Coinrule is one rule. Most strategies are several, and they interact.

What Coinrule is

Coinrule is the most approachable thing in this category, and that is a deliberate and well-executed choice. You build automation as if-this-then-that rules from drag-and-drop menus: if BTC drops 5% in an hour, buy $200. There are 350+ templates, a free tier, a demo exchange to practise on, and — unusually — coverage beyond crypto into stocks and ETFs through broker integrations.

If you have a simple idea and want it running without learning a new mental model, nothing here beats it. That is a real product achievement, and most platforms are worse for pretending simplicity is easy.

Where it stops

The clarity comes from the constraint: one trigger, one action. That is what makes IFTTT readable, and it is also exactly what a real strategy stops being after about a week.

Rules run independently. They do not know about each other. So when you have five of them on the same account, nothing is coordinating them:

• →Two rules both fire on the same dip and you are twice the size you intended.
• →One rule buys while another is still unwinding the same position.
• →Your daily loss limit is a rule, so it can only act after another rule has already sent the order.
• →You want the third entry smaller than the first — but each rule fires identically every time, because it does not know it is the third.

You can sometimes patch around this with carefully worded conditions, and experienced Coinrule users do. But you are encoding coordination into rules that were designed not to coordinate, and the encoding lives in your head.

The other limit is that a rule is flat. There is no "and then". Ladder in, watch, switch behaviour, scale out in thirds, trail the remainder — that is a sequence, and a sequence of independent rules is not the same thing as a sequence.

What anvas does differently

The connections are the product. A graph is a set of steps with defined order and shared state, so the coordination that Coinrule leaves to you is something the strategy holds itself.

One position, tracked in one place, so two entry conditions cannot double-size it. One risk guard that sits between every order node and the exchange, so it refuses the order rather than reacting to it afterwards — that is one risk limit across everything. A branch, so "and then do something different" has somewhere to go. A counter, so "the third entry" is a thing the graph knows.

This costs you the thing Coinrule is best at — you have to think about structure, and a canvas is not as immediately readable as one sentence.

The actual question

If your strategy fits in one sentence, write it as one sentence — Coinrule is the right tool and this page is not trying to talk you out of it.

If you are running five rules that all quietly depend on each other, you have already outgrown the model. The dependencies did not disappear; they just moved into your head, where they cannot be tested.

Keep reading