Smart Money Concepts (SMC)

A modern vocabulary for price action — swing structure, zones, and gaps — that largely relabels ideas classical technical analysis has described for decades.

What it is

Smart Money Concepts (SMC) is a school of price-action analysis, popularized online through ICT (Inner Circle Trader) material, built around the narrative that large institutions ('smart money') leave readable footprints on a chart. Its vocabulary includes Break of Structure (BOS), Change of Character (CHoCH), order blocks, fair value gaps, liquidity sweeps, and premium/discount zones. Read honestly, most of these terms map closely onto classical technical analysis: a BOS is a breakout past a prior swing point, a CHoCH resembles Wyckoff-style trend-change structure and Wilder's failure swings, order blocks are a variant of supply-and-demand zones, fair value gaps are price imbalances or gaps, and a liquidity sweep is what older literature calls a stop run or false breakout. That mapping does not make SMC useless — the framework organizes swing structure systematically, and several of its patterns have fully deterministic definitions that can be coded and tested. But the institutional-intent narrative attached to the labels is unverified: chart patterns do not reveal who traded or why. SMC is a descriptive vocabulary, not a predictive system. This overview is educational only, not financial advice, and all trading involves risk of loss.

How it works

SMC analysis starts from swing structure: identifying swing highs and lows, typically via pivots (a bar whose high or low exceeds its neighbours on both sides). In an uptrend defined as higher highs and higher lows, a close above the most recent confirmed swing high is labeled a bullish BOS — trend continuation. A close through the most recent higher low is labeled a CHoCH — the first structural evidence the trend may be changing. From that skeleton, SMC layers in zones: order blocks (commonly defined as the last opposing candle before a strong displacement move), fair value gaps (three-candle imbalances), liquidity pools (clusters of highs or lows where stop orders plausibly rest), and premium/discount (upper and lower halves of a dealing range measured from swing low to swing high). Each element can be given a mechanical definition, which is what makes SMC testable when done rigorously. The critical implementation detail is confirmation lag: a pivot-based swing point only confirms several bars after it forms, so structure labels drawn at the pivot bar describe the past. Any backtest that reads a BOS or CHoCH before its underlying pivot was confirmable is peeking into the future, which is a common flaw in SMC scripts.

How traders read it

• BOS (Break of Structure): price closing beyond the most recent confirmed swing point in the direction of the prevailing trend — classical TA calls this a breakout to a new swing high or low.
• CHoCH (Change of Character): price closing through the most recent swing point against the prevailing trend — comparable to a failure swing or the first lower low after an uptrend.
• Order blocks: the last opposing candle before a displacement move, watched as a zone price may revisit — a rule-based cousin of classical supply and demand zones.
• Liquidity sweeps: a wick through an obvious prior high or low followed by reversal — long known as a stop run or false breakout; the SMC name adds narrative, not new mechanics.
• Premium/discount: position within the current dealing range, above or below its 50% midpoint — a range-position measure, similar in spirit to where price sits between support and resistance.
• All of these are descriptions of completed price behaviour; none is a signal with established predictive power, and structure labels are only trustworthy after their pivots confirm.

Common settings

SMC itself has no canonical parameters, which is precisely why definitions vary between educators and scripts. Coded implementations typically expose a pivot strength (bars each side of a swing point, commonly 3-10), a choice between wick-based and close-based breaks for BOS/CHoCH, a displacement threshold for order blocks, and the timeframe hierarchy used (many practitioners read structure on a higher timeframe and entries on a lower one). Because every one of these choices changes which structures get labeled, any two 'SMC indicators' can disagree on the same chart while both claiming to be correct.

Strengths

• Systematic treatment of swing structure: the BOS/CHoCH framework forces explicit definitions of trend and trend change rather than vague eyeballing.
• Several components (FVGs, pivot-based structure, range midpoints) are fully deterministic and therefore codeable and falsifiable.
• The vocabulary is internally consistent once defined, which makes charts annotated by rule-following tools reproducible.
• Learning SMC alongside its classical equivalents deepens understanding of both — the mapping itself is instructive.
• Popularity means abundant material and tooling, so rule-based versions are easy to compare against each other.

Pitfalls to watch

• The institutional narrative is unverifiable: candles do not record who traded, and no public evidence establishes that 'smart money' leaves the specific footprints SMC claims.
• Definitions vary by educator: without pinning down pivot strength and break rules, BOS and CHoCH become retrospective stories that fit any outcome.
• Confirmation lag and lookahead: pivots confirm bars after they occur, so naive scripts and backtests that label structure at the pivot bar are peeking into the future.
• Hindsight selection: annotated examples in SMC content are chosen after the fact; unannotated charts contain many failed order blocks, unfilled gaps, and broken structures.
• Relabeled concepts can be double-counted: treating an order block, a demand zone, and a support level as three independent confirmations is counting one observation three times.
• No predictive power is established for any component: these are descriptive tools, and trading them carries the full risk of loss.

Pine v6 example

//@version=6
indicator("SMC Structure Example (BOS)", overlay = true)

pivotLen = input.int(5, "Pivot strength (bars each side)", minval = 1)

// Swing points confirm pivotLen bars AFTER they occur — no lookahead
swingHigh = ta.pivothigh(high, pivotLen, pivotLen)
swingLow  = ta.pivotlow(low, pivotLen, pivotLen)

var float lastSwingHigh = na
var float lastSwingLow  = na
if not na(swingHigh)
    lastSwingHigh := swingHigh
if not na(swingLow)
    lastSwingLow := swingLow

// Close-based break of the last CONFIRMED swing (SMC: BOS / classical: breakout)
breakUp   = not na(lastSwingHigh) and ta.crossover(close, lastSwingHigh)
breakDown = not na(lastSwingLow) and ta.crossunder(close, lastSwingLow)

plot(lastSwingHigh, "Last confirmed swing high", color = color.teal, style = plot.style_linebr)
plot(lastSwingLow,  "Last confirmed swing low",  color = color.maroon, style = plot.style_linebr)
plotshape(breakUp and barstate.isconfirmed, "Break above swing high", style = shape.triangleup, location = location.belowbar, color = color.teal, size = size.tiny)
plotshape(breakDown and barstate.isconfirmed, "Break below swing low", style = shape.triangledown, location = location.abovebar, color = color.maroon, size = size.tiny)

Pro tip: Translate every SMC term you use into its classical equivalent and into a line of code. If a concept survives both translations — you can name what old-school TA called it and write an unambiguous rule for it — it is testable and worth studying. If it only survives as a story about what institutions 'must' be doing, treat it as narrative, not analysis. Either way, structure labels describe the past: educational context only, no framework predicts price, and all trading involves risk of loss.

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