How we measure

The same rules for every channel, every day. Nothing is picked by hand.

1. Which channels

Public Telegram channels that publish gold (XAUUSD) signals. New channels are found by a search that runs every day and start on trial like everyone else. Channels that stop posting, post things we cannot read, or keep losing are dropped: 1,929 so far. No channel pays to be listed, ranked or removed.

2. Reading a signal

Our reader follows the channels around the clock and reads each new message within seconds. A message is a signal when it has a direction (buy or sell), an entry price or zone, a stop loss and at least one target. Corrections to the stop published right after the signal are applied only when they tighten it.

3. Checks before opening

• Price already gone: if the market has moved too far from the entry (more than the larger of $4 and half the distance to the stop), the signal is refused.
• Late signals: more than 30 seconds after the message, the signal opens only if the price is still within $2 of the entry.
• Wrong stop: a stop on the wrong side of the price is refused.
• Repeats: the same signal posted again by the same channel within 10 minutes is not opened twice.
• Targets already passed: a target that the price has already reached is not opened.

Refused signals are listed in the dashboard with the reason.

4. Execution

Each signal becomes one order per target on a MetaTrader 5 demo account at a broker, each with the channel’s own stop loss. The broker closes each order at its target or at the stop: we do not move stops or close orders by hand. Median time from the Telegram message to the order at the broker: a few seconds. Results include the broker’s spread and execution as they happen.

5. Results

• Size: results are shown for 0.1 lot per order, in euro. They are measured at 0.01 lot and multiplied by 10: results in money grow in proportion to size, but real execution at a larger size can differ.
• Trading day: 18:00 to 17:00 New York time. A signal belongs to the day it was opened.
• All-time net: the sum of every closed order of the channel since we started following it, wins and losses.
• Return R: euro earned for every euro put at risk (distance to the stop times the number of orders). It compares channels that risk a lot with channels that risk little.
• Proof: how strong the evidence is that a channel really earns (return per signal divided by its standard error). Above 3.5 with at least 60 signals; below that, a good run can still be luck.
• Reliability: solid, uncertain or losing, judged on worst-case results rather than averages.

6. Elite roster

At most 10 channels that pass every gate: positive net result, at least 8 signals over at least 8 trading days with at least 2 losses, worst signal and average loss within limits, at least +€50 per trading day, a rising trend over the last 10 trading days and at least one signal in the last 3 market days. A newcomer must pass the gates at two closes in a row; with the roster full it must beat the weakest member by 30%. Channels join and leave every day on their results.

7. The free trial

The trial shows the same dashboard 3 hours behind live, with channel names hidden and without the figures from which a channel’s total could be rebuilt. The Elite roster, session ranking, new candidates and rejected signals are for subscribers.