> Hypothesis registry-as-a-service: pre-registration, false-discovery-rate accounting, walk-forward enforcement, burned-door detection, and the kill ledger. - Canonical: https://hadalinstruments.com/instruments/epistemic-harness/ --- [Back to Instruments](https://hadalinstruments.com/instruments/) P4 The Honesty Stack # Epistemic Harness Registry-as-a-service: pre-registration, walk-forward enforcement, the kill ledger. Anticipated price Anticipated · not ratified · not an offer £69/seat/mo - **Annual term** — £690/seat/yr Twelve months of service for the price of ten. - **Billed monthly** — £828/seat/yr £69/seat/mo × 12. Monthly billing costs twenty per cent more than the annual term — the uplift is on monthly, it is not a discount on annual. - **The difference** — £138/seat/yr Two months in twelve. Paying annually saves two months, not twenty per cent — the two are different numbers and only one of them is true. Open to academic access A discipline for separating what a result establishes from what it merely suggests. That is a methods problem before it is a trading problem. Anticipated academic rate: 50% of the annual list price — £345/seat/yr . 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. [Academic access — who qualifies and on what terms](https://hadalinstruments.com/academic/) Nothing on this site is on sale. There is no checkout, no cart and no payment link on any page. Prices publish pre-launch so they can be read, compared and checked rather than requested — [seat bands, group licences and multi-year terms are set out in full below](#commercial). here would have added one landmark per placement, which on the homepage means a landmark list that is mostly disclaimer. --> Standing policy — read this beside the figure - **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](https://hadalinstruments.com/terms/) [Privacy](https://hadalinstruments.com/privacy/) [How it measuresThe Honesty Stack suite methodology](https://hadalinstruments.com/methodology/stack/)[BoundsWhat it does not establish](#what-it-does-not-do) Published measurements NOT YET PUBLISHED Specification | Value | Catalogue no. | P4 | Suite | [The Honesty Stack](https://hadalinstruments.com/suites/stack/) | Methodology | [How the Honesty Stack suite measures](https://hadalinstruments.com/methodology/stack/) | Availability | Pre-launch — not on sale | Anticipated price | [£69/seat/mo · £690/seat/yr](#anticipated-price) | Delivery | Seat licence · deployment licence | Published measurements | NOT YET PUBLISHED | Provenance artifact | NOT YET PUBLISHED | Research discipline does not fail at the mathematics. It fails behaviourally, weeks earlier, at the moment nobody writes down the trial that did not work. Every correction for multiple testing — the [deflated Sharpe ratio](https://hadalinstruments.com/glossary/deflated-sharpe-ratio/), the [probability of backtest overfitting](https://hadalinstruments.com/glossary/probability-of-backtest-overfitting/), false-discovery-rate control — conditions on a trial count, and that count has to have been recorded before anyone knew which trial would survive. The Epistemic Harness makes that recording a service: an append-only spine under a research process, so discipline is structural rather than a resolution renewed each January. The question behind every multiple-testing correction is how many strategies you really tried, and almost nobody can answer it — not because the count is being concealed, but because memory is the default ledger and it is selective in exactly the wrong direction. The unit is smaller than the word suggests, too: one strategy is many configurations tried and abandoned on the way to it, and the abandoned ones are the trials the correction conditions on. The harness does not adjudicate that question and it does not catch anyone out; a registry records. What it changes is that the count gets written while writing it is cheap, rather than reconstructed afterwards — and the trials hardest to reconstruct are exactly the ones that did not work. ## What it does A dated file in version control carries the same idea; the harness adds the enforcement a text file cannot supply. - **[Pre-registration](https://hadalinstruments.com/glossary/pre-registration/) before data contact.** A hypothesis is registered — its form, its data, its decision rule — before the data is fetched. [Pre-registration](https://hadalinstruments.com/glossary/pre-registration/) is worth only the provability of its ordering, so registrations are timestamped into an append-only record and the ordering is the evidence. - **The [kill ledger](https://hadalinstruments.com/glossary/kill-ledger/).** Every configuration that dies is written down at the moment it dies: what was tried, on what data, with what result, and why it was abandoned. A [kill ledger](https://hadalinstruments.com/glossary/kill-ledger/) is append-only, contemporaneous and complete — a thirty-second experiment counts the same as a month-long one, because the selection statistics do not care how long a trial took. - **False-discovery-rate accounting across the trial book.** The correction is applied across the whole book of registered trials rather than one test at a time, so a surviving result carries the significance the search actually earned, not the significance its final run displays. - **Walk-forward enforcement.** [Walk-forward validation](https://hadalinstruments.com/glossary/walk-forward-validation/) is consumable: the first pass is a test, and the fifth pass after four rounds of reshaping is fitting wearing a test’s clothes. The harness counts the passes and holds the boundary, rather than trusting memory. - **Burned-door detection.** Data fetched before its hypothesis was registered destroys the evidential value of that registration, and the harness flags the inversion. A [burned door](https://hadalinstruments.com/glossary/burned-door/) is a shape in the record — work claimed without a matching cost, a receipt that precedes its artifact — and an append-only record is what makes that shape visible. ## What it reads on my own book, right now The harness runs against this estate’s own research, and these are its live readings — not illustrations. They are also the least flattering numbers I hold, which is rather the point: an instrument that has never embarrassed its owner has not been pointed at anything. - **30 nominal tests collapse to ~6.24 effective independent ones.** Measured across the pair book. A correction applied as though thirty independent trials had been run would be answering a question nobody asked. The gap between 30 and 6.24 is the gap between a defensible threshold and a decorative one. - **Nine of ten registered trials could not have rejected at any effect size.** Every test has a resolution floor — the smallest p its own design can return. On my live trial book that floor sat *above* the level the tests were being judged at, so they were structurally incapable of producing a finding, and the online procedure spent budget on each one regardless. **Nothing was over-claimed** — every one reads “not rejected” either way — but the alpha they consumed bought nothing. - **The alpha budget shrank 13.7× across ten trials.** That is what spending a budget on tests that cannot pay out looks like from the account’s side. The third reading is the one worth dwelling on, because it is not a coding error and no amount of care would have surfaced it: a trial that cannot reject produces exactly the same output as one that looked hard and found nothing. Only an instrument that computes the floor separately can tell them apart. ## What it does not do The Epistemic Harness does not make a hypothesis true. Nothing does. It makes the evidence trail behind a hypothesis auditable — a strictly smaller claim, and the only honest one available. It is not a compliance product and it certifies nothing to any regulator. It also cannot stop a researcher lying to it — a registry records what is entered into it — which is exactly why the ordering constraints and the append-only spine are structural rather than advisory: the price of the lie is that it must be committed in advance, in writing, where it can later be found. ## Who it is for Independent quants and small research teams whose multiple-testing corrections need a trial count they can defend; allocators and prop firms reviewing an external track record who want the trial book, not the winner; and any desk that has watched a strategy die out-of-sample and been unable to reconstruct how many things were tried first. ## The ship gate No instrument is sold until it does what this page says it does. Where a page is written in the future tense, that tense is a statement about timing rather than a hedge about capability: the instrument is not finished, so it is not listed as available, not priced as available, and not sold. It waits. Nothing described in this catalogue is a placeholder that will quietly disappear. An instrument that turns out to be wrong gets a [kill-ledger entry](https://hadalinstruments.com/kill-ledger/), not a deletion — which is the only version of that promise anyone can check. ## Commercial terms Published in full, pre-launch, so they can be read and checked rather than requested. Every figure is an anticipated indication I have set and not yet ratified, and every derived figure is the arithmetic of the one above it — shown, not asserted. Nothing here is purchasable: there is no checkout on this site. ### Why Epistemic Harness is priced the way it is - **What the figure buys** — A seat on the registry spine: hypotheses registered — form, data, decision rule — before the data is fetched, timestamped into an append-only record where the ordering is the evidence; a kill ledger that takes every abandoned configuration at the moment it dies, a brief experiment on the same terms as a long one; false-discovery-rate accounting applied across the whole registered trial book rather than one test at a time; walk-forward passes counted and the boundary held rather than trusted to memory; and burned-door detection for the case where data reaches a hypothesis before its registration does. The edge is that a registry records — it does not adjudicate. It cannot make a hypothesis true, it certifies nothing to any regulator, it is not a compliance product, and it cannot stop a researcher lying to it. What it buys is that the lie has to be committed in advance, in writing, into a record that only grows. - **Why it is priced this way** — Per seat, because pre-registration is an act performed by a person and the ordering is only evidence if it is attributable — a shared login produces a trial book nobody can be held to, which is the one thing a defensible trial count cannot be. A second researcher is a second seat because a second researcher generates trials, and the trial count is the object being sold; charging for the software while the trials arrive uncounted would price the wrong thing. Recurring rather than one-off because the discipline is continuous: a registry with a lapse in it has that hole in exactly the place where the trials nobody wanted to record would have sat. - **What the alternative costs** — A dated file in version control, appended to honestly, with commits serving as timestamps. That genuinely is the idea, it costs nothing, and a disciplined researcher can run it for years — what separates the two is enforcement rather than concept, because nothing in a text file counts your walk-forward passes, applies the correction across the whole book, or notices that the data was fetched before the hypothesis was written down. The other alternative is reconstructing the trial count from memory afterwards, which is also free, and which is precisely the failure the multiple-testing corrections exist to catch: the trials that get forgotten are the ones that did not work. Every figure on this page is an anticipated indication awaiting ratification, and nothing here is purchasable. The reasoning above is published for the same reason the arithmetic below is: a price you can interrogate is worth more than a price you have to accept. ### The two-SKU split One route. This instrument has no marketplace equivalent, and the cell says so rather than sitting empty — an empty cell reads as an omission, and this is a fact about how the instrument is sold. Route | What it is | Anticipated | Marketplace SKU | None. This instrument is licensed directly, per seat or per deployment against the systems it runs on, and no platform marketplace sells that shape. There is no third-party SKU between you and it. | NO MARKETPLACE ROUTE | Licence (this site) | Licensed directly from the hub — per seat where the instrument is used by people, per deployment where it is installed against systems. Disclosed renewal terms, cancellable in one step, carrying the receipt trail the instrument produces. | £69/seat/mo · £690/seat/yr | The price rule The at-or-below-marketplace cap binds the instruments that ship through a platform. This one does not, so the cap has nothing to bind. What holds in its place is simpler and checkable: the figure above is the only figure. There is no separate, higher list price it is discounted from, nothing on this site has ever been sold at any price, and so there is no “was” anywhere to strike through. ### The commitment ladder One month less paid per year of service, per step, taken off this instrument's own anticipated annual rate of £690/seat/yr . The rate is held at the figure you sign for the whole term, so a multi-year commitment fixes the price as well as reducing it. Term | Months paid per year | What it means | Anticipated | Monthly billing | 12 | £69/seat/mo × 12. Twenty per cent more than the annual term — that is the uplift for paying monthly, not a discount for paying annually. | £828/seat/yr | 1-year term | 10 | Twelve months of service for the price of ten. This is the annual rate every band below is taken off. | £690/seat/yr | 2-year term | 9 | Ten per cent off the annual rate, held at that figure for the whole term. | £621/seat/yr | 3-year term | 8 | Twenty per cent off the annual rate, held at that figure for the whole term. | £552/seat/yr | Three years for what two years of monthly billing costs Eight months paid per year, across three years, is 24 months paid for 36 months of service — one year in three carries no charge. On this instrument the three-year term totals £1,656 per seat. That is 24 × £69 = £1,656 , and two years billed monthly is £828 × 2 = £1,656 . The same money. It buys three years instead of two, and the arithmetic is on the page so you can check it rather than take it. ### Seat bands This instrument is licensed per seat, so the seat count is a band. Bands are taken off the annual rate. Three of the twenty-one instruments are per-seat; on the rest a seat band would be a category error, and the scaling axis is entities instead. Seats | Band | Anticipated, per seat | 1–4 seats | The annual rate. | £690/seat/yr | 5–9 seats | Ten per cent off the annual rate. | £621/seat/yr | 10–19 seats | Twenty per cent off the annual rate. | £552/seat/yr | 20+ seats | Past nineteen seats the licence is negotiated as a group licence rather than counted seat by seat. | Group licence, negotiated | The two bands multiply — they do not add A seat band and a term band are applied one after the other, not summed. Twenty per cent and twenty per cent is not forty per cent off; it is sixty-four per cent of the annual rate. Worked end to end on this instrument, at 12 seats on a three-year term: - **Annual rate per seat** — £690 - **Seat band, 10–19 · ×0.8** — £552 - **Term band, 3-year · ×0.8** — £441.60 per seat per year - **× 12 seats** — £5,299.20 per year - **Over the three-year term** — £15,897.60 None of those figures is rounded to look tidy. A buyer who can reproduce the number has a reason to trust the vendor who published it, and a rounded arbitrary figure gives them nothing to reproduce. ### Commercial routes Pricing scales on seats and on term — bands from five seats, a group licence past nineteen, set out above with its own figures. [Licensing — seat bands, group licences and multi-year terms in full](https://hadalinstruments.com/licensing/) ### Support - Tier **Standard** at this instrument’s base contract — ticket, first response targeted at three business days . Support tier follows the annual contract value, not the price of a single unit — more seats, a suite licence or a group agreement raise the contract value and can raise the tier with it. [Support — the tiers, the ticket-only channel model, and what a target does and does not promise](https://hadalinstruments.com/support/) ### Trial mechanics The trial runs on a seat — thirty days, disclosed in plain words before you start it and cancellable in one step; none of it is live yet. [Trial mechanics in full, by delivery class](https://hadalinstruments.com/licensing/#trial). ### Read before you commit The documentation is published ahead of the product on purpose — intended behaviour is only a commitment if it exists first. Start with [installation and first run](https://hadalinstruments.com/docs/install/), then [the limits](https://hadalinstruments.com/docs/limits/): the conditions under which this instrument refuses to produce a number are the part worth reading before you pay. The full centre is at [/docs/](https://hadalinstruments.com/docs/), and the [support model](https://hadalinstruments.com/support/) states what a ticket does and does not cover. ## Questions and answers Answered from what this instrument publishes about itself. Nothing below is attributed to a customer, because there are none yet. ### What stops a researcher lying to the registry? Nothing can, and the harness does not claim otherwise — a registry records what is entered into it. What the ordering constraints and the append-only spine change is the price of the lie: it must be committed in advance, in writing, where it can later be found. ### Why must a hypothesis be registered before the data is fetched? Because pre-registration is worth only the provability of its ordering. Data fetched before its hypothesis was registered destroys the evidential value of that registration, and the harness flags the inversion. ### Why does a thirty-second experiment belong in the kill ledger? Because the selection statistics do not care how long a trial took. The ledger is append-only, contemporaneous and complete, and every correction for multiple testing conditions on the count it holds. ### Can I licence this instrument today? No. Nothing on this site is on sale — there is no checkout, no card capture, and no product account to create. Every figure on this page is an anticipated indication I have set so it can be read and compared, not an offer, and final pricing awaits my ratification. The launch list is the only thing you can join today. ## Change log NOT YET PUBLISHED Epistemic Harness has not shipped, so there is nothing to record. When it does, every version lands here — dated, append-only, written by a person, and including the changes that removed a capability rather than added one. ## Where this sits Epistemic Harness is one of the instruments in the [Honesty Stack suite](https://hadalinstruments.com/suites/stack/). How that suite measures — the per-instrument battery, and the receipts each measurement will carry — is set out in the [Honesty Stack methodology](https://hadalinstruments.com/methodology/stack/), part of the site-wide [measurement methodology](https://hadalinstruments.com/methodology/). ### Also in the Honesty Stack suite Epistemic Harness shares the Honesty Stack suite with three other instruments. - P5 [Agent-Governance Constitution](https://hadalinstruments.com/instruments/agent-governance-constitution/) The rails that caught a forged ratification — packaged. - P15 [Provenance-Grade Ingestion](https://hadalinstruments.com/instruments/provenance-grade-ingestion/) Vendors sell data. Nobody sells provenance discipline around data. - P19 [Fail-Visible Ops](https://hadalinstruments.com/instruments/fail-visible-ops/) In trading infrastructure, silence is a fault — not a status. ### Research behind this instrument Epistemic Harness draws on nine research notes on this site. - [The edge that was real and worth half a pip](https://hadalinstruments.com/research/the-edge-that-was-real-and-worth-half-a-pip/) Asked as: "can a trading edge be statistically significant but not profitable" - [Do order blocks predict anything?](https://hadalinstruments.com/research/do-order-blocks-predict-anything/) Asked as: "do order blocks actually predict anything" - [Do round numbers matter in forex?](https://hadalinstruments.com/research/do-round-numbers-matter-in-forex/) Asked as: "do round numbers matter in forex trading" - [Does a wick rejection mean anything?](https://hadalinstruments.com/research/does-a-wick-rejection-mean-anything/) Asked as: "does a wick rejection actually mean anything" - [Does COT positioning predict reversals?](https://hadalinstruments.com/research/does-cot-positioning-predict-reversals/) Asked as: "does cot positioning predict reversals" - [The break that cleared neither bar](https://hadalinstruments.com/research/the-break-that-cleared-neither-bar/) Asked as: "how do you know a detected regime change is real" - [The floor that never fired](https://hadalinstruments.com/research/the-floor-that-never-fired/) Asked as: "how do you know a check that never fails is working" - [The level that remembered nothing](https://hadalinstruments.com/research/the-level-that-remembered-nothing/) Asked as: "do price levels remember what happened there before" - [How do I test a backtest for overfitting?](https://hadalinstruments.com/research/how-to-test-a-backtest-for-overfitting/) Asked as: "how to test a backtest for overfitting" --- ## Raw artifact — NOT PUBLISHED FOR THIS PAGE No downloadable artifact ships with this page. Eight published measurements do, each content-hashed so a reader can verify the figures independently. Where a measurement is published here without one, that is a gap rather than a policy, and it is stated rather than left to be noticed. [Measurements that ship their data](https://hadalinstruments.com/research/) [Join the launch list](https://hadalinstruments.com/pricing/#waitlist)[How this suite measures](https://hadalinstruments.com/methodology/stack/)