PDF page 1 of 4
Manager Desk Read
Systematic equity strategy
SAMPLE-002 | 25 September 2026 | Version 1.2 | Illustrative work product
Documents only. No manager strategy code was executed, and performance reproducibility was not independently verified. All materials and figures in this sample are synthetic; this is not client work or a real track record.
Decision this read supports
Initial screening of a proposed $20 million allocation to SAMPLE-MANAGER-01, a fictional scenario. The deck presents 10.7% CAGR and Sharpe 1.44 after estimated trading costs; the documents do not reconcile that basis. The figures match returns labelled gross. Deducting the stated 60-basis-point annual cost budget evenly across months gives about 10.1% CAGR. The configuration was selected from 48 variants on the same five years.
Scope and access
Synthetic allocator brief D0; manager deck D1; 60 gross backtest months D2 (2021 to 2025); questionnaire D3; cost note D4; six manager-reported live months D5 (January to June 2026). No strategy code, trades or independent account statements were supplied.
Principal findings
01 Material. The net headline is not substantiated. Applying the stated 60-basis-point budget evenly across months gives 10.1% CAGR, not the displayed 10.7%. This is a conditional reconciliation test. See page 2.
02 Material. Selection reuses the evaluation history. No untouched period was supplied. If the 48 variants form the complete fixed family and each individual test is valid, Bonferroni gives p = 0.101. See page 3.
03 Material. Turnover implies a different cost budget. The documents imply annual cost budgets of 60 versus 180 basis points. Available capacity for the proposed $20 million remains unsubstantiated. See page 3.
What remains unresolved
Net returns need a monthly cost bridge (Q1); selection needs the complete trial log (Q2); costs need trade records (Q3); timing and universe need dated histories (Q4); capacity needs current assets and liquidity workpapers (Q5); live results need independent statements (Q6).
Questions for the manager
The six requests on page 4 identify the records needed to resolve these gaps and can be forwarded as written. Missing evidence does not establish a code defect.
Reviewer
Evgenii Azarov, StatGazer. Prepared 25 September 2026. This note assesses technical evidence; it does not recommend whether to invest.
PDF page 2 of 4 · SAMPLE-002 · Synthetic, not client work.
The net return claim is unreconciled
FINDING 01 | MATERIAL | RETURN BASIS NOT RECONCILED
Evidence
D1.1 labels the 2021 to 2025 results after estimated trading costs and before management and performance fees. The deck's rounded statistics match D2.1, which is explicitly gross of all costs and fees. That match alone does not establish their provenance.
| Statistic | Deck labelled net | Recomputed gross |
|---|---|---|
| Compound annual growth rate | 10.7% | 10.73% |
| Annualized Sharpe ratio | 1.44 | 1.44 |
| Maximum month-end drawdown | 7.4% loss | 7.449% loss |
| Annualized volatility | 7.3% | 7.30% |
Sources: D1.1 and D2.1. Calculations use unrounded inputs. Drawdown of 7.44853% rounds to the deck's 7.4%; it is measured at month end. Sharpe uses the stipulated zero cash return, not an observed cash benchmark.
Test against the stated cost budget
Assume the annual cost budget is deducted evenly: subtract one twelfth of it from each gross monthly return, then compound. Under that convention, the highest annual cost consistent with a CAGR that rounds to 10.7% is about 7.6 basis points a year. The lower rounding boundary is 10.65%.
D4.1 budgets 60 basis points a year. Applying that budget gives 10.076% CAGR, or 10.1%. The deck therefore does not reconcile under this convention. A budget is not a record of costs deducted; the monthly allocation is an explicit assumption. This scenario is not verified net performance.
Action
Request the actual monthly gross-to-net bridge, cost timing and fee terms. Identify trading costs, financing, borrow and manager fees separately; reconcile the deck, the selected model and D4.1. The bridge should show whether the gap is a labelling issue or a different cost calculation.
What the short live record can establish
D5.1 compounds to +1.68% for January to June 2026. These fictional manager-reported months are after commissions but before other costs and fees. Their period and basis differ from the gross backtest; no administrator or broker corroboration was supplied.
This checks arithmetic and comparability. Six months cannot establish persistence or deterioration here. We do not annualize this segment or join it to the backtest.
Calculation conventions
CAGR compounds 60 monthly returns. Annualized volatility uses the sample standard deviation times the square root of 12; Sharpe uses mean divided by that standard deviation with the same annualization. Full formulas and conditional checks are in the evidence package.
PDF page 3 of 4 · SAMPLE-002 · Synthetic, not client work.
Selection and cost assumptions
The evaluation history was also used for selection
FINDING 02 | MATERIAL | EVALUATION DESIGN
Evidence. D3.1 reports choosing the strongest Sharpe among 48 variants on the same 60 months shown in D1. No untouched evaluation period or returns for the other variants were supplied.
Check. The nominal zero-mean t test gives t = 3.22, 59 degrees of freedom and two-sided p = 0.0021 under independent, identically distributed normal monthly returns. If the 48 variants form the complete fixed search family and each test is valid, Bonferroni-adjusted p = 0.101 (48 times the unrounded p-value).
Why it matters. Bonferroni does not require independent variants and can be conservative. Under the stated conditional 48-test procedure, p = 0.101 does not pass the 5% threshold; this is not evidence that no effect exists. Undisclosed or adaptive research choices, or invalid individual tests, can invalidate the calculation.
Action. Request the full trial log, selection criterion, design-freeze date and evaluation history. Check for an untouched period and disclose subsequent tuning. The 48-test calculation is conditional until the search history is established.
The two turnover figures imply different cost budgets
FINDING 03 | MATERIAL | INCONSISTENT INPUTS
Evidence. D4.1 uses turnover of 6 times NAV; D3.2 states 18 times. Both count absolute traded notional divided by average NAV, with each purchase and sale counted once. At 10 basis points per unit:
| Input and implied cost budget | Cost note D4 | Questionnaire D3 |
|---|---|---|
| Annual turnover | 6 times NAV | 18 times NAV |
| Implied annual cost budget | 60 bps | 180 bps |
Why it matters. The linear gap is 120 basis points per year. It is not a measured CAGR reduction. Fees, financing and borrow are excluded; market impact is not independently modeled and its treatment within the budget is unspecified.
Capacity and action. D1.2 and D4.2 claim $150 million capacity; D0 proposes a $20 million allocation. Current strategy assets and liquidity workpapers are absent, so available capacity is not established. Reconcile turnover to trades and test execution costs at total assets after the proposed allocation (Q3 and Q5).
Material means the issue could change interpretation of the supplied performance evidence. It is not a manager rating or investment recommendation.
PDF page 4 of 4 · SAMPLE-002 · Synthetic, not client work.
Questions and review coverage
Questions for the manager
1. Return basis. Please reconcile the deck to actual gross and net monthly returns, with each cost and fee component, deduction timing and applicable mandate terms.
2. Model selection. Please supply the complete trial log, selection criterion, design-freeze date and an evaluation period not used for selection or later tuning.
3. Turnover and costs. Which turnover is correct, 6 or 18 times NAV? Please reconcile it to trades and identify what the 10-basis-point assumption covers.
4. Timing and universe. Your questionnaire says the universe is "intended to include delisted names." Please provide historical membership and delisting records, revisions and signal-availability timestamps showing what was implemented.
5. Capacity. For our proposed $20 million allocation, please provide current strategy assets, other committed capital and trade-size, liquidity and participation workpapers supporting the $150 million capacity claim. Include market impact and borrow where relevant.
6. Live record. Please reconcile the six live months to independent statements and a consistent cost basis. Describe any model changes since December 2025, or confirm there were none.
Coverage of the six Desk Read checks
| Check | Result and boundary |
|---|---|
| Headline statistics | Recomputed from supplied rows; deck precision matches gross figures. Finding 01. |
| Significance and selection | Conditional Bonferroni p = 0.101; complete search history unverified. Finding 02. |
| Costs, turnover and capacity | Turnover gap quantified; $20m allocation headroom and $150m capacity unverified. Finding 03. |
| Backtest versus live | Six-month arithmetic checked; horizons and cost bases differ. Page 2. |
| Methodology risks | "Intended" inclusion of delistings describes intent, not implementation. Timing and survivorship risk unresolved; Q4 requests records. |
| Document consistency | Return labels, trial history and turnover cross-checked across D1 to D5. |
Evidence and limits
D0 to D5 are StatGazer-authored fictional inputs in input-materials.md and supplied-returns.json. The companion analysis.py and methodology.md explain each calculation. No benchmark series, manager strategy code, trade ledger or independently corroborated live record was reviewed.
No real manager was contacted and no factual-check response is implied. This sample demonstrates a documents-only technical review. It does not establish code correctness, operational due diligence, regulatory assurance or suitability of an investment.