How many signals a day should you take?
We publish roughly five signals a day. The question we get is whether you should take all of them, and the honest answer is that the count is the wrong thing to worry about.
Taking all of them is a real strategy
The expectancy we publish — 11.6 pips a trade across 1031 trades — assumes every signal is taken. It is the outcome of the whole set, losses included. Take the set and you get the set's result.
Skip the ones that feel uncomfortable and you are no longer running that strategy. You are running your own, using our entry prices, and its expectancy is unknown to both of us. That is not automatically worse, but nobody can tell you what it is.
Where filtering does have a case
One filter is supported by our own record rather than by instinct. Broken down by when a trade was opened:
- London — 15.4 pips per trade
- New York — 7.8 pips per trade
- Asia — 10.3 pips per trade
- Late session — 1.3 pips per trade, with 52.6% stopping out
The late session is the one clear outlier: more than half of those trades stop out and the average is close to nothing. If you are going to skip anything, skip by the clock rather than by how a chart looks to you at the time. The sample there is small, so treat it as a lean rather than a law.
The limit that actually bites
Not how many you take. How many are open at the same time.
We can have up to eight positions running at once. At the standard 1% risk per trade, that is 8% of the account exposed simultaneously — and currency pairs move together, so those are not eight independent bets. EURUSD and GBPUSD disagreeing with you at the same moment is ordinary, not unlucky.
Almost every guide tells you to risk 1% per trade. Almost none of them mention what happens when five of those are live at once.
What to do instead of counting signals
- Cap total open risk, not trade count. Decide the most you will have exposed at once — 4% is a common figure — and skip new signals while you are at it.
- Size for a full book. If eight open positions would frighten you, your per-trade size is too large, regardless of how it looks on one trade.
- Skip on time, not on feeling. The late-session numbers above are evidence. A chart that "looks wrong" at 9pm usually is not.
- Count how long you are exposed. Our median trade runs 3.0 hours, and losers resolve fastest at 1.6. Positions do not sit open for days.
Five signals a day is not a lot to execute. Five signals a day at careless size is how accounts end. The whole record is at teachtrades.com/results if you want to count the overlaps yourself.