MACD-V explained: is MACD divided by ATR comparable across markets?

Contents

1. What is MACD-V and how is it calculated?
2. What do the 150 and -150 levels mean?
3. How does MACD-V differ from standard MACD?
4. Are MACD-V readings consistent across markets?
5. How do you test a MACD-V crossover rule?
6. What should you check before adding MACD-V to a strategy?
7. Frequently asked questions
8. References

What is MACD-V and how is it calculated?

MACD-V is the standard MACD line divided by the 26-period Average True Range and multiplied by 100, so it measures the gap between the two moving averages in units of the market's own volatility instead of price points. That puts every market on one scale, with 150 and -150 marking a strong but stretched trend. One scale does not mean one behaviour, so each market still needs its own test.

Three terms carry the whole idea. The MACD line is the 12-period exponential moving average of price minus the 26-period exponential moving average, which is the usual MACD you already know. An exponential moving average (EMA) is an average that gives more weight to recent bars than to old ones. The Average True Range (ATR) is the average size of a bar's full range over a set number of bars, including any gap from the previous close, so it is a plain measure of how much the market moves per bar. The MACD-V indicator uses a 26-period ATR on purpose: it is the same length as the slow average, so the gap and the yardstick cover the same stretch of history.

The only formula you need:

MACD-V = (MACD line ÷ ATR(26)) × 100

Here the MACD line is the 12-period EMA minus the 26-period EMA in price points, ATR(26) is the 26-period Average True Range in the same price points, and × 100 turns the ratio into a number such as 120 rather than 1.2. The points cancel, which is the whole trick: a reading of 100 means the two averages are exactly one ATR apart, on any market and at any price level. Think of describing the distance between two cars as "two car lengths" instead of "nine meters": the first description still makes sense when you switch from a small car to a truck.

Among the technical analysis tools and indicators you can actually test, MACD-V belongs to the group whose output has a fixed meaning, which is what lets you write a rule with named levels instead of eyeballing a chart.

What do the 150 and -150 levels mean?

A MACD-V reading above 150 or below -150 means the fast average is more than one and a half ATRs away from the slow average, a zone where the trend is strong but stretched, while readings between -50 and 50 mean directional momentum is small compared with the market's own noise. Because the scale is fixed, these levels keep their meaning on every market and in every year, which standard MACD levels cannot do.

Zone	MACD-V reading	What it describes
Strong and extended up	above 150	Averages more than 1.5 ATR apart; trend strong, move stretched
Rising trend	50 to 150	Clear upward momentum, between half an ATR and 1.5 ATR
Ranging or weak momentum	-50 to 50	Gap smaller than half an ATR; noise dominates direction
Falling trend	-150 to -50	Clear downward momentum
Strong and extended down	below -150	Averages more than 1.5 ATR apart to the downside

"Stretched" is a description, not an instruction. A reading of 180 says the market has moved far from its slow average relative to its usual bar size; whether that leads to a pause, a reversal or a further run is a question for a test on your market, not something the level decides. The practical use of the bands is as filters in a rule: for example, take a signal-line crossover only when MACD-V sits between 50 and 150, or avoid new entries above 150. Each such filter is a setting, and the last section explains why the count of settings matters.

How does MACD-V differ from standard MACD?

Standard MACD is measured in price points, so the size of its swings changes with the price level and with current volatility, while MACD-V is measured in volatility units and keeps one scale; zero-line crossovers fall on the same bars in both, but signal-line crossovers can land on different bars. The signal line is a 9-period EMA of the indicator itself, so on MACD-V it is an EMA of a ratio whose denominator moves every bar, and that shifts where the two lines meet.

Property	Standard MACD	MACD-V
Unit	Price points	Volatility units (ATR × 100)
Scale over time	Grows with price and volatility	Stays within a similar band
Fixed levels	Only the zero line is fixed	Zero line plus 50, 150 and their negatives
Zero-line crossovers	Same bars	Same bars (ATR is always positive)
Signal-line crossovers	EMA of a price-point series	EMA of a ratio; can shift by a bar or more
Histogram	In points, hard to compare across years	In volatility units, comparable across years

The zero-line result follows from the formula: dividing by a positive number never changes the sign, so MACD-V is above zero exactly when the MACD line is. The signal-line result is less obvious. If volatility rises while the gap between the averages stays the same, the MACD-V value falls even though the MACD value does not, and the 9-period EMA of MACD-V bends down a bar or two before the EMA of MACD would. A strategy built on signal-line crossovers therefore produces a different trade list on the two indicators, and the only way to know which list you prefer is to run both on the same bars. The entry and exit rules in MACD indicator explained: calculation, signals and settings are the right baseline for that comparison.

Are MACD-V readings consistent across markets?

Dividing by ATR removes the price-level difference between markets, so a MACD-V of 120 on Nasdaq-100 futures, gold futures and Mini-TAIEX describes the same stretch of momentum relative to each market's own noise, but it does not make the markets behave alike after that reading. The number is comparable; the follow-through is not, so every market still needs its own test.

What stays different is everything the indicator does not measure. Session structure, the share of overnight gaps in the daily range, how often a stretched reading turns into a reversal, and the cost of a round trip in ticks all differ by market, and all of them shape what happens after a MACD-V of 120. Comparable readings are useful for two things: writing one rule with named levels that you can carry from market to market without rescaling, and comparing the results of that rule across markets on one scale. Neither of those means the rule will pass on the second market because it passed on the first. Run the same test on each market and compare the trade lists.

How do you test a MACD-V crossover rule?

Write the entry, exit, levels, costs and session before the first run, test with commission and slippage on the same bars as a plain MACD crossover baseline, and read any kept result as one candidate out of many tried. The baseline is the important part: MACD-V only earns its place if its trade list is different from, and more acceptable than, the list from the indicator it replaces.

1. Write the rule in full. For example: go long when the MACD-V line crosses above its signal line while MACD-V is between -50 and 150; exit when it crosses back below; mirror for shorts. Write down the 12, 26, 9 and ATR 26 settings and the band limits.
2. Set commission, slippage and the session. Use the same values for every run in the comparison and decide whether the test uses day-only or full-session bars.
3. Run the identical rule with standard MACD on the same bars. This is the baseline. Same dates, same costs, same execution on the next bar.
4. Compare trade count, average trade after costs and maximum drawdown. A different trade count tells you how many signal-line crossovers moved; the other two tell you whether the move helped.
5. Run neighbouring settings and the out-of-sample period. Check that ATR 22 and ATR 30 behave like ATR 26, and that the rule holds on data it never saw, before you decide.

When the MACD-V rule and its MACD baseline both run through the backtesting of AlphaPrime (Valynox) with the same commission, slippage, session and next-bar execution, the only thing that differs between the two trade lists is the indicator, which is exactly the comparison this test needs.

To show what a report from such a test looks like, we ran a search on NQ daily bars from 2019-01-02 to 2025-06-30, long and short, one contract, with $20 commission and $10 slippage per trade. The run used an N Day rule as its only entry and exit, with no stop, target or filter; it is a stand-in built from standard indicators and does not test MACD-V itself. The result below is the one candidate the search kept out of 1,024 it tried, chosen for its return against its drawdown. Picking the kept candidate after the fact flatters it: the best of a thousand tries always carries a selection bonus, which is the point Bailey, Borwein, López de Prado and Zhu (2014) make about any search that keeps its best result.

Read the summary the way step 4 says. Net profit of $145,750 sits against a maximum drawdown of $129,920, so almost the whole gain was at risk at one point, and 78 trades over six and a half years is a thin sample for judging the average trade of $1,868.59. The time in market of 90.69% says this rule is nearly always positioned, which on a market that rose over most of the period is a large part of what you see. None of these numbers describe what MACD-V would do or what a reader can expect; they show which lines of a report to read first and how a kept candidate should be questioned.

What should you check before adding MACD-V to a strategy?

Count how many parameters you are tuning, check whether a smoothed input changes the signal count, and decide whether a context filter deserves a separate test, because every added choice is one more way to fit the past. MACD-V tempts you to add filters because its levels are fixed and easy to name, and that is exactly when the parameter count climbs without anyone noticing.

1. Count the settings. A plain MACD-V crossover already has four (12, 26, 9, ATR 26). Each band limit you use as a filter adds one more. Divide your in-sample trade count by the number of settings; a low ratio means the fit has more freedom than the data can discipline.
2. Test the input bars separately. Some traders feed MACD-V with smoothed candles. Heikin Ashi candles and how they are tested explains why the smoothing changes both the averages and the ATR, so the signal count can move a lot. Run the rule on ordinary bars first, then on the smoothed ones, and compare the two trade lists before you keep either.
3. Keep context filters as their own candidates. A filter such as gamma exposure levels and how to test them is a second hypothesis. Test it on its own, then combined, so you can see which part of any improvement it contributed.
4. Prefer the fewest levels that make the rule complete. If a rule works with the zero line and the signal line alone, the 150 band is a refinement to test later, not a starting condition.

A concrete next step: take a market you already trade, write a MACD-V signal-line crossover rule and its standard MACD twin on one page, run both on the same bars with your real commission and slippage, and note how many trades the two lists share. That number tells you how much MACD-V actually changed before you look at a single profit figure.

Frequently asked questions

Does MACD-V use the same 12, 26 and 9 settings as MACD?

Yes. The MACD line is still the 12-period EMA minus the 26-period EMA, and the signal line is still a 9-period EMA of the indicator. The one addition is the 26-period ATR used as the divisor, chosen to match the slow average's length. The multiplication by 100 only changes the display scale, not the signals.

Why can MACD-V signal-line crossovers differ from MACD crossovers?

The signal line is an EMA of the indicator itself. On MACD-V the indicator is a ratio whose denominator, the ATR, changes every bar, so its EMA bends differently from the EMA of the raw MACD line. Zero-line crossovers are unaffected, because dividing by a positive ATR never changes the sign. Signal-line crossovers can move by a bar or more, so the trade lists differ.

Is a MACD-V reading of 120 on one market the same as 120 on another?

As a description, yes: on both markets the fast average sits 1.2 ATR above the slow one. As a prediction, no: how often such a reading leads to a continuation or a reversal depends on each market's own behaviour, costs and sessions. Use the common scale to write one rule and compare results, then test the rule on each market separately.

Can MACD-V be used on intraday bars?

Yes, the formula does not depend on the bar size. On intraday bars the ATR reflects the range of that bar length, so the volatility units adjust automatically. Session breaks and overnight gaps enter the ATR through the true range, so decide whether your test uses day-only or full-session bars and keep that choice fixed across the comparison.

References

• Wilder, J. W. (1978). New Concepts in Technical Trading Systems. Trend Research.
• Bailey, D. H., Borwein, J., López de Prado, M., & Zhu, Q. J. (2014). Pseudo-Mathematics and Financial Charlatanism: The Effects of Backtest Overfitting on Out-of-Sample Performance. Notices of the American Mathematical Society, 61(5), 458–471.
• Pardo, R. (2008). The Evaluation and Optimization of Trading Strategies (2nd ed.). John Wiley & Sons.