Gold Signal Performance — The Record So Far
The published result of every completed XAUUSD simulated signal, including the measurements that do not flatter the system.
Completed
10
1 calibration order excluded
Net result
+4.41R
average 0.44R per signal
Win rate
60.0%
profit factor 3.013
Max drawdown
-1.15R
longest losing streak 3
Profit retention
54.3%
of peak profit kept
Profit given back
5.06R
peak minus realised
Simulated-account statistics - not proof of future profitability. This is a young forward record on a simulated account and the sample is small. A strategy family with two closed trades tells you essentially nothing, and is shown anyway rather than hidden until the numbers improve.
Equity and daily results
Closed trades only — realised · UTC calendar day, dated by when a trade closedNet P/L
+$130.05
realised
Net R
+3.00R
average 3.00R per trade
Trades
1
0 opened in period
Wins
1
closed in profit
Losses
0
closed at a loss
Win rate
100.0%
Equity 20291 → 20448 (+156.65) 67 samples
Daily performance
| Date | Trades | Wins | Losses | Win rate | Net P/L | Net R |
|---|---|---|---|---|---|---|
| 2026-09-02 | 1 | 1 | 0 | 100% | +$130.05 | +3.00R |
| 2026-09-01 | 9 | 5 | 4 | 56% | +$120.75 | +1.41R |
By strategy family
| Group | N | Net R | Average | Win | PF | Retention |
|---|---|---|---|---|---|---|
| bko | 3 | -0.87R | -0.29R | 33% | 0.169 | -0% |
| cont | 2 | +1.59R | +0.80R | 100% | Not recorded | 130% |
| rvs | 2 | +0.31R | +0.16R | 50% | 1.314 | 161% |
| swg | 1 | +0.52R | +0.52R | 100% | Not recorded | 17% |
| tpb | 2 | +2.86R | +1.43R | 50% | 20.732 | 61% |
By direction
| Group | N | Net R | Average | Win | PF | Retention |
|---|---|---|---|---|---|---|
| BUY | 2 | -2.04R | -1.02R | 0% | 0 | Not recorded |
| SELL | 8 | +6.45R | +0.81R | 75% | 43.769 | 54% |
By session
| Group | N | Net R | Average | Win | PF | Retention |
|---|---|---|---|---|---|---|
| london | 5 | +1.25R | +0.25R | 80% | 2.198 | 45% |
| london ny overlap | 4 | +0.16R | +0.04R | 25% | 1.142 | 29% |
| new york | 1 | +3.00R | +3.00R | 100% | Not recorded | 100% |
Trend versus reversal
| Group | N | Net R | Average | Win | PF | Retention |
|---|---|---|---|---|---|---|
| breakout | 3 | -0.87R | -0.29R | 33% | 0.169 | -0% |
| reversal | 2 | +0.31R | +0.16R | 50% | 1.314 | 161% |
| trend | 5 | +4.96R | +0.99R | 80% | 35.312 | 54% |
By layer depth
L1 is a first entry; deeper layers are additions to an existing thesis| Group | N | Net R | Average | Win | PF | Retention |
|---|---|---|---|---|---|---|
| L1 | 10 | +4.41R | +0.44R | 60% | 3.013 | 54% |
What these numbers do and do not show
Net R is after measured costs of about 0.19R per round trip. Retention is realised result divided by the best the trade showed, aggregated over trades that actually went into profit — not an average of per-trade ratios, which would be meaningless across trades that never won.
What this record cannot show is whether the edge persists. A forward simulated record of this length is a check that the system does what it was built to do, not evidence about future returns. The research behind it is deliberately kept separate from a sealed holdout that has never been examined.
Every individual signal and its decision record →