trade secret. The instrument implements the published tests, so an academic using it is checking my arithmetic — which is the point.
- P4The Honesty Stack
Epistemic Harness
Registry-as-a-service: pre-registration, walk-forward enforcement, the kill ledger.
A discipline for separating what a result establishes from what it merely suggests. That is a methods problem before it is a trading problem.
- P5The Honesty Stack
Agent-Governance Constitution
The rails that caught a forged ratification — packaged.
AI-agent governance is an active research area, and my own founding incident is a first-party case study in it. Researchers studying agent oversight should not be paying full price to examine the constitution that caught a fabricating agent.
- P10Backtest Honesty
Data Forensics
Your stats can be clean and your data still lying.
Dataset provenance, survivorship and point-in-time integrity are dataset-science problems that reach far outside finance. Any field that inherits a dataset inherits its defects.
- P18Backtest Honesty
Reproducible Verdict Kernel
Same inputs, same verdict, byte for byte — or it is not a verdict.
Byte-identical recomputation of a stated result is the reproducibility problem itself. Withholding this one from researchers would be self-refuting.
- P21Integrity Services
Gate Falsification Kit
A gate that has only ever passed proves nothing.
"Can every gate actually fail?" is a software-engineering research question with a universal answer surface, and vacuous-green CI is endemic well beyond trading.
The list figures above are each instrument’s own anticipated indication — set by me, not yet ratified — reproduced here exactly as its page carries it. Every one of them is indicative and awaiting ratification, and there is no checkout anywhere on this site to use them in. The halving is settled, though — the academic rate is a ratified decision, not an indication. What moves when the list figures are finalised is the number the rate is taken from, never the rate itself.
What it does not cover, and which of those I could change
Publishing the gap with its reason is the house style; an unexplained omission is the thing I do not ship. Three of the four classes below are walls — constraints that are not Hadal's to lift, at any price. The remainder is a choice I have made, and a choice can be revisited. Those are different sentences and this page does not blur them.
Licensed-market-data instruments
WALLNot Hadal's to lift
They carry third-party data licences that do not permit onward grant. This exclusion is not Hadal's to lift at any price, and I would rather say so than quietly leave them off the list.
Live-market and broker instruments
WALLNot Hadal's to lift
They measure a live commercial relationship with a venue and need the account data that comes with one. There is no research-shaped form of them for a discount to apply to.
Deployment-shaped products
WALLNot Hadal's to lift
Their cost is support, not licence — halving the fee does not halve the work, and support capacity is the real constraint on this business. Read-only demo access and the published receipts corpus serve examination instead.
Commissioned audits
CHOICEMy decision, and revisitable
A service delivered by a person rather than a licence granted to one. Bounded research collaborations are discussed case by case.
My eligibility register places every other instrument in one of the lists above and has not yet placed this one. Until it does, it is not covered. An unsettled question publishes as an unsettled question here, rather than being resolved in my favour or quietly left off the page.
The terms, in full
The terms below are conditions of the rate, not requests. They are also the answer to the obvious commercial question, so they are published in full and in one place — copy them into a departmental resource page without asking first.
- Verified institutional affiliation — an institutional address and a public faculty, lab or departmental page naming you.
- Non-commercial research and teaching only. Consulting, fund work, or any revenue-generating use requires a commercial licence, including where the same institution has a commercial arm.
- Citation in any published work that used the instrument. Each instrument page carries its canonical citation block.
- Findings publish without any veto of mine. The gag-free rule applies to academics exactly as it applies to the venues I measure.
- Annual term, renewable. The rate applies instead of the seat and term bands, never on top of them.
Why a university pays half. A commercial buyer may reasonably ask why a university pays half. The answer is that the academic rate is not the same licence at a lower price — it carries conditions a commercial buyer would not accept: non-commercial use only, mandatory citation, and findings published without any veto of mine. Anyone willing to take those terms is welcome to that rate, and almost no commercial buyer is.
There is no free tier and no application queue attached to any of this. A free grant needs an assessment process to staff, and a process I cannot staff would fail the first department that relied on it — so the programme is one rate, one set of conditions, and no gate to wait behind.
What support looks like at this rate
Half the fee buys the same instrument and a different queue position, and it is better to say so here than to let it be discovered later. Every request is a ticket — there is no phone line and no live chat at any tier, academic or otherwise.
- Tickets from the named holder. Students reach support through their supervisor, never directly — that routing is what makes a two-hundred-student cohort one relationship instead of two hundred.
- No onboarding session, and no commissioning support.
- Answered after the paid queues, honestly and without apology: the rate is half, and the queue position reflects it.
- A defect report is escalated to the normal defect path immediately. Discounted support is slower; it is not less careful.
First response is targeted at five business days, which is a target rather than a guarantee, and a first response rather than a fix. The reason the rate works at all is the routing: because support reaches me through the holder and only the holder, a cohort of two hundred students is a single relationship to carry rather than two hundred. That is a support-model rule, not a pricing one, so it holds whatever the rate becomes.
The support model in detail — tiers, what decides yours, and what support does not cover
Citation is what I am paid in
The rate is not a favour and citation is not a courtesy — it is the consideration in the exchange, which is why it sits in the terms rather than in a footnote. What I do with it is the part that decides whether the programme is credible or promotional:
Cited work is listed as it is published, never quoted selectively, and a citation reporting a negative result about one of my instruments is listed exactly as prominently as one reporting a positive result. A citation list that only carries flattering findings is a testimonial wall wearing a lab coat.
An instrument is cited by its name, its catalogue number and the URL of its page — for example, Overfit Auditor (P2) at its instrument page, with the date you accessed it. Every instrument in the catalogue is listed on the instruments index, and the reuse terms for the published measurement corpus are set out in the data licence. This page carries its own copy-ready citation below.
Why I run this, without the coyness
There is a commercial motive and I would rather state it than dress it as philanthropy. Researchers become practitioners. Departments become institutional buyers. And a citation in a methods section is third-party scrutiny that no marketing budget can buy — the strongest thing that can be said about an instrument is that somebody with no stake in me used it and reported what it did.
Stating the motive is also how you check the terms. A programme run for citations has an obvious temptation — to list the flattering ones — which is exactly why the commitment above binds in the direction that costs me: a published negative result about one of these instruments is listed as prominently as a favourable one. The terms are good regardless of the motive, and you should read them knowing what mine is.
Questions
Can a student hold an academic licence?
No. The licence is held by a named academic, and students are covered under that holder’s licence for teaching and supervised research. Support runs through the holder and only the holder, which is what turns a two-hundred-student cohort into one relationship rather than two hundred.
Is there a free tier or an application queue?
No. There is one academic rate on the eligible instruments and nothing to queue for. A free tier needs an application process to staff, and a process I cannot staff is worse than a rate I can honour.
Can an excluded instrument be added later?
Three of the four excluded classes are walls rather than choices — licensed-market-data instruments; live-market and broker instruments; deployment-shaped products — and a wall does not move for anyone. The rest are choices I have made, and a choice can be revisited.
Does the academic rate stack with the seat and term bands?
The academic rate applies instead of the seat and term bands, never on top of them. Stacked, the three would land below the cost of serving the account, and a rate I cannot honour is worse than one I never offered.
How does an academic apply today?
There is no application form, because academic access is not open yet — but the contact routes are published, so an enquiry can be sent today and will be read. The launch list is the notification channel: it takes an email address, and optionally a name and one line on what you would want measured. It is not a route to a reply, so an application cannot be lodged through it, and joining it is how you are told first when academic access opens. Have four things ready for when it does: your institution and role, a public faculty or lab page that names you, the group the licence would cover, and which instruments you want. Joining commits you to nothing.
How to register interest today
Plainly: there is no application form on this site, no academic address is published, and nothing is purchasable. Those are absences rather than omissions — an inbox that does not exist would route a real request into a void, and a form posting to it would take your message and drop it. Every contact route and its current status is published rather than implied.
The launch list is a notification list rather than a message channel: it takes an email address, and optionally a name and one line on what you would want measured. That line shapes what gets researched, not what gets replied to, so a request cannot be lodged through it. Join it to be told when academic access opens — and for an actual question, the published contact routes are open today. Have these four things ready for when it does — they are what the licence needs, and they are what will be asked for first:
- your institution, and your role in it;
- a public faculty, lab or departmental page that names you;
- the group the licence would cover — research group, module, or supervised cohort;
- which of the eligible instruments you want, and what you intend them for.
None of that is a form you are filling in now. It is the shape of the request, published in advance so a department can see whether it qualifies before anyone spends time on it.
Join the launch list
Academic access opens when the catalogue does, and the launch list is told first. For a question rather than an announcement, the contact routes are open.
Standing policy
- No promise of profit, ever. Hadal makes no performance claims and carries no implied edge. Past measurements describe instrument behaviour — never future returns.
- You own risk management. Hadal cannot control it and does not insure it. Good tools do not fix bad discipline — and this site says so.
- Analytical tools, for discretionary use. Nothing here is investment advice or a recommendation to trade. Every decision, and every outcome, is yours.
The same statement stands in the footer of every page.Terms Privacy
Related: how Hadal prices ·the full instrument catalogue ·how the instruments measure