Key takeaways
- SMT divergence is a disagreement between correlated assets: one makes a new high or low, the other does not.
- Bullish SMT appears at lows, bearish SMT at highs. It shows that one market’s move is not confirmed by the other.
- It is only worth looking for at significant liquidity and during active hours; elsewhere divergences are random.
- SMT is a filter and a reversal confirmation, not an entry. Decisions still come from your model and risk rules.
What is SMT divergence
SMT stands for Smart Money Technique (sometimes Smart Money Tool). It is an ICT concept based on correlation: many markets move almost in sync because they depend on the same money and the same news. The Nasdaq and S&P 500 indices, EURUSD and GBPUSD, gold and silver.
SMT divergence occurs when that sync breaks at a key point. For example, the Nasdaq sweeps the previous low while the S&P 500 forms a higher low at the same moment. One market shows continuation lower, the other does not. ICT calls this a “crack in correlation”.
Why the divergence happens
Large capital does not build positions in every instrument at the same time and in the same way. If selling has weakened in one market while the other is still being pushed down to the liquidity below its lows, the charts diverge. The asset that fails to make a new extreme shows the real strength: sellers there can no longer extend the move.
That is why SMT is most valuable where liquidity sits. A low swept in one asset but not in the other often means the purpose of the move was to run stops, not to continue the trend.
Bullish and bearish SMT
- Bullish SMT forms at lows. One asset makes a new low while the correlated one makes a higher low. It hints that sellers are weakening and a reversal up is possible.
- Bearish SMT forms at highs. One asset makes a new high while the correlated one does not. It hints that buyers are weakening and a reversal down is possible.
Which assets to compare
SMT only works with pairs that have a stable correlation. With inverse correlation the logic is mirrored: the low of one asset is compared with the high of the other.
| Pair | Correlation | Notes |
|---|---|---|
| NQ and ES | Positive | The most popular pair for index futures; YM is often added |
| EURUSD and GBPUSD | Positive | The classic forex pair for the London and New York sessions |
| EURUSD and DXY | Inverse | The dollar index moves opposite to the euro |
| Gold and silver | Positive | XAUUSD and XAGUSD, as well as GC and SI futures |
| BTC and ETH | Positive | Works, but correlation in crypto is less stable |
Where and when to look for SMT
Assets diverge all the time. Only divergences in the right place and at the right time matter:
- At significant liquidityPrevious day, week and session highs and lows are where stops accumulate. SMT when such a level is swept matters far more than at a random local extreme.
- On higher cyclesSMT between weekly, monthly or quarterly candles helps with direction over a larger period. On lower timeframes it pinpoints the moment of reversal.
- During active hoursThe London and New York opens and ICT macro windows are when the algorithm seeks liquidity. A divergence then is more informative than in quiet hours.
- In the direction of contextA bullish SMT that agrees with a bullish higher-timeframe context is a strong argument. Against the context, it is only a reason to be more careful.
How to compare charts correctly
Most SMT mistakes are technical. To keep the comparison fair:
- Compare identical candles. Same timeframe, same close time and same time zone on both charts.
- Compare matching extremes. Low with low, high with high, taken over the same time window.
- Mind the contracts. On futures, make sure both charts use the front contract, especially during rollover weeks.
- Remember inverse correlation. With DXY, bullish SMT for the euro looks like a dollar high that failed to break.
Doing this by hand takes time, so I built indicators for myself: SMT LTF and SMT HTF for TradingView pick the correlated asset automatically and mark divergences by cycle, from yearly and quarterly down to intraday windows.
Common mistakes
- Entering on SMT alone. A divergence shows a move is weakening, but not where or when to act. You need an entry model and a predefined risk.
- Looking for SMT anywhere on the chart. A divergence in the middle of a range with no liquidity sweep means almost nothing.
- Using pairs without stable correlation. If the assets diverge often anyway, the “divergences” will be random.
- Comparing different times. An NQ low at 9:35 and an ES low at 10:10 is not SMT.
- Forgetting context. A bullish SMT on the 1-minute chart against a clear daily downtrend is a weak argument.
Video breakdown
In this video I walk through SMT divergence on a real chart and show how to build it into a trading strategy as a confirmation. The video is in Russian; YouTube auto-translated subtitles are available.
More on the topic: SMT as entry confirmation and an overview of the SMT, AlgoTime and Liquidity indicators.
FAQ
What does SMT stand for?
SMT stands for Smart Money Technique. The term was introduced by ICT (Inner Circle Trader) for divergence between correlated assets. You may also see Smart Money Tool — the meaning is the same.
How is SMT different from a regular oscillator divergence?
A classic divergence compares price with an indicator such as RSI or MACD. SMT compares price with price: two real markets that usually move together. It does not depend on indicator settings and shows how the assets themselves behave.
Which pair works best for SMT?
For index futures, NQ and ES are the most common, sometimes with YM added. In forex, EURUSD and GBPUSD, or EURUSD against the inversely correlated DXY. What matters is stable correlation and identical candle times.
Which timeframe should I use for SMT divergence?
Any, but the weight differs. SMT between weekly and daily candles helps with direction. Intraday, it pinpoints the reversal at liquidity. It works best when a lower-timeframe SMT agrees with the higher-timeframe context.
Can SMT be used in crypto?
Yes, for example between BTC and ETH. But crypto correlation is less stable than for indices or currency pairs, so there are more false divergences. Test your rules on the history of the specific pair.
Want to see this on a live chart?
On YouTube I show these concepts in real time, and on Telegram I post new breakdowns and materials. If you want the whole system, there is one-on-one mentorship.
This material is educational and is not investment advice or a trading signal. The diagrams are teaching illustrations, not real trades. Trading financial markets involves high risk; test any idea on historical data and a demo account before using it.