Closed trades by what produced them. Expectancy in R is the number to watch: positive and stable means the process has an edge after costs.
Real trades only. Instruments with negative expectancy are candidates to disable or re-playbook.
No closed trades in this window
[D / bad_entry] The bearish setup had a defined invalidation and attractive planned reward relative to the original stop. However, the manager had identified a confirmed M5 close below 1.11790 before the limit filled as a reason to treat the order as stale, then left it active through further fresh lows. The later fill therefore did not follow the stated entry condition. Price eventually hit the original stop; the stop was not moved, and the recorded loss was 1R plus costs. The loss alone does not invalidate the short thesis, but the stale-order handling was a process error.
For EURUSD London bearish-breakdown retest limits, cancel the resting order if an M5 candle closes below 1.11790 before entry, rather than keeping the order active for a later rebound.
[D / bad_entry] The original short had a defined retest entry and invalidation, but the limit was not filled until more than four hours later, with no recorded revalidation of the setup. It then stopped out immediately. The stop loss itself was not shown to be poorly placed, so the loss alone does not make the trade idea bad; the problem was allowing an old order to enter without confirming the thesis still held. Costs were about 10% of the stated risk, a meaningful drag but not the main issue.
Cancel an unfilled limit order when its setup reaches the stated staleness condition; if it remains unfilled for more than one hour, require a fresh setup check before leaving it active.
[D / bad_exit] The bearish pullback entry had a stated setup and an initial stop and target consistent with it. The exit management was not: the stop was moved from 1.11996 to 1.11946 at the same recorded time as the fill, with no recorded price evidence that the trade had first moved more than 1R in favor. The stop then closed the short just 0.2 pip below entry; that small gross gain did not cover the recorded costs, so the trade lost $0.60 net. The recorded claim that the trade had earned more than 1R conflicts with the timeline and should have been verified before tightening the stop.
Before moving a stop to breakeven, verify from recorded prices that the trade actually reached the required favorable-move threshold; for a short, do not make that move before price has moved at least 1R below entry.
When moving a stop on EURUSD to protect a short, set it far enough below entry to cover the trade's estimated spread, commission, and other costs; a 0.2-pip gross gain did not cover the recorded costs here.
[D / bad_exit] The entry had a defined breakout-retest rationale and a favorable planned reward-to-risk ratio, although it was counter to the higher-timeframe trend. The trade was manually closed at 4186.585, below the 4188.1 initial stop, for a 1.399R loss rather than the planned 1R risk. The record does not explain why the stop was exceeded; the main failure evident here is allowing the loss past the original risk limit.
On XAUUSD trades, do not manually hold beyond the original stop; if execution or slippage takes the loss past 1R, record the specific reason.
[B / good_trade] The short had a coherent trend-following basis: lower lows, falling M1 EMAs, and sustained M5 selling, with a stop above the pullback high. The target was reached for 2.594R from the fill. The entry was close to identified 4170 support while RSI was oversold, so continuation was not certain; the trade worked, but that outcome alone does not remove the risk of entering before support breaks. The fill was 0.99 worse than the intended entry, a meaningful cost relative to the stop distance, though the remaining reward was still substantial.
For XAUUSD shorts entered near a support level identified in the trade rationale, wait for a break below that level before entering.
[C / bad_entry] The bearish setup had a coherent multi-timeframe rationale and a defined invalidation level, so the stop-out alone does not show that the idea was bad. But the sell filled at 4162.52, 1.90 below the 4164.42 signal price—about 44% of the planned stop distance in adverse slippage. That materially worsened the trade's risk/reward: from the actual fill, the stop was 6.23 away and the target 11.82 away. The stop was then hit; no management changes are recorded.
Before accepting a market entry, recalculate risk/reward from the actual fill; reject or reprice if adverse slippage exceeds 0.25 of the original stop distance or leaves less than 2:1 reward-to-risk.
[D / bad_exit] The rejection setup had a clear invalidation level and a reasonable planned reward-to-risk ratio. However, the sell filled 3.49 points below the planned entry, materially worsening execution, and was manually closed 39 seconds later for only 0.034R, far short of the structural target with no recorded reason the thesis had failed. The small gross gain may not have covered trading costs.
For a market entry, skip the trade if adverse slippage exceeds 25% of the planned distance from entry to the initial stop.
[C / bad_exit] The short had a defined resistance-based entry, a stop beyond the swing high, and a planned reward-to-risk ratio of about 2.08R. It was manually closed for only 0.305R, well before the target, with no recorded invalidation or reason for exiting. The entry looks reasonably planned; the early exit limited the payoff. Transaction costs are not recorded, so the net result cannot be assessed.
For XAUUSD rejection trades with a predefined target, do not manually close a profitable position before the target unless the setup is invalidated or a management trigger was specified before entry.
[D / bad_entry] The bearish rejection rationale was plausible when the order was created, but the limit order remained active from October 2 until October 5. It filled and hit its stop at the same recorded time, so the original intraday setup was stale by execution. The loss alone does not show that the rejection idea was wrong; the avoidable issue was allowing an old order to enter without revalidating it.
Cancel intraday pending orders at the end of the trading session; before carrying one into a later session, revalidate the setup and replace the order only if the original conditions still hold.