ACDS DecisionAgricultural Capital Decision Standard

Public validation register

See what has—and has not—been validated.

This register separates internal software testing, a second calculation implementation, agricultural-professional review, and field testing for every current ACDS model area. Passing either internal stage does not mean the model has been independently certified or validated on a real farm decision.

Ask about participatingReview model governance

Status by model area

A claim-by-claim validation record.

“Passed” means the named, dated control completed. For calculation reconciliation, that means a separately implemented formula path, not third-party assurance. “Not completed” means no acceptable record exists for that stage.

Passed · documented internal evidence Not completed · no qualifying record Not applicable · no numeric calculation
ACDS-F-01

Equipment Investment Analyzer

v0.25.0
Model areaInternally testedSecond-calculation reconciliationReviewed by an agricultural professionalField-testedOutstanding validation work
Whole-life economics and alternative ranking
PassedAutomated production release gate · September 4, 2026
Not completedNo documented reconciliation to a separately implemented calculation.
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Reproduce keep, purchase, lease, rental, and custom-hire results in an independent calculation; investigate every difference before blind testing.
Capacity, downtime, and timeliness
PassedAutomated production release gate · September 4, 2026
Not completedNo documented reconciliation to a separately implemented calculation.
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Compare modeled work windows with actual harvest or field records and obtain an equipment professional’s review of capacity definitions and downtime treatment.
Financing and annual cash requirements
PassedAutomated production release gate · September 4, 2026
Not completedNo documented reconciliation to a separately implemented calculation.
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Reconcile loan and lease schedules with independently prepared lender or accountant calculations using identical payment timing and terms.
Sensitivity, break-even, and decision thresholds
PassedAutomated production release gate · September 4, 2026
Not completedNo documented reconciliation to a separately implemented calculation.
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Independently reproduce threshold results, then compare forecast decision drivers with actual outcomes in qualifying field cases.
ACDS-F-02

Farmland Acquisition Analyzer

v0.91.0
Model areaInternally testedSecond-calculation reconciliationReviewed by an agricultural professionalField-testedOutstanding validation work
Buy-rent-wait economics, price/rent, and supported price
PassedAutomated production release gate · September 4, 2026
Passed80-check separate-formula reconciliation across six scenario families · September 4, 2026 · internal technical evidence, not third-party assurance
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Reconcile the same outputs to a completed lender or accountant case and investigate definition or timing differences before blind testing.
Generational wealth and indefinite ownership sensitivity
PassedAutomated production release gate · September 4, 2026
Not completedAutomated invariants and direct formula checks pass, but no separately implemented professional-case reconciliation has been documented.
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Reconcile retained-land wealth and the financed growing-perpetuity comparison to an independently prepared long-horizon case; test permanent-growth, succession, and rental-continuity assumptions with agricultural professionals.
Loan amortization, DSCR, and source-of-funds
PassedAutomated production release gate · September 4, 2026
Passed80-check separate-formula reconciliation across six scenario families · September 4, 2026 · internal technical evidence, not third-party assurance
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Match a lender-prepared debt schedule and DSCR definition, then document any policy-driven difference from the ACDS classification.
Liquidity, leverage, and borrowing headroom
PassedAutomated production release gate · September 4, 2026
Passed80-check separate-formula reconciliation across six scenario families · September 4, 2026 · internal technical evidence, not third-party assurance
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Compare post-closing liquidity, security availability, and repayment-limited headroom with a lender-prepared completed case.
Seasonal cash and debt maturity
PassedAutomated production release gate · September 4, 2026
Not completedNo documented reconciliation to a separately implemented calculation.
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Compare modeled operating-line pressure with dated farm statements and actual monthly or quarterly cash timing across qualifying cases.
Parcel evidence, offer structure, and constraints
PassedAutomated production release gate · September 4, 2026
Not applicableEvidence sufficiency and professional judgment do not have one numeric calculation to reconcile.
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Obtain agricultural lender, accountant, appraisal, and farm-advisor review of evidence boundaries, conditions, and practical decision use.
Lender underwriting file and coordinated downside
PassedAutomated production release gate · September 4, 2026
Not completedAutomated integrity tests cover the underwriting helper, but no independently implemented lender-file calculation has been documented.
Not completedNo dated review by an objective agricultural professional.
Not completedNo qualifying development or blind case recorded under the public protocol.
Reconcile a sanitized completed case with an agricultural lender, including normalized repayment cash, covenant definitions, collateral policy, coordinated downside assumptions, risk coverage, structured alternatives, and closing cash timing.
Technical reconciliation · September 4, 2026

Farmland core formulas · current v0.91.0

Passed · 80 checks

What was independently recalculated in code

  • Buy, rent, and wait NPV plus economic supported price
  • Loan payments, remaining balances, DSCR, and liquidity
  • Price/rent yield and the exact break-even-rent crossover
  • Borrowed closing cash, lender fees, and operating-credit effects
  • Collateral-limited and repayment-limited borrowing headroom
  • Base, all-cash, high-rent, high-price, shortage, and downside cases
  • Retained-land generational equity versus rent-and-invest wealth is tested separately from the 80-check core NPV reconciliation
  • The indefinite-ownership check is tested separately for direct growing-perpetuity reconciliation, invalid growth, narrow spreads, and no terminal land-value double count

Corrections made before this release

The downside headroom screen now recalculates stressed land and source-of-funds payments directly. The post-closing asset ratio now includes lender fees correctly, and a cash deficit remains visible in working capital even after current assets reach zero.

This is a separate-formula technical reconciliation. It does not count as a development farm case, blind case, lender review, appraisal, approval, audit, or assurance engagement.

Version 0.20.0 also includes automated integrity tests for the lender-file calculations: accrual cash normalization, global DSCR, collateral value reductions and advance rates, coordinated downside results, structured alternatives, and HST cash timing. Those tests are separate from the 80-check core-model reconciliation and do not represent lender or professional validation.

Version 0.21.0 adds an explicit parcel-only scope. Its integrity gate confirms that farm financial inputs are not required and that the parcel Decision File cannot display whole-farm DSCR, liquidity, leverage, borrowing headroom, or a credit-ready conclusion. This changes workflow and disclosure, not the reconciled core economic formulas.

Version 0.22.0 adds review focuses for credit, appraisal, accounting, and farm-advisory work plus a live section queue. Automated integrity checks confirm that the selected focus persists in saved projects, legacy files receive a safe default, and the workflow changes navigation and visible detail without changing the calculation model.

Version 0.23.0 adds sector-adaptive operating profiles and a deterministic, device-only CSV or TSV import for current financials, historical results, and debt schedules. Integrity checks cover sector normalization, strategic adjustments, capital mapping, local import parsing, and the parcel-only professional scope boundary. These additions improve workflow and documentation without changing the reconciled core economic formulas or issuing a professional opinion.

Current version 0.91.0 deducts recurring water and energy operating costs separately for specialty and irrigated productive land. Automated checks cover the main use, added productive areas, annual cash-flow timing, and the saved-project transition that prevents silent double-counting. The remaining reconciled acquisition, financing, liquidity, and ranking formulas are unchanged while the 10 development and 5 blind external cases are open.

Development cases0 / 10
Blind cases0 / 5
Public findings reportNot issued

Independent case series not yet completed. This is model validation—not a testimonial campaign or a request to approve a transaction. Counts change only after a sanitized case satisfies the protocol below.

Who can contribute

Completed decisions and objective reviewers.

Farm business

A recent equipment or farmland decision

A completed or seriously evaluated proposal with enough records to reconstruct the assumptions and outcome.

Professional reviewer

Agricultural lending or accounting experience

A lender, accountant, farm advisor, appraiser, or other qualified reviewer able to challenge definitions, calculations, and limits.

Case requirement

Permission and a sanitized record

No public naming is required. Confidential or identifying information must be removed before any case is transferred.

Comparison protocol

A repeatable six-stage review.

Development cases may identify corrections. Blind cases are held back until the model version is frozen so the comparison is not tuned to the answer.

  1. 01

    Agree on scope and permission

    Define the decision, reviewer role, confidentiality boundary, permitted use, and whether results may be reported only in aggregate.

  2. 02

    Remove identifying information

    Use a sanitized case without names, account numbers, legal descriptions, contact details, or unnecessary third-party information.

  3. 03

    Normalize definitions

    Reconcile cash available, debt service, current portions, lease treatment, timing, evidence dates, and valuation basis before comparing outputs.

  4. 04

    Run the frozen ACDS version

    Record the model version, inputs, warnings, recommendation, constraints, and calculation fingerprint without changing formulas during the case series.

  5. 05

    Compare and investigate

    Document every difference against the existing professional analysis, including definition differences—not only arithmetic differences.

  6. 06

    Publish aggregate findings

    Report case counts, procedures, corrections, unresolved limitations, and reviewer roles without naming participants or implying assurance not performed.

Information boundary

Contact ACDS before sending any case material.

Do not email complete financial statements, banking details, legal descriptions, customer information, employee information, or identifiable third-party records. Initial contact should describe only the type of decision and the reviewer role.

Contact support@acdsdecision.ca