Docs·Web App·Auto Trade

History — closed trades

The performance ledger. Every closed AUTO trade with its full lifecycle, plus aggregated win-rate per source, per symbol, per timeframe, and the cumulative equity curve. The single page where you decide which configs are working and which need adjusting.

The History tab is your performance review. Every closed AUTO trade lands here with its full lifecycle — entry orders, every TP that fired, the final close (TP hit, SL hit, manual close, detach), realized P&L, fees paid, and time-in-trade. Aggregated views on top let you slice by source, symbol, timeframe, or side so you can find out which configs are actually making money and which to dial back.

Trade table — what each column means

ColumnNotes
ClosedTimestamp of the final close.
SourceWhich signal feed fired the trade (currently Crypto USD-M Signal).
SymbolClick to load the chart with the trade's lines marked.
SideLONG or SHORT.
EntryAverage fill price across all entry orders (market + limit fills).
ExitAverage close price across all closing fills (TP fills + final close).
OutcomeOne of TP6 (ran to final target), TP3 + SL@TP2 (partial run, locked profit on remainder), SL (full stop), Detached (you took over), Manual close (you hit Close).
TPs firedComma list of which TPs filled — e.g. 1,2,3.
Realized P&L $Net of fees, in USDT.
Realized P&L %Percentage of margin used.
FeesTotal exchange fees paid across all fills.
Time in tradeDuration from entry fill to final close.
DetailOpens the per-trade timeline dialog (every order placed, every fill).

Aggregated cards

Above the table, summary cards let you read performance at a glance:

CardWhat it shows
Equity curveCumulative realized P&L over the date range. Click to zoom in on a window.
Win rate by sourcePer-source win rate + total P&L.
Win rate by symbolTop 20 traded symbols sorted by win rate.
Win rate by timeframePer-TF aggregation (15m vs. 1h vs. 4h, etc.).
Win rate by sideLONG vs. SHORT.
Best / worst tradeSingle biggest winner and biggest loser in the range.
Avg time-in-tradeHow long a trade typically lives — useful for sizing future capital exposure.

Read the equity curve before reading any single number. A 60% win rate with a falling equity curve means winners are smaller than losers; a 40% win rate with a rising curve means winners are big enough to pay for the losers. The curve is the truth.

Filters

The same filter bar you'd expect from any closed-trades surface:

FilterOptions
Date rangeToday / 7d / 30d / 90d / custom range.
SourceMulti-select; mute underperformers visually before muting them in config.
OutcomeTP-hit only / SL-hit only / breakeven only / detached / manual close.
SideLONG / SHORT.
SymbolSearch by ticker substring.
Win / lossShow winners only / losers only.

Filters compose — Last 30d + Source = Crypto USD-M + Outcome = SL shows every stop-loss in the last month, ordered by P&L impact.

Per-trade detail dialog

Clicking a trade opens a dialog with the full timeline. Every event is timestamped to the second:

EventDetails
Signal receivedSource, side, suggested entry / SL / TP, leverage, AI score.
Orders placedEach entry order — market or limit, price, size, working / filled status.
FillsEach entry fill — price, size, fee, exchange order ID.
TP fillsWhich TPs hit, when, what size at each.
Trailing eventsEach SL move logged with the new price and which TP triggered the move.
Final closeLast fill — TP6, SL, or manual — with price and size.
P&L breakdownGross P&L, total fees, net realized P&L, realized % of margin.

The exchange order IDs cross-reference to your Bottom panel → Order History and to Binance itself if you want to verify any single fill on the exchange side.

CSV export

The cog menu in the table header has an Export CSV action. It dumps every trade in the currently-filtered range with all columns plus the order IDs. Useful for:

  • Tax / accounting filings at year end.
  • Strategy review in a spreadsheet you control.
  • Sharing with a trading group or accountant.

The export respects the active filters — if you filter to last 7 days SL-only, the CSV contains only those rows.

How to actually use the History tab

The whole point is to inform config changes. A few high-leverage workflows:

Weekly review (every Sunday, 10 minutes)

  1. Filter to Last 7d.
  2. Sort the trade table by Realized P&L $.
  3. Look at the Win rate by source card. If a source is below 40% AND the cumulative P&L is negative — pause it on the Trade Config until you understand why.
  4. Look at the Win rate by side card. If LONG and SHORT are wildly different, the market is one-sided — adjust risk accordingly.
  5. Click into the 3 biggest losers. Was the SL hit by noise, or was the trade genuinely against the trend? If noise → consider widening SL via a custom SL setting; if trend → check whether you should have rejected those signals in Semi-Auto.

Monthly review (every month, 30 minutes)

  1. Filter to Last 30d.
  2. Read the equity curve. Trending up, sideways, or down?
  3. Win rate by symbol — which symbols make you money and which lose? Adjust the whitelist / blacklist in Trade Config.
  4. Time-in-trade distribution — are TP6 hits clustering at a specific TF? Maybe focus that TF in your config.

Per-trade post-mortem after a big loss

Click the trade. Read the timeline. Ask:

  • Did the entry fill at signal price or with significant slippage?
  • Did the SL get hit by a wick or a sustained move?
  • Were the TPs realistic given the chart at signal time?
  • Was the AI score high or low? Should you have rejected this in Semi-Auto?

A few of these post-mortems per week is how you actually learn what your config does.

What History does NOT show

  • Open positions — see Running Positions. History is closed trades only.
  • Skipped signals — see the Skipped filter on Signal Live Monitor. Skips are not "trades" since no order was placed.
  • Manual trades that never touched the engine — they appear in the chart-terminal trade history, not here. History is scoped to trades that went through the AUTO pipeline.

Common pitfalls

  • Reading a 1-day sample as truth. Sample size too small — you might have caught one trend day and called it a strategy. Minimum 50 trades or 14 days for source-level conclusions.
  • Comparing P&L across periods without normalising by volume. If you ran 2× the volume in week 2, raw P&L will look better — sort by win rate or by Realized % instead.
  • Muting a source after one bad week. One losing week doesn't kill a strategy. Three losing weeks in a row probably do — that's the bar for muting.
  • Forgetting that the engine's wins compound silently. It is easy to remember the big losses (they hurt). The equity curve is the only honest measure.

What's next