A repainting indicator changes what it showed you in the past. That single behaviour is enough to make every backtest, every screenshot and every performance claim about it worthless, because the history you are evaluating is not the history anyone could have traded.
An indicator repaints when the values it displays for past bars are different from the values it displayed at the time those bars were forming or closing. You look at a chart today and see a buy arrow at a perfect low. That arrow may not have existed when that bar was live. It may have appeared two bars later, once the low was confirmed, and been drawn backwards onto the chart.
There are several distinct mechanisms and they are worth separating, because they are not equally bad.
A backtest is an attempt to answer one question: if I had followed these rules, what would have happened. That question only has meaning if the information the rules used was available at the time of each decision. Repainting breaks exactly that condition.
Consider an arrow that only appears once price has moved thirty pips in its favour. On a historical chart it sits at the turning point and looks like a superb entry. In reality, at the moment the bar closed, there was no arrow. By the time it appeared, thirty pips of the move were gone, and the trader would have been entering at a much worse price with a much worse stop. The backtest reports the good entry. The live account gets the bad one. The difference is not slippage or execution, it is that the backtest was answering a different question.
The effect on performance figures is not marginal. A repainting entry can turn a system with no edge whatsoever into one that appears to win seventy or eighty percent of the time, because it is effectively being told the answer before it commits. This is why an indicator with an implausibly good historical chart should increase your suspicion rather than your interest, and why so many people describe the same experience: perfect in the tester, useless live.
The related trap is optimisation on repainted history. If you tune parameters against a series that contains future information, you are not fitting to the market, you are fitting to the leak.
You do not have to take anyone's word for it, including ours. There are four practical checks and they take very little time.
A useful rule of thumb: if an indicator marks tops and bottoms with striking accuracy on a historical chart, assume retroactive plotting until you have proved otherwise with the visual tester. Our longer guide on whether MT5 indicators repaint works through the tests in more detail.
MSP locks its state on the closed bar. It reads structure, trend, momentum, levels, volatility, volume and session in one pass, resolves them into a single verdict (TRADE, TRANSITION or NO TRADE) with a confidence percentage and an A, B or C grade, and once the bar closes, that result is fixed. It does not go back and improve it when later bars reveal what happened next.
The plain-English reason is part of that locked state as well, which is more useful than it sounds. When you review a trade from three weeks ago, you can see not only what the verdict was but what it was based on, exactly as it was presented at the time. That is what makes a review process worth doing.
Because there is one verdict rather than a set of separate signal lines, verification is straightforward. There is a single output to check, and the four tests above apply to it directly. Run them during the free trial. An indicator that asks you to trust its history without letting you verify it is asking for more than it should.
The practical consequence for trading is confidence under pressure. If you act on a verdict and it can never be un-shown, then a subsequent loss is a loss, not a reason to distrust the tool. That distinction matters a great deal for whether you can follow a process consistently.
It does not mean accurate. A non-repainting indicator can be consistently wrong; all it guarantees is that it is wrong in the same way live as in history, which is precisely what makes evaluation possible. Non-repainting is a prerequisite for trustworthy testing, not a claim about performance.
It does not mean instant. Locking on the closed bar means waiting for the bar to close, and on an H4 chart that is four hours. That is a genuine trade-off: you get reliability and you give up immediacy. Anyone offering both is offering the first one falsely.
It also does not mean the indicator will never change its mind. A verdict on the next bar can differ from the verdict on this one, because conditions changed. That is not repainting. Repainting is when the verdict on this already-closed bar is different tomorrow.
Repainting is the difference between a chart that shows you what happened and a chart that shows you what would have been nice. If an indicator's historical signals were not visible when those bars closed, then nothing you conclude from its history applies to your account.
Market Structure Pro locks its verdict, confidence, grade and reasoning on the closed bar, so the history on your chart is the history that was displayed. Test that yourself: watch a live bar, screenshot and compare, step through the visual tester, and reload the indicator.
It remains decision support for MetaTrader 5. It does not place trades, is not an EA or a signal service, and guarantees nothing. Free 7-day trial with no card required, and a money-back guarantee on paid plans, in the pricing section.
Market Structure Pro reads structure, trend, momentum, levels, volatility, volume and session in one pass and returns a single verdict with the reasoning attached. Free 7-day trial, no card required.
Start free trialIt means the indicator's displayed values for past bars are different from what it displayed at the time. A signal you see on a historical chart may have appeared several bars later, once the outcome was already known, and been drawn backwards. It can also mean a signal that appears on the current unfinished bar and vanishes when that bar closes.
Because a backtest is only meaningful if every decision used information that existed at that moment. A repainted signal effectively knows the outcome before committing, so the test measures a strategy nobody could have traded. A system with no real edge can show a seventy or eighty percent win rate this way, which is why implausibly good historical charts should raise suspicion rather than interest.
Four checks. Watch a forming bar for twenty minutes and see whether signals appear and disappear. Screenshot the chart and compare the same region hours later. Run the strategy tester in visual mode and step through bar by bar, watching for markers that appear on earlier bars. Remove and reapply the indicator, or switch timeframe and back, and see whether the history changes.
No. Its state locks on the closed bar, so the verdict, confidence, grade and written reason for a completed bar are fixed and do not change afterwards. You can verify this yourself during the free 7-day trial using the tests above, and we would encourage you to, for any indicator you rely on.
Yes, and it is open about it. ZigZag identifies swing points only after they are confirmed, then draws them on the historical bar where they occurred, so its most recent leg regularly moves. That is fine for what it is, a visual aid for identifying structure in history. It becomes dangerous when someone builds entry rules on it and backtests them.
No. It means the indicator behaves the same way live as it does in history, which is what makes honest evaluation possible. A non-repainting tool can still be wrong frequently. Non-repainting is a prerequisite for trusting a test, not a claim about results, and no indicator predicts the future.
Some do it innocently, because their calculation genuinely requires later bars, as with centred moving averages or swing detection. Some compute on the unfinished bar and were never intended to be acted on intrabar. And some do it because a repainted chart sells extremely well, since it looks like the tool catches every turn perfectly.
Yes, and that is the trade-off you are accepting. Reliability costs immediacy. On an M5 chart the wait is five minutes; on H4 it is four hours. Many traders find this improves their results anyway, because acting on unfinished bars is a significant source of impulsive entries.
