Order Blocks

A rule-based definition — the last opposing candle before a displacement move — for a concept that is usually defined only after the fact.

What it is

An order block, in smart-money-concepts terminology, is a candle (or the price zone it covers) said to mark where institutions accumulated positions just before a strong move, with the expectation that the zone will matter again if price returns to it. Honesty first: there is no verifiable public evidence that institutions systematically leave retrievable 'blocks' of orders at these candles. Institutional execution is deliberately fragmented across venues and time precisely to avoid leaving footprints, and the story as usually told is unfalsifiable — any zone that holds 'proves' it and any zone that fails is explained away. What can be done honestly is to strip the narrative and keep a testable pattern: define an order block objectively as the last opposing candle immediately before a displacement candle — for example, the final down candle before an up candle whose body exceeds a stated multiple of ATR. That definition is measurable and reproducible, which means its retest behavior can actually be studied rather than asserted. The marks it produces describe past price arrangements only; nothing about the definition predicts future behavior. This is educational material, not financial advice, and trading involves risk of loss.

How it works

The rule has two parts. First, displacement: a candle counts as a displacement candle when its body (the absolute distance from open to close) exceeds a chosen multiple of the Average True Range — say 1.5x ATR(14) — which normalizes 'unusually strong' across instruments and volatility regimes instead of relying on eyeballing. Second, the opposing candle: a bullish order block is the down candle immediately preceding an up displacement candle, and a bearish order block is the up candle immediately preceding a down displacement candle. The zone is then defined from that opposing candle — its full high-to-low range, or the body only; both variants exist and the choice must be stated because they produce different zones. A retest (often called 'mitigation') is simply price later trading back into the zone, which is equally codeable. The parameters — ATR length, displacement multiple, zone variant — control everything: a stricter multiple marks fewer, larger events; a looser one floods the chart. Because every piece is explicit, you can count how many zones a rule produces, measure how price behaved on return visits across a large sample, and test claims about order blocks instead of curating flattering examples in hindsight.

How traders read it

• A marked zone is a candidate area produced by a stated rule — treat it as a hypothesis about where reactions might be observed, never as a promise that price will respect it.
• Check which zone variant the chart uses (full candle range versus body only) — the same rule name with different variants marks different prices.
• The displacement threshold drives everything: at low multiples nearly every candle qualifies as 'displacement', which quietly manufactures zones everywhere.
• A retest of a zone is an observation, not a signal — price slices straight through marked zones routinely, especially in trending markets.
• The common claim that 'fresh' (untested) zones behave differently from already-retested ones is a testable hypothesis, not an established fact — a rule-based definition lets you check it on your market.
• Read zones only from confirmed bars; a displacement candle is not a displacement candle until it closes.

Common settings

A typical starting point is ATR length 14 with a displacement body threshold of 1.5-2.0x ATR, and the zone taken from the opposing candle's high and low. Body-only zones are a stricter variant. Lower thresholds mark far more zones; there is no correct value, only stated ones whose consequences you have measured.

Strengths

• Converts a vague, narrative-driven concept into a falsifiable rule that can be tested on historical data
• ATR-normalized displacement makes 'strong move' consistent across instruments, timeframes, and volatility regimes
• Every parameter is explicit, so the zones are reproducible and auditable rather than curated by eye
• Simple to compute — two candles and an ATR comparison — and easy to combine with structure analysis as an organizing framework

Pitfalls to watch

• The institutional narrative is unverified: naming a candle an 'order block' does not make it special, and no public order-book evidence supports the claim that these zones hold resting institutional orders.
• Hindsight bias dominates the teaching material — examples are almost always selected after the outcome is known, which makes any pattern look reliable.
• Parameter sensitivity: changing the ATR length, displacement multiple, or zone variant redraws every block on the chart; a compelling chart is often one flattering setting.
• In trending markets, price runs through most counter-trend zones; the failures are rarely shown in promotional material.
• Some published order-block scripts delete or move zones once price violates them, leaving only the survivors visible — a survivorship-biased picture of how the tool behaves live.
• No predictive power: the rule marks a past price arrangement; it says nothing about what happens on a return visit, and trading involves risk of loss.

Pine v6 example

//@version=6
indicator("Rule-Based Order Block Marker", overlay = true)

atrLen   = input.int(14, "ATR Length", minval = 1)
dispMult = input.float(1.5, "Displacement Body (x ATR)", minval = 0.5, step = 0.1)

atrVal = ta.atr(atrLen)
body   = math.abs(close - open)

// Displacement: a candle whose body exceeds dispMult x ATR
dispUp = close > open and body > dispMult * atrVal
dispDn = close < open and body > dispMult * atrVal

// Falsifiable rule: the LAST OPPOSING candle immediately before displacement.
// Bullish OB = down candle followed by an up displacement candle; mirrored for bearish.
bullOB = dispUp and close[1] < open[1] and barstate.isconfirmed
bearOB = dispDn and close[1] > open[1] and barstate.isconfirmed

// The shape marks the prior (opposing) candle via a display offset
plotshape(bullOB, "Bullish OB (prior bar)", style = shape.triangleup,   location = location.belowbar, color = color.teal,   offset = -1)
plotshape(bearOB, "Bearish OB (prior bar)", style = shape.triangledown, location = location.abovebar, color = color.maroon, offset = -1)

Pro tip: If you want to know whether order blocks mean anything on your market, do the unglamorous thing: fix the rule, mark every zone it produces — including the ones that fail — and measure behavior on return visits across hundreds of cases. A definition you cannot write as code is a definition you cannot test, and a zone that is only identified after price reacted to it is a story, not a tool. Educational context only, not financial advice; no pattern predicts price, and all trading carries risk of loss.

Educational only — not financial advice, not a recommendation to trade. No indicator is predictive; trading involves substantial risk of loss.