Eligible record
Closed status, a recorded entry date, realized P&L, and a qualifying source such as supported CSV import, broker workflow, SnapTrade, or TradingView ingestion.
Published methodology / reviewed August 28, 2026
How imported trades become an evidence-backed finding.
TradeInsights separates deterministic historical measurement from interpretation. This page documents the qualifying data, calculations, finding selection, coverage labels, and limitations behind the First Value Report.
Direct answer
After at least 20 qualifying imported, closed trades are available, TradeInsights calculates reproducible historical metrics, runs documented comparison screens, and selects only a flagged result as the primary evidence-backed finding. The report exposes the method, sample coverage, warnings, and limitations. If no screen flags, the diagnostic does not invent a finding.
1 / Data qualification
The evidence threshold uses imported records, not sample activity.
Excluded from activation
Manual exploration and sample records do not unlock the evidence threshold. Missing fields can also exclude a record from a specific calculation.
Reconciliation required
Provider availability, field coverage, history depth, fees, timestamps, and refresh timing vary. Compare imported totals with the source statement before interpreting a result.
2 / Deterministic measurements
The same eligible input produces the same calculations.
The report calculates historical net P&L, gross profit and loss, win rate, average winner and loser, payoff ratio, profit factor, expectancy, streaks, and drawdown. It also segments available records by symbol, weekday, direction, and duration. Calculation versions are returned with the report so changes can be identified.
3 / Current finding screens
Each screen states what it compares—and what it cannot establish.
High-volume-day comparison
Compares recorded win rate on days with at least twice the sample's average trade count against other recorded days. Day boundaries are UTC, and the comparison does not establish that trade count caused an outcome.
Entry-time comparison
Compares recorded entry-hour groups with the full sample and checks the first and last UTC activity windows. Local-session interpretation requires appropriately localized timestamps.
Post-loss size screen
When position-size fields are available, checks whether a trade immediately after a loss was at least 1.5 times the eligible sample's average recorded size. It does not infer emotion, intent, or causation.
Winner-versus-loser hold time
When enough exit timestamps exist, compares median recorded hold duration for winners and losers. Duration alone cannot explain why an exit occurred.
4 / Finding selection
A negative screen is not presented as evidence.
Every available screen can explain what it checked. Only flagged screens are eligible to become the primary finding. When more than one flags, the report prioritizes sample-size coverage and then the magnitude of any supported historical impact field. A completed diagnostic without a flagged screen remains useful, but it does not satisfy the evidence-backed First Value milestone.
5 / Coverage labels
Coverage describes sample size—not certainty.
- Low
- Fewer than 30 eligible records
- Medium
- 30–99 eligible records
- High
- 100 or more eligible records
These labels are not statistical confidence intervals, repeatability probabilities, causal estimates, or predictions.
6 / Interpretation limits
Historical association is not diagnosis, causation, or advice.
- A result describes only the imported sample and fields available to the calculation.
- P&L and trade sequences do not reveal emotion, discipline, motivation, or intent on their own.
- Changes in strategy, regime, instrument mix, costs, and data quality can affect comparisons.
- TradeInsights does not predict markets, place trades, guarantee outcomes, or replace independent judgment.
Questions / answers
Methodology questions, answered directly.
How many trades does TradeInsights need for the First Value Report?
The evidence threshold is 20 qualifying imported, closed trades. A qualifying record must include an entry date and realized P&L and arrive through a supported imported source. Manual and sample records can help explore the product, but they do not satisfy this threshold.
Does a high coverage label mean a finding will repeat?
No. Low, medium, and high are sample-size coverage labels, not confidence intervals, probabilities, causal estimates, or predictions. Current coverage bands are fewer than 30, 30 to 99, and at least 100 eligible trades.
Does TradeInsights diagnose trading psychology?
No. It can compare recorded sequences, timing, sizing, outcomes, and user-authored context, but trade outcomes alone do not establish emotion, discipline, intent, or a psychological diagnosis.
What happens if no screen finds a supported pattern?
The report can still show deterministic performance measurements and explain what it checked. A negative screen is useful diagnostic context, but TradeInsights does not relabel it as an evidence-backed finding.
Inspect the method against your own history.
New accounts can store up to 30 real trades. Import 20 qualifying closed trades to run the deterministic First Value Report.