Questions before a brief
Short answers on what a run costs, what it checks, what the report can and cannot say, and how your material is handled.
01
What is Validraft?
Validraft is an independent validation desk for trading hypotheses. You describe a strategy idea in a brief; the desk scopes the data, assumptions and checks it needs, runs the backtest and validation on its own research data lake, and delivers a descriptive report with the evidence and its limits. It is research and simulation, not investment advice.
How it works02
How much does a backtest or a validation cost?
Research studies start at $79. A candidate backtest costs $79–$299, a final validation $179–$249 and a diagnostic review $79–$199. These prices cover the run before data-depth and universe adjustments; the final price is agreed in the scope before work starts, and a higher price never buys a more favourable verdict.
Pricing03
What is the difference between a backtest and a final validation?
A candidate backtest tests a rule, or optimizes pre-declared parameters, to develop and select a candidate; its report is candidate evidence only. A final validation takes one frozen candidate and tests it on protected out-of-sample or lockbox data, with a stress review on the same freeze. Only a final validation can produce a final GO decision.
Runs and reports04
What does a validation check?
Each run gets a profile drawn from 72 public controls in 10 areas, from economic rationale and point-in-time data to execution costs, statistical validation, stress testing and governance. They include walk-forward analysis, look-ahead and survivorship checks, permutation tests, the Deflated Sharpe Ratio, the Probability of Backtest Overfitting and Monte Carlo path risk. The strategy family and the run type decide which controls are required, conditional, diagnostic or not applicable.
Methodology05
What verdicts can a validation report give?
Three, and all three are delivered as a full report. GO: Every blocking gate passed and a named reviewer signed off the result. Conditional GO: Gates passed, but a documented, ticketed threshold override is on record and must be read with the result. NO GO: At least one blocking gate failed. You still receive the report — which gate, which value, and why. A GO means every blocking gate passed within the declared scope and a named reviewer signed off. It is not a recommendation to trade, a profitability guarantee, a safety claim or approval for live trading.
Sample reports06
Do I need a coded algorithm?
No. Discretionary, systematic and hybrid hypotheses all start the same way: a brief in plain English with the asset, timeframe, rules and what success looks like. The desk turns it into a precise, testable specification and checks data coverage and feasibility before any compute runs.
Strategy families07
Can I check feasibility before paying?
Yes. An initial feasibility enquiry needs no account, code or payment: describe the research question and the desk replies with what is feasible and any missing inputs. A written feasibility check, which maps the idea to the correct run type, data requirements and known blockers, is a separate $19 service agreed before any charge. Data, deliverable, price and timing are agreed before work starts.
Check feasibility08
Which data does Validraft use?
The research lake holds 3.9 TB of raw market, fundamental, macro and alternative data from 6 providers (Databento, Massive, Financial Modeling Prep, ALFRED / FRED / ECB, FMP News and Quiver Quantitative), with history from 1990 to 2026. It covers CME futures, US equities, fundamentals, macro series, news sentiment and alternative data. If a hypothesis needs a feed the lake does not hold and it is available and licensable, the desk sources it and documents the data assumptions before validation starts.
Data coverage09
Is my strategy kept confidential?
Briefs and submitted strategy details are treated as confidential by default, subject to any written agreement and legal obligations. They stay tied to the engagement they came from: client hypotheses are not added to a public template library, a marketing example or a cross-client benchmark, and they are not used for training. An NDA is available on request, and may be required for sensitive, institutional or investor-facing work.
Privacy notice10
Is a Validraft report investment advice?
No. Reports are descriptive, not prescriptive: they describe historical performance, statistical evidence, validation outcomes, assumptions and limitations. They are not recommendations to buy, sell or hold any instrument, and the desk does not execute orders, manage capital or run paper or live trading. A backtest is a historical simulation; it cannot prove that a strategy will work in the future.
Legal notice11
How long does an engagement take?
Delivery timing is agreed in the scope, before work starts. It depends on the run type, the data the hypothesis needs and the desk's capacity; a rush option is listed on the pricing page.
Pricing12
What happens after the report is delivered?
The report is posted to your workspace. Optional monitoring then compares new observations with the report's baseline and posts descriptive snapshots: Monitor Lite at $19 a month (monthly), Monitor Standard at $39 a month (weekly) or Monitor Desk at $79 a month (monthly / quarterly). Monitoring notes describe drift; they are never trading instructions.
Monitoring plansNext step
Your hypothesis. Independently validated.
Start with a sample, then ask whether your question is feasible. We agree on the analysis, data, price and delivery before any work starts.
72
Public controls
Documented on the methodology page
6
Validation phases
From data integrity to sign-off
3
Possible outcomes
Each one is a deliverable
$79
Starting price
Research and backtests, scoped first