What happens to an order placed during an economic release

A signal lands two minutes before the US employment figures are published. The message gives an entry price, a stop loss and two targets. The order goes out, it fills, and the ticket shown in the terminal bears little resemblance to what was written in the channel.

That gap is not a bug in the copier. It is how a market behaves during an economic release, and it happens with the same regularity at every broker. The only useful question is whether you want your orders going out at all under those conditions.

What changes in the book during a release

During a release, two things happen at once. The spread widens, so the distance between the bid and the ask grows. And slippage increases, so the order fills at a price further from the one requested.

The practical result is simple: the order fills at a price far from the signal's. Not by a few tenths of a pip, sometimes by much more. On a market entry you pay the widened spread when you open, then you pay it a second time when you close if you exit inside the same window.

This applies to the entry, but also to the exit. A stop loss touched during a release is not guaranteed to fill at its level. It becomes a market order the moment price crosses it, in a book with fewer counterparties than there were a minute earlier.

Why the problem is not the copier's speed

The instinctive reaction is to look for faster execution: a VPS closer to the server, lower latency, a shorter processing delay. Those things matter in the normal life of a copied account. They do not fix this particular problem.

No copier can fill at a price that no longer exists. When price moves during a release, it does not slide, it jumps. Between the level quoted in the message and the next price your broker offers, there is sometimes nothing at all. Gaining a few milliseconds does not change that, because the intermediate price was never available.

The consequence is arithmetic. If your entry is further away than planned and the stop loss stays at the level the signal gave, the distance between the two is no longer the one you used to size your lot. Your risk in money is not what you think it is. And your reward-to-risk has already degraded before the position has begun to live.

First safeguard: refuse trades around releases

Botty can refuse trades around economic releases. The signal is received, it is read, and it is not executed because the window is closed. No order goes out.

This is the most direct filter, and it is also the one that demands the most honesty from you. It will also refuse signals that would have worked well. Some channels publish precisely around releases because that is their strategy; if you follow a channel of that kind, this filter will make you miss most of its activity. You need to know that before enabling it, not after comparing your results with the channel's.

For a channel that trades normally through the rest of the day, the trade-off is more comfortable. You give up a few entries a month and you remove the worst fills of the month.

Second safeguard: close before the release

Refusing new orders says nothing about positions already open. A position taken three hours earlier, running and slightly in profit, will go through the release with its stop loss in place, in a book where that stop loss is precisely less reliable.

Botty can close open positions before a release. You come out of the market ahead of the window, and you go back in when a new signal arrives.

The price of that comfort is clear: you take whatever the result is at that moment. A position that would have finished on its second target is closed halfway. A position slightly in the red is closed at a loss, without waiting for the recovery it might have had. This setting swaps one tail of the distribution for another.

A position closed before a release does not reopen by itself. Botty does not resume the trade after the window: it will take a new signal from the channel. If your channel sends a single message per trade and then manages the position through updates, you lose the rest of its management on that particular trade.

Third safeguard: time windows

Botty can restrict trading to set time windows. It is a blunter filter than the previous one, and that is often the point.

Many releases come at fixed hours. US statistics land in the early European afternoon, central bank decisions in the mid-afternoon, Asian figures during the European night. A well-chosen time window covers a whole family of events with a single setting, without you having to reason event by event.

Time windows do something else as well. They exclude periods where the spread is structurally wider with no release involved: the first few minutes after the Sunday evening open, the end of the day at the value-date rollover, the thin hours when the book is empty. A signal received at those moments suffers the same kind of degradation, with no calendar to warn you.

Fourth safeguard: specific dates

Botty can also restrict trading to specific dates. Some days are better handled as a whole day than as a window.

• Public holidays in a major financial centre, when liquidity is thin for the whole session and not just at one hour.
• Monetary policy decision days, where the announcement is followed by a press conference that stretches the turbulence over several hours.
• Quarter ends and expiry days, where the flows have nothing to do with what your channel is reading.
• Days when you cannot watch the account, which is a perfectly valid reason not to trade.

This is the filter people forget to remove. A blocked date stays blocked. If you exclude a week of holiday, write it down somewhere, or you will spend two days wondering why signals are arriving and no orders are going out.

Combining them without smothering the channel

The four settings stack, and stacked without thought they end up letting nothing through. A reasonable order of setup is to start with the broadest and tighten only if the tickets justify it.

Start by watching. Over a few weeks, note the trades whose fill price departs clearly from the price given in the message. Look at what time they were taken. If the pattern is concentrated in a few slots, a time window is enough. If the pattern follows releases, turn on the refusal around releases. If the problem mainly concerns positions you hold for a long time, closing before the release is the right tool. If the problem is a whole day, use the dates.

A filter installed without having seen the problem it solves is a filter you will switch off in the first disappointing month, usually at the worst moment.

What these settings do not do

None of these four filters predicts anything. They do not know whether the figure will come in above or below consensus, nor which way the market will react. They only decide whether an order goes out inside a window you defined in advance.

They also do not recover slippage you have already taken. An order filled far from the signal price is filled; the filter exists so that the next one is not. And they do not replace position sizing: if your lot is calculated on a stop loss distance that is no longer the right one, the problem remains untouched outside releases too.

Finally, you have to accept the most uncomfortable consequence. As soon as you filter, your history stops matching the one the channel publishes. You will have missing trades, early exits, empty days. The provider will show a result you did not get, and you will not be able to know for certain whether your filter protected you or merely cost you an opportunity. That is the price of these settings, and it is a price better accepted when you enable them than at the end of the month.