Service · Digital assets

Technical crypto due diligence.

StatGazer independently reviews digital-asset models and strategy backtests before an allocator, fund, or investment committee relies on the result. We focus on the evidence underneath the performance claim: data timing, universe construction, costs, liquidity, model risk, and reproducibility.

Scope a technical review Open review standard

Technical diligence only. StatGazer does not provide investment advice, trading signals, execution, custody, or performance guarantees.

Sample work product

Does the result survive the market it claims to trade?

Start with the reported result, then rebuild it under declared daily implementation assumptions for a 365-day market. This deterministic synthetic perpetual-futures example isolates fees and spread/slippage, the funding ledger, and size-dependent liquidity impact. It is not client work, a real strategy, or investment evidence.

StatGazer / Digital-Asset Diligence Memo Ref. SYNTHETIC-CRYPTO-001

Illustrative · synthetic data · not a client strategy

Performance reconstruction extract

Perpetual-futures strategy — the gross headline weakens after implementation.

Question: At the proposed synthetic size, how much of the gross result remains after explicitly modeled implementation drag?
Setup: 12 generic perpetual contracts · 730 synthetic daily observations · dollar-neutral 1.0× gross exposure · no real tokens, venues, managers, or dates.
Reconstruction: Weights use prior-day synthetic signals; daily P&L uses next-day synthetic returns. Fees and spread/slippage, position-weighted funding, and liquidity impact are applied from synthetic trade and position ledgers.
Conventions: Annualized mean return = 365 × daily mean; Sharpe uses a 0% cash hurdle and √365; max drawdown uses compounded daily returns.

Gross-to-net results from an illustrative synthetic perpetual-futures reconstruction, not client performance, a real strategy, or investment evidence.
Stage	Annualized mean	Sharpe	Max drawdown
Synthetic reported gross	+30.2%	1.74	−15.4%
After fees + spread/slippage	+23.8%	1.37	−16.5%
After funding ledger	+20.7%	1.20	−16.9%
Reproduced net incl. liquidity impact	+14.2%	0.82	−18.0%

Annualized synthetic drag: fees + spread/slippage −6.4 pp; funding −3.1 pp; model-based liquidity impact −6.5 pp. Average one-way turnover 28.4%; synthetic trade/ADV participation median 0.42%, 95th percentile 1.78%.

Declared synthetic schedules. Fee 1.5 bp + base spread/slippage 1.0 bp + 0.6 bp × clipped prior-day volatility state, per absolute traded weight; funding 1.0 bp × current synthetic signal + a normal shock with σ = 0.4 bp; liquidity impact 32.5 bp × √(synthetic trade/ADV participation).

Material · implementation-sensitive

Annualized mean return and Sharpe are 53% lower than the synthetic gross headline.

Evidence. With the return path held fixed, explicit implementation deductions reduce annualized mean return from +30.2% to +14.2% and Sharpe from 1.74 to 0.82. Max drawdown deepens from −15.4% to −18.0%.

Limit. All contracts, signals, returns, funding, and liquidity inputs are synthetic. Funding can be a debit or credit; venue and contract rules differ. The liquidity term is a model-based proxy at an illustrative size, not observed fills or a capacity estimate. This is not a client result, a backtest of a real strategy, a track record, a forecast, or evidence of investability.

Action. Reconcile every trade and funding cash flow at venue and contract level; rerun against point-in-time fees, depth, and volume at proposed capital; document the gross-to-net bridge and size sensitivity before relying on the headline.

Waterfall chart for a synthetic 12-contract perpetual-futures strategy. Annualized mean return starts at plus 30.2 percent gross; fees and spread reduce it by 6.4 percentage points, funding by 3.1, and model-based liquidity impact by 6.5, leaving a reproduced net plus 14.2 percent. This is not real performance. — Code-rendered deterministic synthetic example. Not client data, a real strategy, a track record, investment evidence, or a forecast. RNG_SEED 20260818. Open the full-size chart in a new tab

Scope

Evidence for the decision, not a token endorsement.

The scope is anchored to a specific technical question: whether a strategy result is reproducible, whether assumptions are decision-material, and what remains unresolved before capital or governance depends on the claim.

What we examine

Point-in-time market, reference, on-chain, exchange, and vendor data.
Token universe construction, listing and delisting history, and survivorship.
Signal timing, 24/7 market calendars, rebalance logic, and look-ahead risk.
Fees, spread, slippage, funding, borrow, market impact, liquidity, and capacity.
Out-of-sample behavior, parameter stability, stress tests, and regime sensitivity.

What you receive

A technical memo separating verified evidence from unresolved claims.
A reproducibility artifact where data and access permit one.
An issue register ranked by impact on the decision, not cosmetic severity.
A record of assumptions, limitations, and evidence still required.
A technical readout for investment and quantitative stakeholders.

Decision contexts

Useful before allocation, acquisition, or scale.

Independent review is most valuable while the decision can still change. The objective is not to predict token prices; it is to determine whether the quantitative evidence supports the strategy or manager claim being evaluated.

Good fit

An allocator is evaluating a digital-asset manager or quantitative strategy.
A fund is increasing capital behind a backtest that has not been independently reproduced.
A committee needs a technical record of data, model, liquidity, and implementation risks.
An internal quant team wants independent challenge before launch or scale.

Not a fit

You want a buy, sell, hold, allocation, or token recommendation.
You need trading signals, execution, brokerage, exchange, wallet, or custody services.
You want promotion, fundraising claims, or a guaranteed performance outcome.
You need legal, tax, regulatory, sanctions, smart-contract security, or financial-statement assurance.

Review design

Separate market risk from research error.

A strategy can lose money even when the research is sound, and it can make money temporarily despite a broken backtest. The review separates economic exposure from defects in data, timing, implementation, or validation so the decision record remains useful after market conditions change.

Claim map

We identify the performance, risk, diversification, liquidity, and scalability claims that matter to the decision and the evidence each one requires.

Independent challenge

We reproduce material results, test plausible alternatives, and document which assumptions or missing artifacts could change the conclusion.

Decision record

The final memo distinguishes confirmed defects, material assumptions, limitations, and unresolved questions so follow-up work has a clear owner.

Professional boundary. Independent technical model, data, and backtest review only — not investment advice, trading signals, brokerage, execution, exchange, wallet, custody, legal, tax, regulatory, smart-contract security, financial-statement audit, or performance assurance.

Related routes

For allocators evaluating a digital-asset manager: the same read, engaged by you →

Broader diligence

Need a manager or platform review beyond digital assets?

Quant due diligence covers strategy evidence, models, data infrastructure, monitoring, documentation, and governance across asset classes.

See quant due diligence

Next step

Start with the claim and the decision it has to support.

Send only high-level context in the first message. We can sign an NDA before reviewing strategy logic, holdings, source code, data, or non-public manager materials.

Scope a technical review