Technical Scanner
The technical scanner watches live price data and turns it into trade setups. What it looks for, where it enters, and how it manages the trade are all yours to configure — you build a strategy in the dashboard’s Technical tab, test it against history, and only then link it to an account.
This page explains how a strategy is put together. The field-by-field reference is linked from the ? next to every setting in the builder.
A strategy has four layers
A strategy is not one big form. It is a small stack of reusable pieces, so that two strategies that share (say) the same stop-loss rules can point at the same one rather than drifting apart.
| Layer | What it decides | Reusable? |
|---|---|---|
| Engine profile | Which timeframes to read, and how market context is built on them — swing highs and lows, structural breaks, supply/demand zones, ATR | Yes, across strategies |
| Entry profile | What pattern to look for inside that context, and what order to place when it appears | Yes |
| Strategy profile | Stop loss, take profit, break-even, and every filter that can reject a signal — sessions, daily caps, minimum R | Yes |
| Technical strategy | Ties one of each together and points it at a list of instruments | This is the thing you link to an account |
The three profile layers are shared objects. If four strategies use the same engine profile, editing it changes all four. The builder tells you how many strategies a profile is used by, and the compose form has a Fork & edit button that copies the profile under a new name and points only the current strategy at the copy. Use it whenever you want to change one strategy without touching the rest.
Timeframes have roles
An engine profile does not read a single timeframe. It reads a list of slots, and each slot has a role that says what that series is used for.
| Role | Required | What it does |
|---|---|---|
| Entry | Yes, exactly one | The series detection runs on. A signal fires on an entry-timeframe bar. |
| Trend | Optional | Sets the higher-timeframe trend direction that the HTF direction filter reads. |
| Structure | Optional | Where swings, structural breaks and zones are built. |
The default is a single entry slot — one timeframe doing everything, which is how the scanner has always worked. The other two roles let you split it up:
- Entry M5 + trend H1 — trade M5 setups but only in the direction of the H1 trend. Without a trend slot, the HTF direction filter falls back to a long lookback on the entry timeframe itself, which is not really a higher timeframe.
- Entry M15 + structure H1 + trend H4 — mark zones on H1, wait for a trade inside one on M15, and only take it with H4 behind you. This is the classic higher-timeframe-zone, lower-timeframe-entry split.
Two rules the builder enforces:
- A non-entry slot must sit at or above the entry timeframe. A “higher-timeframe trend” read from a lower timeframe is not a weaker signal, it is a wrong one.
- Every slot is a separate stream of candles from your terminal, and each needs roughly 120 bars before it produces anything. More slots means a longer warm-up after your EA reconnects. While a slot is still filling, the strategy reports warming up and emits nothing — it does not quietly fall back to a substitute.
The timeframe lives on the engine profile and nowhere else. The strategy form shows what the linked engine resolves to, read-only.
Building a strategy
- Pick or fork an engine profile. Start from one of the shipped presets — a watcher-zone M5 engine, an M1 scalp engine, an M15 swing engine, an H1 position engine. Fork it if you want different timeframes.
- Pick an entry profile. This is the biggest decision: the scan mode determines what pattern the scanner hunts for, and it also determines which of the other settings do anything at all. Fields that cannot reach the runtime for your chosen scan mode are greyed out and listed under “Set but ignored”.
- Pick a strategy profile for stops, targets and filters.
- Add your instruments. Leave the list empty to match every symbol your terminal streams.
- Read the validation panel. It reports schema errors, incompatible combinations, and realism warnings — settings that are legal but will behave differently live than they look on paper.
Testing before you trade it
Two tools, both under the Backtest tab:
- Backtest runs the strategy over a date range and reports trades, win rate, average R, drawdown and net R, with execution friction modelled — spread, slippage, commission and swap.
- Day Simulation replays a single day slice by slice with chart images, so you can see exactly what the scanner saw and why it did or did not fire.
Both read the strategy’s own configuration, including its timeframes. Neither lets you override the timeframe from the form: a result you cannot reproduce live is not a result. If a backtest looks good on a timeframe, that is the timeframe the strategy actually trades.
A backtest tells you what has worked on past data so that edge can be repeated on data you have not seen. It is not a tool for finding favourable historical windows — live trading happens on future data, where those windows are unknown in advance.
Going live
Link the strategy to an account from the account’s settings. From then on every closed bar your terminal streams is scanned against it, and setups that pass every filter appear in your dashboard for approval, or go straight to execution if the linked risk preset is set to auto.
Editing a live strategy applies immediately. The builder shows you how many accounts and in-flight setups an edit will touch before it saves.