Average True Range (ATR)

Average True Range measures how much a market typically moves in a given period — its normal "size of a bar." It doesn't tell you direction; it tells you scale.

In 30 Seconds

• ATR measures volatility — the typical size of a bar's range, not which way price is going.
• Bigger ATR = bigger moves. It rises when a market gets choppy and falls when it goes quiet.
• It's a ruler, not a signal. ATR tells you *how far* price normally travels — used for sizing stops and targets.
• MPM uses ATR as a distance unit — "how stretched" or "how far from a level" is often measured in ATRs, not raw points, so it's comparable across markets and eras.

Definition

Average True Range is the average of the "true range" over a chosen number of periods (commonly 14). It was introduced by J. Welles Wilder in 1978 as a way to measure volatility that accounts for gaps between bars.

The true range of a single bar is the largest of three distances:

1. The current bar's high minus its low.
2. The current bar's high minus the previous bar's close.
3. The current bar's low minus the previous bar's close.

Taking the largest of these three captures the full extent of a bar's movement — including any gap from the prior close that a simple high-minus-low would miss. The Average True Range is then the moving average of that true range over the lookback period.

The result is a single number in the market's own price units. If ES has an ATR of 12 points on the hourly, a "typical" hourly bar has spanned about 12 points. If it climbs to 20, the market has become more volatile; if it drops to 6, it has quieted down.

"ATR measures how far price moves, not which way it moves."

Why It Matters

Every risk decision depends on knowing how much a market normally moves. A stop placed 10 points away means something completely different in a market with an ATR of 5 than in one with an ATR of 30 — in the first, it's far outside normal noise; in the second, it's well inside it and likely to be hit by ordinary movement.

ATR turns "how volatile is this?" from a feeling into a number. That makes it useful for three things in particular:

• Sizing stops and targets to the market's actual movement rather than to a round number that ignores volatility.
• Judging whether a move is unusual. A 30-point move means little on its own; a 30-point move when ATR is 8 is a roughly-four-ATR move — genuinely large.
• Comparing markets on equal footing. Raw points can't be compared across a stock index and a commodity, but "two ATRs" means the same *relative* thing in both.

How It Works (a simple example)

Suppose a market's last three hourly bars had true ranges of 8, 12, and 10 points. A simple 3-period ATR would be their average: (8 + 12 + 10) / 3 = 10 points. So the "typical" recent bar spanned about 10 points.

Now imagine the next bar spans 30 points. That single bar is three times the recent ATR — a clear signal that volatility just expanded sharply. ATR itself would begin to rise as that larger range feeds into the average. This is the core behavior: ATR climbs as ranges widen and falls as they narrow, lagging slightly because it's an average.

Note what ATR did *not* tell you: whether that 30-point bar was up or down. Volatility is directionless. A violent rally and a violent selloff can produce identical ATR.

How MPM Uses ATR

Within MPM, ATR is used as a measurement tool — not a trading signal. It is one of the standardized units HIE uses when expressing distance and volatility (alongside standard deviation, z-score, and percentile), because an ATR-based measure stays comparable across the different markets and eras HIE covers.

• As a distance unit. Concepts like "stretched from its average" or "how far price sits from a level" can be expressed in ATRs rather than raw points, because an ATR-based distance is comparable across markets and volatility eras.
• As a volatility descriptor. Whether a market is currently "quiet" or "volatile" is grounded in volatility measurement that ATR is part of the family of — so a condition like "after a quiet stretch" has an objective, disclosed meaning.

HIE uses a 14-period ATR — with a nuance worth stating honestly: the primary volatility measure (behind the "quiet"/"volatile" regime language) uses Wilder's smoothing, while the reference-level "near" bands use a simple 14-period average of true range. Both are period 14; they differ in smoothing by surface. There is no special per-timeframe handling — 4-hour conditions are produced by resampling 1-hour bars and running the same computation. ATR is also ranked against its 500-bar history, which appears in reports as "ATR in the top/bottom/middle third of its 500-bar history."

In practice, "stretched" or "overextended" isn't a silent threshold HIE applies behind your back — it's a guided question. Because "stretched" is ambiguous, the engine never runs it silently; it asks which reference you mean: stretched far above the 20 EMA, far above the 200 EMA, or far below the 200 EMA. Each resolves to a fixed distance of 2.5 ATR from the chosen moving average, and the choice you made is shown in the "What ran" line — for example:

'stretched' → Stretched far above the 20 EMA

This distance is measured in ATR units internally, which keeps it comparable across markets and eras — there is no percent-distance version such as "6% above the 200-day average." If you state your own explicit number in a question, HIE uses your number rather than the 2.5 default.

Interpretation

• Rising ATR — the market is making larger moves; ranges are expanding. Often associated with breakouts, news, or trend acceleration — but ATR alone doesn't say which.
• Falling ATR — the market is quieting; ranges are contracting. Often associated with consolidation or low-participation periods.
• A single bar far larger than ATR — an outlier move relative to recent normal; worth noting, but one bar is not a trend.
• ATR compared across markets — only meaningful when expressed relative to price or as a ratio; 10 points of ATR is large for one instrument and tiny for another.

The honest reading of ATR is always relative: it tells you what's normal *for this market, right now*, so you can judge whether something is unusually large or small against its own recent behavior.

Common Mistakes

"Treating ATR as a direction signal."

ATR is directionless. A high ATR means big moves, not up moves. Any tool or claim that reads rising ATR as bullish or bearish is misusing it.

"Comparing raw ATR across markets."

An ATR of 15 is meaningless without context. Fifteen points is enormous on one instrument and trivial on another. Compare ATR relative to price, or use it within a single market.

"Assuming ATR predicts the next move's size."

ATR is a trailing average of recent ranges. It describes what has been normal; it doesn't guarantee the next bar. Volatility clusters, so recent ATR is informative — but it's history, not a forecast.

"Confusing a large move with a meaningful one."

A move of several ATRs is large, but "large" isn't the same as "tradeable" or "directional." Size and edge are different questions.

What It Is Not

• Not a direction indicator. It measures scale, never which way.
• Not a prediction. It's a trailing average of past ranges, describing recent normal movement.
• Not comparable in raw form across markets. It only means something relative to the instrument's own price.

Limitations

ATR is a lagging measure — it's an average of past ranges, so it reacts to volatility changes with a delay and can be slow to reflect a sudden regime shift. It's also sensitive to the period chosen: a short lookback reacts fast but is noisy; a long one is smooth but sluggish. And because it's expressed in price units, it must always be interpreted relative to the market's price level to be meaningful. ATR describes volatility; it does not explain *why* volatility changed, nor does it say anything about direction.

How This Fits Into MPM

ATR is one of the volatility tools that underpins how MPM defines market conditions. When MPM expresses a distance or a volatility state, ATR is often the unit doing the work behind the scenes, chosen because it stays comparable across the different markets HIE covers.