شفافیت و راستی‌آزمایی

A trade enters the public record when it has closed (closed_at IS NOT NULL), has a recorded realised ROI (pnl_percent_on_close IS NOT NULL AND <> 0), and was not closed by a risk-engine housekeeping action. Nothing else is filtered — every loss is included. A WIN IS THE SAME TEST ON EVERY BASIS — a trade whose ROI on that basis is strictly greater than zero — and every metric says which basis it used. What differs between the bases is the ROI, not the verdict: basis=ladder (the default) measures a trade at the highest take-profit rung its price reached, on the full position, while recorded and blended measure it at its final exit. How often price reached a rung is published separately, as success_rate_pct and reach_pct, and is a different and smaller number than the win rate. The universe itself — which rows are graded at all — is identical on all three bases, so the three answers are three readings of one set of trades and not three different samples.

positions.telegram_message_id IS NOT NULL — the signal message id was stored before the outcome

At least one follower account's entry order for this signal was accepted by the exchange API (trade_execution_logs stage=entry, status=ok). Proves the order was sent; records no fill.

An exchange-assigned order id is stored in exchange_executions (publishable rows). positions.order_id is NULL on every row and copy-trade logs keep only our client_order_id.

An exchange-reported entry fill (quantity and average price) in exchange_executions.

An exchange-reported exit fill (quantity and average price) in exchange_executions.

Net realised PnL computable from exchange data: gross (reported or fill-derived) - fees + funding.

A recorded and/or ladder ROI computed from price levels. Every gradeable trade has one.

positions.pnl_percent_on_close against the ROI the same monitor wrote on the closing position_events row. Two tables, one exit: they should agree exactly.

positions.pnl_percent_on_close against ROI recomputed from entry_price and the closing event's price_at_event: (exit-entry)/entry*leverage*100, sign-flipped for SHORT. Differs by design on stop exits, where the stored ROI models the fill at sl +/- 0.10 %.

positions.pnl_percent_on_close against ROI recomputed from signal_quality_metrics.exit_price, the market price seen when the exit was detected.

positions.realized_pnl_pct against the ladder rule re-applied independently (SPEC section 2): max(ROI of the highest rung reached, ROI of the recorded exit), where the rung ROI is the TP_HIT event's announced ROI, or ROI(entry, published TP price) when no event carries one.

positions.laddered_pnl_pct against sum(frac * roi) over the same pnl_legs.

Every TP_HIT event's stored ROI against ROI recomputed from entry_price and that rung's published price. The largest absolute difference across the trade's rungs is reported.

Exchange-verified realised ROI (net of fees and funding, on entry margin) against the modelled recorded ROI. Only computable for a verified trade.

Fees and funding as reported by the exchange. There is no modelled counterpart: the model excludes both, so this check reports availability, not a delta.