ArdenAlgoNo repaint, by design
"Does it repaint?" is the most asked and least answered question in the indicator world. The honest answer is that repainting is never an accident — it's the predictable result of three design shortcuts. We refused all three, and you can verify each refusal.
Shortcut 1: computing on the forming bar
Signals that appear intrabar look impressively early — and get silently revised when the bar closes differently. Everything in our engine (and in the free TradingView indicator that mirrors it) evaluates on confirmed bars only. A mark that prints has, by construction, nothing left to change it.
Shortcut 2: pivots that pretend to know the future
A swing high is only a swing high if the next k bars stay below it — which means it is unknowable until k bars later. Many tools draw the pivot at its historical location immediately, creating structure that "was always there" in hindsight. Ours confirms pivots with the full delay and accepts the lag as the cost of honesty.
Shortcut 3: irreproducible logic
If running the same data twice can give different answers — random seeds, session-dependent state, discretionary overrides — then every historical chart is unfalsifiable. Our engine is deterministic, and the automated test suite enforces it: same bars in, same events out, every time. That's also gate G1 of our public evaluation, passed with a bar-by-bar cross-check against TradingView data.
How to check us
Watch any signal land in the ledger, note its timestamp and levels, and look again a day later. Nothing will have moved. Repaint can survive marketing; it cannot survive an append-only database.