Independent strategy validation

Adalia tests systematic trading strategies on Binance perpetual futures for clients who want an outside check before capital is committed. You receive a written report with the figures the evidence supports, and the figures it does not.

Who it is for

Allocators and funds

Offered a strategy by a third party, who want its evidence checked.

Trading teams

Who want an outside test of their own strategy before they scale it.

What we test

Most reviews check the statistics of a backtest. We rebuild the market the strategy would have traded in, and test it there. This includes correcting for backtest survivorship bias: testing on the full set of perpetuals that existed at the time, not only those that survived to today.

No hindsight in the coin list. The strategy is tested on the perpetuals that were listed at the time, including those since delisted, not on today’s survivors. This removes the survivorship bias common in backtests that only include currently listed symbols.

Trading costs at your size. Slippage and capacity are estimated at the capital you intend to run: from order-book depth where historical depth is available, and from traded volume where it is not. The report says which method was used for each period.

Independent rebuild. We re-implement your rules from the written specification and compare the output with your own results.

Every trial counted. All the variants that were tried are counted, and the result is discounted for them.

Leak checks. We test whether any signal uses information that was not available at the time of the trade.

What you receive

A written conclusion on the evidence: supported, supported with reduced figures, or not supported, with the reasons.
The corrected figures for return, risk and capacity, next to the figures you were shown.
A reproducible record: the tests can be rerun on the stored data and give the same result.

Our own research follows the same rules. The test plan is written before the results, every variant is counted, and the ideas that failed - if authorized - are published on perpswatchdog.com.

How it works

A mutual NDA. Your rules, code and results are used only for this review. We do not trade them or use them in our own research.
You send the written specification and, if you have them, your code and results.
We agree the scope, a fixed fee and a delivery date in writing before work starts.
We run the tests and deliver the report. The report is for your own use. Sharing it needs our written consent.

Scope

The report reviews the method and the evidence behind a strategy, on Binance USDT-margined perpetual futures. Whether to invest in it, and whether it suits you, stays your decision. The report is not investment advice and does not forecast future results.

Disclaimer. This is a technical review of method and evidence, not investment advice. Past results, corrected or otherwise, do not guarantee future performance. Decisions about capital allocation remain yours.

Request an audit