Loading...
Loading...
Every signal we publish is locked at the moment it fires and then scored against real market prices. Wins, losses and expiries all count. When a target and a stop fall inside the same candle we record it as a loss, because the candle cannot tell us which came first — a rule that can only make these numbers worse, never better.
Read the full methodology for how signals are generated and scored, or take the raw data — it's free to use and republish with attribution.
This page counts live signals only. For how the engine performs on two years of historical prices with trading costs charged, see the backtest — those are simulated results, not live performance, and are kept separate on purpose.
On 2026-10-22 we moved every signal's target from twice the stop distance (2:1) to the same distance as the stop (1:1). At 2:1 the engine lost about two trades for every one it won. The record below is a rolling 90 days, so for now it mostly measures the old rule. These three figures show the new one. Each is labelled for what it is, and none of them is mixed into the live record.
1 – 0
1 won · 0 lost · 20 running. Win rate shown from 20 finished.
Published at 1:1 and scored against real prices, counted from day one. This is the figure that matters, and it grows every day.
53% wins
63 won · 57 lost (as published: 34 won · 64 lost)
Every signal in the live record re-scored against the real prices that followed, with only the target moved. Average +0.5% per signal, against +1.4% as published: more wins, but smaller ones. These trades were published with 2:1 targets, so this is not a result anyone received. Computed 2026-10-22.
53% wins
863 simulated trades · +0.6% per trade after costs
The engine replayed over 107 assets, scored only on the half of the history it was never tuned on. Simulated, not live. Full backtest.
Every signal is tracked to its target or stop on live market data — no cherry-picking. What matters isn't winning every trade, it's that the wins are bigger than the losses. Here's the last 90 days.
Our signals hit their target 34% of the time — and that's still profitable, because a winner averages +21.3% while a loser is cut at just -9.6% — about 2.2× more reward than risk.
35 wins · 68 stopped out · profit factor 1.10
Reward:risk
2.2 : 1
avg win vs avg loss
Avg. winner
+21.3%
when target hits
Avg. loser
-9.6%
cut at the stop
Net edge / signal
+0.8%
over 138 signals
Our signals hit their target 34% of the time — and that's still profitable, because a winner averages +21.3% while a loser is cut at just -9.6% — about 2.2× more reward than risk.
35 wins · 68 stopped out · profit factor 1.10
What we send vs everything we publish
Sent by alert and email
50% hit target · +4.5% per signal
Published and scored, never sent
24% hit target · -1.8% per signal
How each market is doing
Based on 138 resolved signals over the last 90 days, scored automatically against real market prices. Past performance doesn't guarantee future results. Not financial advice; trading carries risk.
Every asset with resolved history, strongest first — including the ones currently losing money. Each carries its own ledger: what was published, what happened, and how far the trade ran either way before it resolved.
Percentage shown is expectancy — the mean signed return per resolved signal for that asset.
Past performance does not predict future results. These signals are not financial advice and are not tailored to your circumstances. Trading carries risk, including loss of capital.