Most indicator vendors won't tell you how their tools decide anything — because "proprietary" sounds better than honest. Here is exactly how ours work, and exactly what they don't do.
"No repaint" is the most abused phrase in the indicator market, so here is our exact definition, the one our code follows:
The honest cost of this honesty: confirmed signals arrive one candle "later" than repainting tools appear to. Their history is fiction; ours is reproducible.
A fair value gap is a three-candle event. Our engines apply one strict refinement: inside bars are not counted as candle 1, 2 or 3 of the pattern. An inside bar adds no new range information — counting it manufactures gaps that were never real imbalance. Fewer zones, cleaner zones.
A high or low is only confirmed as structure once its inducement — the minor liquidity resting in front of it — has been swept. This removes the false swing points that make typical fractal-pivot SMC tools whipsaw.
Order blocks are graded ★ to ★★★ from measurable inputs: volume, body percentage, inside buy/sell pressure and location in premium/discount. A grade is a quality description, not a prediction.
Our tools ship with changelogs and semantic versions — the flagship suite is on its sixth major generation, and fixes are documented release by release (feature releases as x.1, x.2; fix-only patches as x.0.1). When a detection rule changes, the changelog says so. When something was wrong, the changelog says that too — the ICT Pro v6.1 release notes include a first-touch counting bug we found and fixed ourselves.
If you find a plot that looks wrong, report it with symbol, timeframe and settings — repaint reports are treated as the highest-severity bug class we have.
If this level of honesty is what you want from your charts, you'll like what's on them.
