Docs

For the moment something does not match what you expected

Most people never open documentation. These are written for the moment a number surprises you, so each one answers a single question and stops.

Answers

Four things people ask most, answered against what the code actually does.

Why did my live results differ from the test?

Costs, almost always. We apply spread, commission and a realistic delay on every trade, but your broker’s spreads widen during news in ways ours do not, and requotes and slippage are real. That is why every result is also run again at 1.5× those costs and both numbers are shown — the gap between them is how much of the result depended on getting cheap fills.

What counts as a trade in the trade count?

Every close, including a partial one. If a strategy closes half at 1R and the rest at 2R, that is two trades in the count, not one — each close is recorded separately with its own result. It is worth knowing before comparing the trade count against a platform that counts round trips instead.

Can I edit the code after downloading?

Yes. Anything you download is yours to inspect, change and deploy — the terms say so explicitly. It is also yours to be responsible for: we never connect to your broker and never place a trade on your behalf.

Why can I not download my strategy yet?

Downloads are not open during the pilot. Building strategies and testing them is free and unlimited in the meantime, and nothing you build is lost — the file becomes available to you when downloads open.

The guides

Each answers one question and stops. Read the first three in order if you have not used this before.

Start here

Write your first strategy4 minWhat to say, what to leave out, and how much detail is enough.Reading the review card3 minEvery rule we parsed, in the order it fires, before anything runs.Reading your results6 minNet, drawdown, trade count, and the stressed line next to each of them.

The strategy canvas

Nodes and what they mean5 minPrice reads, indicators, sessions, filters, zones and exits.Editing a rule after a run3 minChange one node, rerun, and compare the two side by side.The scrub bar4 minStep the canvas through the test to see when a zone became readable.

Running it for real

Downloading the file2 minWhat you get, and where this currently stands.Connecting your broker5 minWhere costs will differ from the test, and by how much.When live and test disagree7 minThe four usual reasons, in the order worth checking them.

Words we use precisely

These appear on the results screen and mean exactly this, every time.

Drawdown
The largest fall from a high point in your balance to the low that followed it, before a new high.
Repainting
When an indicator changes its past values after the fact. Nothing here repaints.
Slippage
The gap between the price you asked for and the price you got.
Stressed result
The same test with costs raised by half again, to show how much of the result survives.
Spread
The distance between the bid and the ask. It is a cost you pay on every trade.
Out of sample
A stretch of history the strategy was not tuned on.

Still stuck? Write to help@algolo.app and a person answers.

See the data sources