Backtest Discipline accepts a small number of individually scoped evaluation and reproduction requests.
The service is for people who have a concrete published strategy, backtest, research paper, or quantitative claim and want an independent answer to a narrower question: can this result be reproduced, or does the strategy still meet a predeclared test when evaluated under an agreed specification?
This is not a signals service, a strategy marketplace, or an automated backtesting platform. Strategy Lab is research software used to support bounded investigations; not every proposed strategy can be onboarded without disproportionate custom work.
What can be commissioned
Suitable requests usually start with something concrete enough to inspect before any evaluation begins, for example:
- a published academic or practitioner strategy with explicit rules;
- a backtest whose headline result you want independently checked;
- a numerical result or statistic that should be reproducible from identified public evidence;
- an existing strategy for which the data, execution model, benchmark, costs, and decision rule can be made explicit.
A request may be declined if the source is too ambiguous, required data cannot be obtained appropriately, implementation would require building a new general-purpose system, or the question cannot be tested without silently making material assumptions.
How the work is scoped
Each accepted engagement is deliberately narrow.
- Feasibility review. You send the source material and the question you want answered. I determine whether a defensible test is possible with the existing research capabilities and available evidence.
- Written test contract. Before numerical evaluation, the relevant interpretation, data source, period, execution timing, transaction costs, benchmark, acceptance criteria, and known substitutions are fixed in writing.
- Deterministic evaluation. The agreed test is run through the research workflow with provenance and preserved outputs. The result is reported whether favorable, unfavorable, or inconclusive.
- Deliverable. You receive a concise written report explaining the question, specification, result, important assumptions, limitations, and selected supporting evidence appropriate to the engagement.
Payment buys the work and the documented process. It never buys a favorable verdict, a trading recommendation, or a claim of profitability.
Private or publishable work
A commissioned evaluation can begin as private work. If public publication is desirable, that must be agreed explicitly before publication.
Public commissioned studies will identify that the work was funded and will not suppress an unfavorable or inconclusive result after the research question and publication terms have been agreed. Confidential material, private credentials, proprietary datasets, and unnecessary client information are not published.
What this service does not provide
- live or paper trading;
- order execution or exchange-account access;
- investment advice or personalized trading recommendations;
- guaranteed profitability or strategy certification;
- open-ended strategy optimization after seeing the result;
- arbitrary strategy uploads into a hosted platform;
- data redistribution rights that the underlying provider does not grant.
The existing research record shows the type of work and evidence discipline behind the service. The methodology explains the general controls and their limits.
Request a feasibility review
Email contact@backtestdiscipline.com with:
- the strategy, paper, repository, or backtest you want examined;
- the exact question you want answered;
- any data or market constraints that matter;
- whether the result needs to remain private or could be published;
- any deadline that genuinely affects usefulness.
The first step is only a feasibility and scope discussion. No research run is authorized until the test is sufficiently explicit and the scope has been agreed.