Indicator Lag and Whipsaw in MT4: How to Measure Both, and Slower Settings Cut Whipsaw in 14 of 16 Grids While Only 4 of 100 Runs Passed
Key takeaways
- You can measure an indicator’s lag and whipsaw on a MetaTrader 4 chart in about fifteen minutes with the Data Window and the crosshair. Below is the method and the numbers to compare against.
- I ran 25 settings of five indicators (EMA cross, Donchian, Supertrend, HalfTrend, RSI) on EURUSD and gold, M5 and M15: 100 runs.
- Slower settings cut whipsaw step by step in 14 of 16 grids. EMA 5/13 on EURUSD M5 flipped back within 6 bars 43.6% of the time, EMA 50/200 4.0%.
- They also made every signal later. The same EMA cross went from 3.21 to 6.24 ATR behind the swing.
- Only 4 of 100 runs passed: EMA 9/21 on gold M5 and EURUSD M15, EMA 12/26 and HalfTrend 6 on EURUSD M15. 22 of 100 made money at all.
“Smooth” and “lagging” are the two words every indicator page uses, and neither comes with a number. In the statistical parameters of 21 indicators I found that both numbers are stable properties of the indicator: they ranked the same way on every chart. That makes them worth measuring on your own MT4 chart before you trust an indicator with money.
How do I measure whipsaw in MT4?
- Open the chart you trade, for example EURUSD M5, and add the indicator with the settings you plan to use.
- Press Ctrl+D to open the Data Window. Hovering any bar shows the indicator’s values on that bar, so you can see exactly where a cross or a colour change happened.
- Scroll back one full trading day. Write down every buy and sell signal and its bar time.
- Count the signals that were followed by an opposite signal within 6 bars. On M5 that is 30 minutes.
- Whipsaw rate = those signals divided by all signals. Do it for three different days and take the average.
Only count closed bars. A signal that appears and disappears inside a live bar is a repaint, not a whipsaw. The ten-minute check for that is in how to check if an indicator repaints.
How do I measure lag in MT4?
- Add ATR with period 14 from Insert, Indicators, Oscillators, Average True Range. Its value shows in the Data Window.
- For each signal, find the lowest low (for a buy) or highest high (for a sell) of the 24 bars before it.
- Press Ctrl+F for the crosshair and drag it from that extreme to the signal bar’s close. MT4 shows the distance in points.
- Divide the distance by the ATR value on the signal bar. That is lateness in ATR.
- Take the median over at least 20 signals, so one news spike does not decide it.
For scale: on the charts below a random bar sits 2.2 to 2.7 ATR from its own 24-bar extreme. A signal near that number is neither early nor late. Well below it is early, well above it is late.
Working in ATR rather than pips means the numbers below apply to your chart even if your broker’s volatility is different. ATR itself is covered in ATR indicator settings for MT4.
What numbers should I expect?
Data: EURUSD M5 June 3 to August 26, 2026, 16,926 bars; gold M5 June 16 to August 26, 13,736 bars; M15 built from them. Profit per trade uses one exit for every setting: stop 1.5 ATR, target 3 ATR, 48 bars at most, 1 pip of cost on EURUSD and $0.30 on gold. A pass needs a profit after cost, a profit in both halves of the period and a result above the 95th percentile of random entries.
EMA crossover (Moving Average, method Exponential)
| Setting | EURUSD M5 signals/day | whipsaw | lateness | pips/trade | Gold M15 signals/day | whipsaw | lateness | $/trade |
|---|---|---|---|---|---|---|---|---|
| EMA 5/13 | 23.6 | 43.6% | 3.21 | −1.25 | 6.7 | 42.9% | 2.59 | −0.40 |
| EMA 9/21 | 15.0 | 33.5% | 3.63 | −1.68 | 3.9 | 30.4% | 2.98 | 0.60 |
| EMA 12/26 | 11.5 | 28.0% | 3.94 | −1.31 | 2.7 | 22.6% | 3.45 | 1.95 |
| EMA 20/50 | 5.3 | 14.1% | 4.84 | −0.85 | 1.5 | 17.3% | 3.85 | −1.72 |
| EMA 50/100 | 2.3 | 7.4% | 6.09 | −0.69 | 0.7 | 9.1% | 4.58 | −0.84 |
| EMA 50/200 | 1.7 | 4.0% | 6.24 | −0.58 | 0.5 | 7.7% | 3.59 | −3.25 |
Supertrend (period 10, multiplier changed)
| Setting | EURUSD M5 signals/day | whipsaw | lateness | pips/trade | Gold M15 signals/day | whipsaw | lateness | $/trade |
|---|---|---|---|---|---|---|---|---|
| Supertrend 10/1.5 | 30.0 | 46.8% | 2.92 | −1.40 | 5.5 | 27.9% | 2.68 | −0.01 |
| Supertrend 10/2.0 | 21.7 | 32.8% | 3.22 | −1.65 | 3.5 | 16.8% | 3.17 | 1.14 |
| Supertrend 10/3.0 | 12.5 | 15.6% | 4.01 | −1.58 | 2.1 | 5.0% | 4.01 | 2.16 |
| Supertrend 10/4.0 | 7.8 | 7.5% | 4.75 | −1.93 | 1.5 | 0.0% | 4.68 | 0.16 |
HalfTrend (amplitude changed)
| Setting | EURUSD M5 signals/day | whipsaw | lateness | pips/trade | Gold M15 signals/day | whipsaw | lateness | $/trade |
|---|---|---|---|---|---|---|---|---|
| HalfTrend 1 | 25.9 | 40.0% | 3.15 | −1.38 | 6.3 | 29.8% | 2.56 | 0.75 |
| HalfTrend 2 | 17.2 | 21.4% | 3.44 | −1.51 | 4.3 | 15.6% | 2.93 | 0.86 |
| HalfTrend 3 | 13.2 | 11.9% | 3.80 | −1.30 | 3.1 | 4.6% | 3.20 | 2.07 |
| HalfTrend 4 | 10.8 | 5.4% | 4.09 | −1.46 | 2.6 | 2.4% | 3.42 | 2.14 |
| HalfTrend 6 | 8.0 | 1.7% | 4.61 | −1.76 | 1.9 | 0.0% | 4.06 | 2.88 |
Donchian channel breakout (period changed)
| Setting | EURUSD M5 signals/day | whipsaw | lateness | pips/trade | Gold M15 signals/day | whipsaw | lateness | $/trade |
|---|---|---|---|---|---|---|---|---|
| Donchian 10 | 30.3 | 8.4% | 5.28 | −1.29 | 9.7 | 9.3% | 4.00 | −0.52 |
| Donchian 20 | 20.7 | 2.4% | 6.23 | −1.45 | 6.4 | 2.2% | 4.74 | 2.07 |
| Donchian 30 | 16.8 | 1.2% | 6.64 | −1.36 | 5.0 | 0.4% | 5.10 | 0.70 |
| Donchian 55 | 12.7 | 0.7% | 7.07 | −1.21 | 3.5 | 0.0% | 5.57 | −0.92 |
RSI exit from oversold and overbought
| Setting | EURUSD M5 signals/day | whipsaw | lateness | pips/trade | Gold M15 signals/day | whipsaw | lateness | $/trade |
|---|---|---|---|---|---|---|---|---|
| RSI 14, 20/80 | 1.1 | 0.0% | 1.32 | −1.50 | 0.5 | 0.0% | 1.42 | −4.18 |
| RSI 14, 25/75 | 2.9 | 0.0% | 1.45 | −1.13 | 1.3 | 0.0% | 1.27 | −2.34 |
| RSI 14, 30/70 | 6.9 | 0.0% | 1.46 | −0.54 | 2.6 | 0.0% | 1.36 | −3.31 |
| RSI 14, 35/65 | 14.6 | 1.6% | 1.48 | −1.19 | 4.7 | 0.4% | 1.33 | −0.73 |
| RSI 7, 30/70 | 22.0 | 5.7% | 1.40 | −1.00 | 7.3 | 2.5% | 1.24 | −0.93 |
| RSI 21, 30/70 | 2.8 | 0.0% | 1.53 | −0.75 | 1.3 | 0.0% | 1.51 | −3.71 |
Gold M15 runs have fewer trades: under 100 for most slow settings and 22 for RSI 14 at 20/80, so read them as a direction, not a verdict.
Which settings passed?
| Setting and chart | Trades | Profit per trade | Whipsaw | Lateness, ATR |
|---|---|---|---|---|
| EMA 9/21, gold M5 | 366 | +$1.16 | 26.0% | 3.04 |
| EMA 9/21, EURUSD M15 | 173 | +0.33 pips | 29.5% | 3.26 |
| EMA 12/26, EURUSD M15 | 144 | +1.51 pips | 22.9% | 3.74 |
| HalfTrend 6, EURUSD M15 | 125 | +0.59 pips | 0.7% | 4.33 |
Four passes in 100 runs. Three of the four were middle-speed EMA crosses with 23% to 30% whipsaw, which is a lot. The lowest-whipsaw settings, Donchian 55 and RSI at 20/80, passed nothing. More moving average periods traded on their own are in moving average settings for MT4.
Which setting should I pick in MT4?
- Pick the whipsaw you can live with, then accept the lag that comes with it. In 14 of 16 grids every step slower cut whipsaw and added lag. You cannot have both low.
- Do not slow down to fix losses. Profit per trade improved with slower settings in 8 of 16 grids and got worse in 8.
- Check the chart, not only the setting. Supertrend 10/3 lost 1.58 pips a trade on EURUSD M5 and made $2.16 a trade on gold M15.
- A near-zero whipsaw rate is a warning, not a feature. It means the rule needs a big move to change its mind, and it will enter late. HalfTrend 6 is the exception that passed, once.
The lag-against-whipsaw trade-off across different indicators, not settings, is in indicator lag vs whipsaw. Supertrend settings traded on their own are in Supertrend settings tested, and Heikin Ashi, the noisiest rule of all, in Heikin Ashi settings for MT4.
Frequently asked questions
How do I measure indicator lag in MT4?
Measure the distance from the 24-bar extreme to the signal close with the crosshair (Ctrl+F) and divide it by ATR(14) from the Data Window.
What is a good whipsaw rate?
There is no good one on its own. The four passing settings had 0.7% to 29.5%. Compare it with lateness, because in 14 of 16 trend grids every cut in whipsaw added lag.
Which EMA settings had the fewest false signals?
EMA 50/200: 4.0% on EURUSD M5 and 7.7% on gold M15. It lost money on all four charts.
Are Supertrend, HalfTrend and Donchian built into MT4?
No. EMA, RSI and ATR are built in; the other three need a download from the MQL5 Code Base.
How many signals should I count?
At least 20 per setting, over three or more days, so one session does not decide the number.
Want a line that does not flip on every wobble?
Get my own enhanced DeMARK Trend Line indicator for MetaTrader 4 and 5. Non repaint, clean, and free.
Forex educator and indicator developer. I build and trade my own MetaTrader tools, and share the ones that genuinely help.