forex-scalping.com

An indicator that repaints changes its past values when new data arrives. The signals look perfect on the chart because they've been adjusted after the fact.

Typical examples include zigzag indicators, which redraw the last leg as price moves, and some pivot or fractal tools. Any indicator that uses future bars to confirm earlier ones has this property. Even moving averages can mislead if the last bar is still forming.

The problem for scalpers is that the chart in hindsight shows entries at perfect highs and lows. In real time, those signals didn't exist when the price was there. They appeared a few bars later.

A simple test: leave the chart running, take a screenshot, and compare it to the same chart an hour later. If signals have moved or disappeared, the indicator repaints.

Another check is to look at the indicator's behavior on the live bar. If the value changes while the bar is forming and settles only at the close, trades based on it need to be tested using that logic, not the final value.

Selling services often show pictures of repainting indicators, and the screenshots look convincing. The only honest evidence is a recorded video of live trading or a forward test.

Some indicators repaint only on the live bar, which is a milder form. A signal appears mid-bar and may disappear before the close. These need testing with the exact rule that will be traded: act mid-bar, with the risk of vanishing signals, or wait for the close, which loses some speed.

Has an indicator ever turned out to repaint, and how was it discovered?

I got caught by one of these early on, a fractal arrow indicator I'd found on a forum that showed what looked like eighty percent winners on the five minute chart. I traded it on GBPUSD for about two weeks and couldn't understand why my live results looked nothing like the history. The arrows were landing on the exact swing highs and lows, but only two bars after the fact.

What finally gave it away was boring. I left the platform running over a London lunch, took a screenshot at noon, and another at two. Three of the arrows from the morning had quietly shifted one bar to the right, and one had vanished altogether. After that I opened the code and saw it was looking two bars forward to confirm each point.

These days I won't trade anything I haven't watched live on a demo for a few sessions with screenshots. If a signal can appear and then disappear mid-bar, I test it strictly on bar close, even though that costs me a pip or so of entry.