What does SMT stand for in trading?

Smart Money Technique, an ICT term. The technique is comparing correlated instruments at their swing points: when one makes a new extreme and the other refuses to confirm it, the divergence is read as smart money running stops on one chart without committing across the whole complex. It has nothing to do with oscillator divergence, which compares price against an indicator.

Which pairs work for SMT divergence?

Anything with a strong, current relationship: equity index futures such as ES, NQ, and YM; EURUSD against GBPUSD; gold against silver. Inversely correlated references like the dollar index work too, but the logic mirrors: a new high in one should pair with a new low in the other, and the divergence is that mirror breaking. Always verify the correlation is currently tight.

Is SMT divergence enough to trade on its own?

No. Correlated instruments diverge briefly all the time, and correlations themselves drift. Most models use SMT strictly as confluence: it strengthens a setup that already has a swept level, a sensible time window, and a structure break, and it gets ignored entirely when the pair has not been tracking tightly. Treat it as a weighting factor, not a trigger.

Concept

Smart Money Technique Divergence

Smart Money Technique Divergence, also known as SMT, correlated-pair divergence, crack in correlation, is a Smart Money Concepts / ICT concept. The Library holds 1 implementation, a working definition you can pull into Quant.

Top Smart Money Technique Divergence indicator

The top custom implementation, built on the original standard Smart Money Technique Divergence formula.

1 total

Want to trade Smart Money Technique Divergence? The implementation below is one prompt away from a backtested strategy in Quant.

SMT Divergences

Indicator

What is Smart Money Technique Divergence?

Smart Money Technique (SMT) divergence is a non-confirmation between two correlated instruments at matching swing points. ES and NQ, or EURUSD and GBPUSD, normally register new highs and lows together; a bearish SMT prints when one takes out its prior high while the other fails to, and a bullish SMT is the mirror: one runs its prior low while the other holds above its own. ICT vocabulary calls the failure a crack in correlation.

The read behind it: genuine institutional interest should express across the whole correlated complex at once, so when only one instrument makes the new extreme, the move looks less like broad participation and more like a targeted run on the stops resting beyond that swing. The divergence proves nothing on its own (correlated instruments desynchronize briefly all the time, and correlation regimes drift), so the read carries weight only when the pair is currently tracking tightly and the non-confirmation forms at a level and time that already matter. Most models treat SMT as a confirmation layer, not a trigger.

How to identify an SMT divergence

SMT is read off two charts at once: same timeframe, correlated instruments, compared swing by swing.

1Choose a pair with a strong current relationship (index futures like ES and NQ, EURUSD and GBPUSD, gold and silver) or an inverse reference like the dollar index, remembering the logic flips when the correlation is inverse.
2Mark the swing highs and lows that formed in the same window of time on both charts; only corresponding swings are comparable.
3Compare the latest test: bearish SMT if one instrument trades above its prior high while the other stays below its own; bullish SMT if one trades below its prior low while the other holds.
4Weight the read by context: it means more during a killzone at a level of interest, and less when the pair's correlation has visibly been drifting.

How it's calculated

A boolean divergence condition between two correlated instruments: one prints a new swing extreme while the other fails to confirm it.

\text{Bearish SMT: } \mathrm{HA}_2 > \mathrm{HA}_1 \text{ and } \mathrm{HB}_2 < \mathrm{HB}_1

\text{Bullish SMT: } \mathrm{LA}_2 < \mathrm{LA}_1 \text{ and } \mathrm{LB}_2 > \mathrm{LB}_1

\text{Inverse pair, bearish SMT: } \mathrm{HA}_2 > \mathrm{HA}_1 \text{ and } \mathrm{LB}_2 > \mathrm{LB}_1

\text{Inverse pair, bullish SMT: } \mathrm{LA}_2 < \mathrm{LA}_1 \text{ and } \mathrm{HB}_2 < \mathrm{HB}_1

A: first instrument of a correlated pair (e.g. ES)

B: second instrument of the pair (e.g. NQ, or DXY in the inverse case)

HA_1: swing high of A at the earlier swing point

HA_2: swing high of A at the later swing point

HB_1: swing high of B at the earlier swing point

HB_2: swing high of B at the later swing point

LA_1: swing low of A at the earlier swing point

LA_2: swing low of A at the later swing point

LB_1: swing low of B at the earlier swing point

LB_2: swing low of B at the later swing point

The two swings must form near the same times on both charts, typically inside the same session or higher timeframe window.

Classic pairs are ES with NQ and EURUSD with GBPUSD; EURUSD against DXY is the usual inverse example.

Some definitions accept an equal extreme (a failure to exceed) instead of a strictly lower high or higher low.

How traders use it

As sweep confirmation: when one instrument runs a resting pool of equal highs and its correlated partner refuses to follow, the failure supports reading the move as a liquidity sweep rather than genuine strength.
As a precondition inside session models: SMT at a higher-timeframe level during the chosen window is the cue to start hunting for displacement and a structure break in the anticipated direction; the entry still comes from that follow-through.
As the non-confirmation leg of a smart money reversal: the divergence marks which swing is suspect and which instrument is likely leading.
As a stand-down filter: when the pair has decoupled, SMT reads are ignored entirely; the technique assumes a correlation that actually exists right now.

SMT divergence vs other divergence reads

Regular Bullish/bearish Divergence: Regular divergence compares price against an oscillator on a single instrument. SMT compares price against price across two instruments (no indicator involved), and it exists only at corresponding swing points.

Intermarket Analysis: Intermarket analysis studies broad relationships across asset classes for macro context over weeks and months. SMT is a narrow, swing-by-swing non-confirmation read on one tightly correlated pair, used for trade timing.

Relative Strength Comparative: Comparative relative strength tracks continuous out- or under-performance against a benchmark, usually as a ratio line. SMT is a discrete event at matched swings; it says nothing about the stretches in between.

Concept family

Smart Money Concepts / ICT

54 concepts mapped · 54 in the Library