Non-Repainting MT4 Indicator: Free Repaint Test

Why does a perfect MT4 signal disappear after you enter?

You have probably seen this happen. A clean arrow prints under a candle. You click buy. You look away for a minute, and when you look back the arrow is gone, or it sits two bars to the left of where it was. Your entry now looks like a mistake, and you cannot say why.

That moment is the core problem with repainting indicators. The signal you acted on was not the same signal the indicator showed later. The chart changed under your feet. Your decision was based on something that no longer exists on the screen.

Many traders blame themselves. They think the market moved too fast, or their timing was off. Sometimes that is true. But when the indicator itself rewrites its own history, no amount of timing skill fixes the problem. You are trading a moving target and calling it a signal.

This matters more than most features. A fancy dashboard, a long list of inputs or a colourful panel is worth nothing if the arrows cannot be trusted at the exact moment you need them. For beginner and intermediate traders on MT4 and MT5, the first question should always be simple: does this signal stay where it was shown?

The problem also hides well. A repainting indicator rarely looks broken on a closed chart. It looks neat, balanced and confident. The damage shows up only when you watch the signal form in real time and then compare what you saw with what is left behind. Most traders never make that comparison, so the problem keeps repeating.

What happened to one trader after a switch from demo to live?

Here is an illustration, not a real person or a real result. Call the trader Sam.

Sam finds a free MT4 indicator with buy and sell arrows. The demo chart looks excellent. The arrows line up with every obvious swing from the last few weeks. Sam is excited, moves to a small live account and starts trading the arrows the same day.

On the first morning, a buy arrow appears while the current candle is still forming. Sam enters at once, because waiting feels like missing the move. Ten minutes later the candle closes, and the arrow has shifted to a different level. The entry is now in a poor position. Sam closes the trade at a loss and tells himself the next signal will be better.

The next signal does the same thing. Sam starts to doubt every arrow. He hesitates, enters late, or skips trades that would have worked. Then, while scrolling back through the chart, he notices that some of the older arrows no longer sit where he remembered them. The backtest that looked so strong no longer matches what he saw live.

Sam did not need a better market or a faster finger. He needed a way to check the indicator before he risked money on it. He also needed to know what the check could and could not prove. That is the purpose of the Indicator Repaint Test Checklist covered in this guide.

Notice what went wrong in Sam’s story. The demo chart gave a false sense of confidence because nobody watched the signals form live. The live account then exposed the problem at the worst possible moment, with real money and an emotional reaction on top. A short, structured test would have caught the issue before the first live trade.

What does repainting actually mean?

Repainting means an indicator changes its values or signal positions on bars that have already printed. The earlier arrow, line or dot is redrawn after new price data arrives. The chart looks different from how it looked when the decision had to be made.

The simplest way to say it: a non-repainting indicator decides on a bar and keeps that decision. A repainting indicator can decide, then decide again after the next candles arrive.

There are two common ways this happens, described here in plain terms. You do not need to read any code to see the results.

The first is when an indicator uses the current, unfinished candle in its calculation. Price on a forming candle keeps moving, so the value changes tick by tick. If the indicator then displays a final value on earlier bars, the history looks clean while the live signal was never stable.

The second is when an indicator uses data from later bars to place a past signal. It looks accurate because it already knows what happened next. That kind of signal is useful for drawing a chart after the fact, but it cannot be traded, because the trader never had the later bars at the time of the decision.

Some vendors call repainting by softer names, such as “recalculation” or “optimisation”. The label does not change the effect. If a signal moved, changed size or vanished after its candle closed, the indicator repainted, whatever the sales page calls it.

It is also worth knowing that a repainting indicator is not always built to deceive. Some are written carelessly. Some were designed for a different purpose, such as drawing levels after the fact. Your job is not to guess the intent. Your job is to test the behaviour you can see.

Why does repainting ruin your backtests?

Backtesting is supposed to show how a system would have behaved in the past. A repainting indicator breaks that promise. When you scroll back through history, the arrows line up beautifully, because the indicator has been redrawn with knowledge of what came next.

That is the trap. The backtest looks honest. The live results do not match it. You then conclude that the system is bad, or that you are bad at trading it, when the real fault sits in the indicator. Many traders lose confidence in the wrong thing and abandon a sound process, or keep trusting a broken tool.

MT4 includes a built-in Strategy Tester. It is useful for replaying history, checking how an Expert Advisor behaves and testing settings. The official MetaQuotes help pages explain how the tester runs. But the tester cannot tell you, on its own, whether an indicator repaints. A replay of history can look perfect even when the live signal was unstable. You have to test the behaviour directly, while the candle is still forming, on a live demo feed.

That gap is what this checklist fills. It does not replace the tester. It asks a different question: did the signal you saw at the moment of decision stay exactly where it was?

There is a practical reason to care about this before you optimise anything. Optimising a repainting indicator simply finds the settings that fit the redrawn history best. The result looks like progress, but it is curve fitting to a signal that cannot be traded. Fix the repaint problem first, then test settings.

How do you test an indicator for repainting yourself?

The fastest honest test takes about thirty minutes. You need a demo account, a chart, and a way to take screenshots with the time visible. You do not need special software.

The idea is simple. You watch a signal form in real time. You let the candle close. Then you check whether the signal moved, changed size or disappeared. If anything changed after the candle closed, the indicator is repainting.

Use a low timeframe, such as M1 or M5. Repainting is easier to catch when candles form quickly, because you get more chances to see a signal in the act. Keep the chart in view while you wait. Take the first screenshot the moment an arrow or line appears, with the clock visible in the frame. Then wait for the candle to close and for two more candles to form. Take a second screenshot of the same area and compare the two.

Ask yourself three questions when you compare them. Did the signal move to a different bar? Did its size or position change? Did it vanish entirely? Any yes means the indicator repainted at that moment.

After that, run the same settings through the strategy tester in visual mode. The signals should appear on the same bars in the tester as they did on the live chart. If the tester shows signals on bars where the live chart showed nothing, or the other way round, treat it as a second warning sign.

Keep a written record. Note the date, the time, the symbol, the timeframe, the indicator settings and what you saw in each screenshot. A written record is much harder to argue with than a memory. It also lets you repeat the test later, on another day, and see whether the behaviour holds.

Do not stop after one signal. One clean test can be luck. Run the test across several signals, in different sessions if you can, before you decide. The aim is to see a pattern, not to win an argument with a sales page.

How do you tell a real tool from a sales pitch?

Some answers tell you a lot before you run any test. The checklist asks five questions. Here they are in plain language, with the signs you want to see and the signs that should make you slow down.

The first question is what data the indicator uses. A good answer says it uses closed candles and states that in the description. A weak answer points to a “proprietary AI formula” with no inputs you can see and no plain explanation of the data.

The second question is whether you can see and change the inputs. A good tool exposes its periods, price sources and thresholds. You can change them and watch the effect. A locked black box makes it impossible to check how it works, and it hides the very behaviour you need to test.

The third question is whether it lags honestly. A trustworthy vendor says how many bars the signal needs before it confirms. Every confirmation rule adds delay, and that delay is a fair trade-off you can measure. Be wary of any claim of no lag, no repaint and near-perfect accuracy all at once. You can believe at most one of those claims, and usually none of them without a test.

The fourth question is how alerts behave. Alerts that fire when the candle closes, and that you can configure, are a good sign. Alerts that fire in the middle of a candle and then re-fire as the signal flickers are a warning. A flickering alert is often a repainting signal in disguise.

The fifth question is how the indicator behaves in the tester. It should work in the tester using the same rules it uses live. If the seller says it is “not suitable for backtesting”, read that as a warning. It often means the history is being redrawn, and the seller would rather not show you that.

Use these questions before you spend time on a test. A tool that fails two or three of them rarely improves after you install it. Still, a good answer on paper is not proof. It only tells you the seller is willing to be specific. The repaint test is what shows you whether the specifics are true.

How do you score an indicator before it touches a trade?

The scorecard turns a feeling into a decision. Give the indicator a clear yes or no on each point. If it fails more than one point, keep it off your live chart. Treat it as something to study, not something to trade.

The first point is that it answers one question well. Trend, momentum, levels or volatility are all valid jobs. A tool that promises all four usually does none of them well, and it is hard to know which part is working when you lose.

The second point is that it agrees with price more often than it argues with it. If you need three extra filters just to make the signal usable, the indicator is probably not the tool. The filters may be doing the real work, and they may be hiding weak signals behind a neat looking output.

The third point is that you can explain, in one sentence, what would make its signal wrong. Every real signal has a failure case. If nothing can make the signal wrong, the indicator is decoration. You cannot manage a risk you cannot describe.

The fourth point is that you have two weeks of forward screenshots before it influences a single trade decision. Forward screenshots show you what the indicator did in real time, not what it looked like after the fact. Two weeks is a minimum, not a target. It gives you enough sessions to see the tool behave on different days.

Above the four points sits one rule. Judge an indicator only by what it showed at the moment of decision. A signal you could not have seen in real time was never a signal. This rule removes most of the arguments traders have with themselves about whether a signal was “really there”.

Keep the scorecard honest. Do not score a tool higher because you like the colours or the sales video. Score it from your screenshots. If the screenshots are missing, the score is a no until you can prove otherwise.

What does the Indicator Repaint Test Checklist cover?

The checklist has three sections. Each one does a different job, and they work best in order.

Section one is the repaint test. It is the thirty-minute test described above. It gives you a repeatable method: watch the signal form, let the candle close, look for any change, and then compare the strategy tester with the live chart. Its only job is to catch repainting. If the indicator fails here, the other sections do not matter.

Section two is the question table. It lists the five questions from earlier, each with a good answer and a bad answer. Use it when you read a sales page, watch a product video or talk to a seller. It helps you spot vague claims that cannot be checked, and it gives you specific follow-up questions to ask.

Section three is the selection scorecard. It contains the four scorecard points from the previous part and closes with the repaint rule as a reminder. Use it after the repaint test, when you decide whether an indicator earns a place in a strategy at all.

The checklist is short on purpose. It fits on a few printed pages, so you can keep it next to your charts while you work. It uses plain words and simple checkboxes, so you can tick off each step during a session without reading long paragraphs.

The checklist also ends with a short risk note. It reminds you that forex and CFD trading carries a high risk of loss, and that most retail accounts lose money. That note is there on purpose. A repaint test makes an indicator more honest. It does not make trading safe.

What does the checklist NOT do?

This part matters as much as the rest. The checklist is a test. It is not a promise, a certificate or a ranking of indicators.

It does not prove that an indicator is safe. An indicator can pass every check and still lose money in your hands. Position size, stop placement, news releases, spreads and your own emotions all sit outside the indicator. A clean repaint result tells you about one part of the system, not about the whole system.

It does not prove the indicator will not repaint on your broker feed. Your broker’s price data, server time and tick flow can differ from another broker’s. A demo test shows how the indicator behaved on the feed you tested, on the symbols and sessions you watched. It cannot promise the same behaviour on every feed, every symbol and every market condition.

It does not replace a demo run. The checklist tells you what to look for and how to record it. Only a longer demo run, with written notes and screenshots across different days, shows how the indicator behaves in a range of conditions. Plan that run before any live money goes in.

It does not measure profit. It does not give you a win rate, a drawdown figure or an expected return. Those numbers must come from your own tests, on data you can verify yourself. Any site that offers a win rate for an indicator, without showing you how it was measured, deserves the same scepticism as the repainting sales pitch.

It does not cover order execution, slippage or your broker’s conditions during fast markets. Those are separate questions. Test them separately, on demo, before you rely on any automated or semi-automated setup.

How do you use the checklist on a demo account?

Start with a demo account. Never run the first test on a live account, even when the test looks simple. Set the demo balance and lot size to something close to what you would really trade, so the test feels realistic without putting money at risk.

Open the checklist and read it once from start to finish. Do not jump straight to the test. The first read tells you what each section is for and what the scorecard expects from you.

Go back to section one and run the repaint test on one symbol and a low timeframe. Keep the notes and the screenshots together in one folder, named by date. Repeat the test on several signals before you move on. If the indicator repaints on any of them, mark it and stop there.

Next, work through the question table using the seller’s own documentation. Write down every answer that is vague or missing. A vague answer is not always a bad answer, but it is a question you still need to answer yourself. Do not accept “trust us” as an answer to any of the five questions.

Then finish with the scorecard. Mark each point yes or no, and write one sentence for the failure case. If you cannot write that sentence, mark the failure-case point as no. Keep the indicator on demo for at least two weeks of forward screenshots. Review those screenshots once a week and note any change you find, even a small one.

Only after that should you consider a live account, and then only with a small size and written loss limits. Set your loss limit before you open the first trade, not after a bad day. The checklist does not decide that number for you. Your own risk rules do.

Where do MT4 testing tools and related guides fit?

MT4 and MT5 both include testing tools, and the official help pages explain how they work. The Strategy Tester replays history and can show signals on a chart. The checklist answers a different question. The tester replays what already happened. The checklist watches what happens while a candle is still forming. You need both views, and you need to know which one you are using.

If you use an Expert Advisor, the same logic applies. An EA that relies on a repainting indicator will look better in the tester than it performs live. Before you trust any EA, run its indicator through the repaint test first. Then check the EA’s behaviour on demo.

Trend filters are a common place for repainting to hide. If you are comparing trend tools, the guide to free Supertrend indicators for MT5 explains what a trend filter should show, and the same repaint checks apply to it. Use the scorecard on any trend tool before you combine it with other signals.

Divergence tools need the same care. The guide to non-repainting divergence indicator settings covers how to set up those tools. Repainting in a divergence tool is especially misleading, because the lines can look convincing after the candle has moved on. Run the repaint test first, then adjust settings.

Stop placement is a separate decision from signal quality. A repainting signal and a poorly sized stop can both hurt your account. The guide to the best ATR multiplier for stop loss explains one way to size stops from volatility. Use it as a second check after the indicator has passed the repaint test.

If you also look at order block tools on TradingView, run the same five questions and the same scorecard. A tool on a different platform can still repaint, and the checklist logic does not depend on which platform draws the lines. Only the screenshots change.

When an Expert Advisor stops working after a terminal update, the cause is often unrelated to the indicator. The guide to fixing an MT4 EA that stopped working after an update covers the common causes. Check that first, so you do not blame a good indicator for a platform problem.

Finally, if you edit an indicator’s source code and the compiler reports errors, read how to fix compile errors in MQL4. Fixing the compile error does not tell you whether the indicator repaints. Run the test after the code compiles, not before.

Which settings and symbols should you test first?

Keep the first test simple. Use the default settings the seller ships. Once the default version has passed the repaint test, change only one input at a time and repeat the test. If a signal shifts when you change a setting, write down which setting caused it. That record tells you which inputs are fragile.

Choose a symbol you already know. A symbol with cleaner, more consistent candles makes the screenshots easier to read. Thin or unusual symbols can produce gaps and spikes that look like repainting even when the indicator is behaving correctly. Use the same symbol for the full test, so you compare like with like.

Be careful with settings that add lookback, smoothing or multi-timeframe calculations. These inputs can change how much of the history the indicator uses. Some of them are the most likely place for a repaint to hide, because they quietly reach into bars that were not complete at the time. Test them with extra care, and check each screenshot against the time on the chart.

Do not choose settings because a backtest looks perfect. A perfect backtest on a repainting indicator is a warning, not a result. Only settings that pass the live repaint test deserve a closer look in the tester.

What should you do next?

Download the Indicator Repaint Test Checklist below. Print it, or keep the PDF open next to your demo chart. Then run section one today on one indicator you already use or are considering.

Keep your notes and screenshots. Do not let any indicator influence a live trade until it has passed the repaint test, scored well on the selection scorecard, and produced two weeks of forward screenshots on demo. If it fails any step, set it aside. A tool you cannot verify does not deserve a place in your process.

Remember what the checklist can and cannot do. It can catch repainting and weak claims. It cannot prove safety, cannot prove the same behaviour on your broker feed, and cannot replace a demo run. Those limits are the reason the final step is still yours.

Trading forex and CFDs carries a high risk of loss, and most retail accounts lose money. Nothing in this guide is financial advice. Test every system on a demo account with a written loss limit before you risk real funds, and stay patient while you do it.