GFWS identifies potentially important market situations. The user must still evaluate higher-timeframe structure, market phase, scheduled events, volatility and personal risk process. The mechanical benchmark below is research context, not the intended way every alert must be used.
GFWS Signal Track Record
Directional Reaction Methodology
GFWS is designed to identify potentially important market situations and directional reactions — not to run an automatic trading system. The platform analyzes market context; users decide if, when and how to act, including entry, exit, risk control and money management.
Signals are recorded continuously. Public audit snapshots are published monthly and archived without overwriting prior reports. The current methodology remains unchanged.
Next monthly snapshot: September 2026 monthly audit, after 4 October 2026.
Preserved exactly as first published. New results are added as separate monthly audit snapshots below.
All recorded LONG PREDICTION alert events included in the available GFWS history.
Repeated alerts from one continuing instrument and timeframe situation were merged into one setup.
269 of 290 fully evaluated setups traded above their recorded signal reference price within the assigned window.
90.24% of all recorded setup episodes have already moved above their signal reference price.
Latest published directional reaction audit
GFWS now publishes one immutable public audit snapshot after each calendar month closes. The measurement logic is unchanged: GFWS records whether a PREDICTION setup subsequently produced a directional reaction inside its timeframe-specific evaluation window.
Median maximum move above the recorded signal reference price: +1.64%. Pending windows are disclosed separately and are never counted in the closed-window rate.
| Timeframe | Setups | Closed-window rate | Pending |
|---|---|---|---|
| 15M | 52 | 90.38% 47 of 52 | 0 |
| 1H | 112 | 95.54% 107 of 112 | 0 |
| 4H | 41 | 87.88% 29 of 33 | 8 |
| 1D | 6 | Pending 0 of 0 | 6 |
| Market | Setups | Closed-window rate | Pending |
|---|---|---|---|
| Crypto | 109 | 93.14% 95 of 102 | 7 |
| Futures | 48 | 84.44% 38 of 45 | 3 |
| Stocks | 16 | 100.00% 15 of 15 | 1 |
| Forex | 12 | 100.00% 10 of 10 | 2 |
| Commodities | 9 | 100.00% 9 of 9 | 0 |
| Etf | 9 | 100.00% 8 of 8 | 1 |
| Indices | 8 | 100.00% 8 of 8 | 0 |
The original beta benchmark is retained unchanged. July 2026 is the transition audit. The calendar-month series begins with August 2026; each published monthly JSON remains available as an immutable download.
What this number means
A LONG setup counts as a directional reaction when the market trades even briefly above the reference price recorded when the PREDICTION was issued. The reaction must occur inside the evaluation window assigned to that signal timeframe.
Whether GFWS identified a market situation that subsequently produced movement in the predicted direction.
It is not an executed-trade win rate, portfolio return or guarantee of profitability.
- Fully evaluated setups
- 290
- Correct directional reactions
- 269
- Still-open evaluation windows
- 48
- 95% statistical interval
- 89.18%–95.22%
Did price react in the predicted direction?
“Correct reactions so far” includes setups whose evaluation window may still be open. The closed-window rate uses only setups whose full timeframe-specific window has elapsed.
| Timeframe | Evaluation window | Setups | Correct reactions so far | Closed-window success | Pending |
|---|---|---|---|---|---|
| 15M | 1 day | 63 | 95.24% 60 of 63 | 94.64% 53 of 56 | 7 |
| 1H | 3 days | 157 | 91.72% 144 of 157 | 91.50% 140 of 153 | 4 |
| 4H | 10 days | 99 | 87.88% 87 of 99 | 93.83% 76 of 81 | 18 |
| 1D | 90 days | 19 | 73.68% 14 of 19 | Not yet available | 19 |
Daily setups are included for transparency, but no 90-day evaluation window had fully elapsed at the audit cut-off. Their closed-window success rate is therefore intentionally left blank.
How the trading record is calculated
Reference price
Each setup begins at the first recorded PREDICTION timestamp and market price. Later duplicate alerts do not create additional wins.
Directional reaction
For a LONG signal, success means that a future market high traded strictly above the recorded reference price before the evaluation window expired.
Timeframe-specific windows
15M signals are evaluated for 1 day, 1H signals for 3 days, 4H signals for 10 days and 1D signals for 90 days.
No duplicate inflation
Timestamped alerts are consolidated into continuous setup episodes by instrument, timeframe and direction.
No look-ahead bias
Only market bars beginning after the recorded signal timestamp are used. Price movement before the alert is excluded.
Closed windows only
Each headline rate includes only setups whose complete evaluation window had elapsed. Pending windows are reported separately.
Important interpretation
This is a signal-reaction record, not an executed portfolio track record and not a trading win rate. GFWS identifies market situations and directional reactions. It does not decide user entries, exits, stop-losses, leverage, position size or money management, and it does not guarantee favorable outcomes or provide personal investment advice.