Build your first strategy

What a condition is

Every condition is the same four parts: an indicator, the timeframe it reads, a comparison, and a value. The builder prints each one as a sentence, and that sentence is exactly what the scanner evaluates.

RSI(14) on 1h  ·  less than  ·  30

Conditions combine as and: the strategy is satisfied on a market when every required condition is true at the same time. There are no nested groups to arrange. If you want "this or that", that is two strategies.

Step by step

1. Name it and pick a direction. Long or short. The direction decides how the entry check and the backtest read a move, and Refine keeps it for every variation it tries, so it is worth getting right at the start rather than editing later.
2. Add a condition. Choose an indicator from the catalogue, set its timeframe, pick a comparison from the ones that indicator supports, and set the value. The builder draws the indicator on a chart of a market of your choice as you edit, so you can see what the line you are comparing against actually looks like.
3. Add the rest. Most strategies that people keep have two or three conditions, and they usually mix a slower timeframe with a faster one: a trend condition on the 4-hour, say, and the entry condition on the 15-minute. Each condition can read a different timeframe.
4. Count the matches. Before you save, the builder can replay the draft against a market's history and tell you roughly how many times it would have matched over the recent window. It is an estimate of frequency, not a measure of quality — but it is the difference between a strategy that fires a few times a week and one that fires twice a year, and that is something to know before the first silent week makes you think the app is broken.
5. Save, then apply. A saved strategy does nothing on its own. The next step is to apply it to a watchlist, which is covered in the next guide.

Required and optional conditions

By default every condition is required, and the strategy only counts as satisfied when all of them are true together. On the Elite plan a condition can be marked optional and given a weight. An optional condition adds to the match percentage you see on the ranking screen, but it is never needed for an alert — only the required ones decide that. Use it for the thing you would like to see but would not refuse an alert over.

Describing it instead of building it

If you know the setup but not which indicator expresses it, the AI strategy builder takes a plain-English description and drafts the conditions for you. What comes back is an ordinary strategy: every condition is visible, editable and testable, and you can change any part of it before saving. The AI drafts the mechanics. It does not judge whether the idea is a good one.

Three habits that save time later

1. Fewer conditions than you think. Every condition you add makes the strategy fire less often. Five conditions on one timeframe can easily never line up at all, and the preview will tell you so.
2. Match the timeframe to how often you want to hear from it. A condition on the 5-minute chart can transition several times an hour. A condition on the daily transitions a few times a month. The fastest timeframe in the strategy sets its pace.
3. Preview on the market you will actually run it on. The same conditions fire at very different rates on a currency pair and on Bitcoin, because the indicators behind them move at different speeds there.