SMT Divergence in Trading: How to Spot and Trade It

SMT divergence explained: which pairs to compare, how to read a divergence at highs and lows, entries after it and the mistakes to avoid.

LSG Club · 25 October 2026 · 14 min read

SMT divergence is a disagreement between two correlated markets at a swing point: one makes a new high or a new low, while the other fails to make it. In ICT trading the market that refuses to follow is read as the stronger hand, and the divergence is used to confirm that a sweep of liquidity may be a turning point rather than the start of a breakout. This guide from the LSG Club journal for SMT divergence trades explains which pairs to compare, how to read bullish and bearish SMT, which timeframes to use, and how to build an entry, a stop and a target around it.

SMT stands for Smart Money Technique (some traders say Smart Money Tool). The idea was popularised by ICT, Michael J. Huddleston, and it is one of the few ICT tools that compares two charts instead of one. On its own an SMT divergence is not a trade. It is a piece of evidence that makes a reversal setup more likely to work, and it is most useful when it appears at a level where you already expected the market to turn.

What is SMT in trading?

Markets that share a driver usually move together. EURUSD and GBPUSD are both priced against the US dollar, so a strong dollar tends to push both down at the same time. The S&P 500 and the Nasdaq 100 share most of the same large companies, so they rise and fall together on most days. When the shared driver moves, both charts print matching swings: a new low on one comes with a new low on the other.

SMT divergence is the moment that matching breaks. If EURUSD trades below its previous low and GBPUSD stays above its own previous low, the two markets disagree. One of them is taking the sell stops below the low; the other is already being bought. ICT’s reading is that the move that took the low was not backed by broad selling, so the low is more likely to be a stop raid than a real breakdown.

Which markets to compare for SMT divergence

SMT only works between instruments with a strong, stable relationship. Comparing two loosely related markets produces divergences every hour that mean nothing. These are the pairings most ICT traders use:

EURUSD and GBPUSD — the classic forex pair for SMT. Both are quoted against the dollar and usually move in the same direction during the London and New York sessions.
EURUSD or GBPUSD and the US Dollar Index (DXY) — an inverse relationship. When EURUSD makes a lower low, DXY should make a higher high. If DXY fails to make that higher high, the dollar is not confirming the move.
ES and NQ futures, or US500 and NAS100 CFDs — the main index pair. The S&P 500 is broader; the Nasdaq 100 is heavier in technology. When one makes a new high and the other does not, the divergence is easy to see.
ES, NQ and YM (the Dow) — some traders check all three indices and look for the one that disagrees.

For the inverse pairing, flip one chart in your head: a higher high on DXY corresponds to a lower low on EURUSD. For more on which pairs behave cleanly in ICT models, see the guide to the best forex pairs for ICT trading.

Check the correlation before you rely on it

Correlations change. A UK-specific news event can push GBPUSD on its own while EURUSD barely moves, and a single large tech company’s earnings can lift NQ while ES stays flat. On those days a divergence reflects the news, not smart money. If the two charts have not been moving together over the last few sessions, skip SMT for that day.

Bullish SMT divergence vs bearish SMT divergence

There are two kinds of SMT, one at lows and one at highs. Both compare the same swing on two charts at the same time.

Bullish SMT at swing lows

Bullish SMT forms when one market makes a lower low and the correlated market makes a higher low (or an equal low) over the same window. The market holding its low is showing relative strength. The setup is strongest when the lower low takes an obvious pool of sell stops, such as the Asian session low, the previous day’s low or a set of equal lows.

Positive correlation: EURUSD prints a lower low, GBPUSD prints a higher low.
Inverse correlation: EURUSD prints a lower low, DXY fails to print a higher high.
Indices: NQ prints a lower low, ES holds above its previous low.

Bearish SMT at swing highs

Bearish SMT is the mirror image. One market makes a higher high and the correlated market makes a lower high. The market that cannot reach a new high is showing relative weakness, and the higher high on the other chart may be a raid on buy stops before a move down.

Positive correlation: GBPUSD prints a higher high, EURUSD prints a lower high.
Inverse correlation: EURUSD prints a higher high, DXY fails to print a lower low.
Indices: ES prints a higher high, NQ prints a lower high.

In both cases the market that took the liquidity is the one that “lied”, and the market that refused is the one telling the truth about direction. That framing helps when you have to decide which chart to trade.

How to identify SMT divergence step by step

Most bad SMT reads come from comparing different swings or different time windows. A fixed routine avoids that:

Choose one correlated pairing and open both charts on the same timeframe, aligned to the same time axis.
Mark the reference swing on both charts: the same session low, the same previous day high, or the same swing point from the same hour.
Wait for price to reach that swing on at least one chart. SMT is read at a level, not in the middle of a range.
Compare: did both markets take the swing, or only one? If only one did, you have a divergence.
Check the context: is the divergence happening at a level from your higher-timeframe bias, such as a discount zone in an uptrend or a premium zone in a downtrend?
Wait for confirmation on the chart you plan to trade: a displacement away from the swing and a shift in short-term structure.
Plan the entry, the stop and the target before the order goes in.

Step 2 matters most. A divergence between “EURUSD’s low at 08:15” and “GBPUSD’s low at 09:40” is not SMT; it is two unrelated swings. The swings must sit in the same window, ideally within the same few candles.

Timeframes for SMT trading

SMT appears on every timeframe, which is both its strength and its trap. A useful way to organise it:

Daily and H4 — SMT at weekly or monthly highs and lows. Rare, but it carries the most weight and can frame the bias for days.
H1 — SMT at the previous day’s high or low or at a session extreme. A good level for intraday traders to read.
M15 and M5 — SMT inside a kill zone, at the Asian range or the London high or low. This is where most intraday traders look for it as an entry filter.
M1 — very frequent and very noisy. Use it only to fine-tune an entry that higher timeframes already support.

Timing helps as much as timeframe. SMT divergences that form during the London or New York kill zones, when liquidity is high and the session tends to set its high or low, are more meaningful than those during quiet hours. The kill zone trading guide covers those windows, and the kill zone time converter shows them in your own time zone.

SMT is confirmation, not a standalone signal

The most expensive mistake with SMT is to treat it as an entry trigger. A divergence can stay a divergence for a long time: EURUSD can keep making lower lows while GBPUSD keeps holding, until GBPUSD finally breaks too. If you buy the first divergence you see, you are buying a falling market on the strength of a comparison.

ICT treats SMT as one layer in a stack. A common sequence looks like this:

A higher-timeframe bias tells you which side you want: buys or sells.
Price runs into liquidity at a key level — a liquidity sweep of a session or daily extreme.
SMT divergence shows that the correlated market did not join the sweep.
Price displaces away from the swing and shifts short-term structure — a market structure shift or a change in the state of delivery.
The displacement leaves an entry array, such as a fair value gap or an order block, that you can use for a limit entry.

SMT sits between the sweep and the structure shift. It raises your confidence that the sweep is a reversal, but you still wait for price to prove it. This pattern often shows up as the early false move of a session, described in the guide to the ICT Judas swing.

Worked example: bullish SMT on EURUSD and GBPUSD

The prices below are illustrative, chosen to make the arithmetic easy to follow. The logic is what matters.

The context

The daily bias on EURUSD is bullish: price has been making higher lows and the previous day’s high at 1.08710 has not been taken. During the Asian session EURUSD forms a low at 1.08420, and GBPUSD forms its Asian low at 1.26750. Both lows have sell stops resting beneath them.

The divergence

At the London open EURUSD drops to 1.08385, 3.5 pips below its Asian low, and takes the sell stops. In the same fifteen minutes GBPUSD only reaches 1.26810, six pips above its own Asian low. EURUSD made a lower low; GBPUSD made a higher low. That is bullish SMT divergence, and it happened at a liquidity level during a kill zone, in the direction of the daily bias.

The confirmation

EURUSD then rallies with a strong M5 candle and closes at 1.08495, above the short-term high at 1.08480 that formed during the drop. That close is the structure shift. The displacement leaves a fair value gap between 1.08440 and 1.08465.

The plan

Entry: a buy limit at 1.08455, inside the fair value gap.
Stop loss: 1.08370, 1.5 pips below the sweep low at 1.08385. Risk = 1.08455 − 1.08370 = 8.5 pips.
Target: the previous day’s high at 1.08710. Reward = 1.08710 − 1.08455 = 25.5 pips.
Risk to reward: 25.5 ÷ 8.5 = 3.0, so the trade is planned at 3R.
Position size on a 10,000 USD account risking 1%: 100 USD ÷ (8.5 pips × 10 USD per pip per standard lot) ≈ 1.18 lots, rounded down to 1.17 lots.

If price trades back below 1.08370, the sweep low has failed, the idea is wrong and the stop takes you out for a planned loss of about 1R. Nothing about the divergence protects you from that outcome; it only improves the odds that the low holds. You can check the sizing with the position size calculator and the reward profile with the risk reward calculator.

Which chart to trade

Here the trade is on EURUSD, the market that swept the low. Its stop sits right under a clear raid, which keeps the risk tight. Some traders prefer the stronger market, GBPUSD, because it held its low and may lead the rally; the trade-off is that its stop has to go under the Asian low at 1.26750, which is usually further away. Neither choice is wrong. Pick one rule, write it in your plan and journal the results so you can compare.

Weak vs strong SMT divergence

Not every divergence deserves the same weight. These factors separate a strong SMT from a weak one:

Location — strong SMT forms at a meaningful level: previous day or week high or low, session extremes, equal highs or lows, or a higher-timeframe premium or discount zone. Weak SMT forms in the middle of a range.
Liquidity taken — strong SMT comes with an obvious stop run on one chart. If neither chart took a visible pool of stops, the divergence has little to say.
Alignment with bias — bullish SMT in a discount zone of an uptrend is far stronger than bullish SMT against a clean bearish daily trend. The premium and discount guide explains how to frame those zones.
Timing — a divergence inside a London or New York kill zone outweighs one in the late afternoon or during the Asian lull.
Size of the disagreement — a clear higher low on one chart against a clear lower low on the other is more convincing than a fraction of a pip of difference, which may simply be spread or data noise.
Follow-through — strong SMT is followed quickly by displacement and a structure shift. If price drifts sideways for hours after the divergence, the signal is fading.

Common SMT divergence mistakes

Comparing different swings — the swings must be from the same window on both charts. Matching a morning low on one chart with an afternoon low on the other produces false divergences.
Using uncorrelated markets — SMT between EURUSD and USDJPY, or between gold and an index, tells you little.
Trading the divergence itself — entering the moment you see SMT, before any displacement or structure shift, means catching a move that may not have finished.
Ignoring news — a currency-specific release can break correlation for a few hours. Check the calendar before relying on SMT around high-impact events.
Reading noise as signal — differences of a fraction of a pip, or divergences that only exist on M1, are often just feed differences between instruments.
Fighting the higher-timeframe trend — counter-trend SMT works far less often than SMT in the direction of the bias.
Moving the stop after entry — the stop belongs beyond the swing the divergence formed at. If that swing breaks, the reason for the trade is gone.

Trading involves risk, and no confirmation tool removes it. Practise reading SMT on a demo account and in your journal before you put real money behind it.

SMT divergence trading checklist

Higher-timeframe bias is clear, and the divergence points the same way.
The two markets have been moving together over recent sessions, with no market-specific news distorting them.
The divergence forms at a meaningful swing: session extreme, previous day or week high or low, or equal highs or lows.
One chart clearly took liquidity; the other clearly did not.
The divergence forms inside a kill zone or another active session window.
Price displaced away from the swing and shifted short-term structure on the chart you trade.
There is a defined entry array (fair value gap or order block), a stop beyond the swing and a target at the next liquidity pool.
The planned reward is at least 2R and the position size matches your risk per trade.

Record which items each SMT trade met and its result in R. After thirty or forty trades you will know which conditions actually make the difference for you. An ICT trading journal with tags and real charts makes that comparison quick.

FAQ

What is SMT divergence?

SMT divergence is a disagreement between two correlated markets at a swing point: one makes a new high or low and the other does not. ICT traders use it to confirm that a sweep of liquidity may lead to a reversal.

What does SMT mean in trading?

SMT stands for Smart Money Technique, a term popularised by ICT (Michael J. Huddleston). It describes comparing correlated instruments to spot when one of them fails to confirm the other’s move.

Which pairs are best for SMT trading?

The most common pairings are EURUSD with GBPUSD, EURUSD or GBPUSD with the inverse US Dollar Index, and ES with NQ (or US500 with NAS100). The pairing must be strongly correlated during the session you trade.

Is SMT divergence a buy or sell signal on its own?

No. SMT is a confirmation tool. Most traders wait for a displacement and a market structure shift after the divergence, then enter from a fair value gap or order block with a stop beyond the swing.

What timeframe is best for SMT divergence?

Intraday traders usually read SMT on M15 or M5 at session extremes and confirm it with H1 context. SMT on H4 or daily charts is rarer but carries more weight.