SMT divergence
Lesson 12 · 4 min read · By DT Terminal
Use a correlated market as a second opinion at a swing that already matters, never as an entry trigger.
Unlock the full lesson in the app
Worked examples, the quiz and the interactive tools are in the app. Free account.
Two markets, one swing
Some markets usually move together: EURUSD and GBPUSD, or the S&P 500 and Nasdaq 100 futures (ES and NQ). SMT divergence, which ICT expands as "smart money technique", is when two of them disagree at the same swing.
- Bullish SMT, at lows: one market makes a lower low, the other makes a higher low.
- Bearish SMT, at highs: one market makes a higher high, the other a lower high.
ICT reads the new extreme as a stop run, a sweep in the sense of Liquidity the ICT way, not a move the whole group made. What you can check is simpler: which market took its old swing and which did not.
Which markets to pair
Pair markets that share a driver, and compare the same swing on the same timeframe and the same bars.
| Pair | Why they move together | How to compare |
|---|---|---|
EURUSD and GBPUSD | Both priced against the US dollar | Same direction: lows with lows |
EURUSD and DXY | DXY is the dollar index | Inverse: a EURUSD low against a DXY high |
ES, NQ and YM | US stock index futures | Same direction: highs with highs |
Line the two charts up on the same timeframe and the same clock, and mark the swing on each before you compare. Decide how far apart two swings may print and still count as the same one, say 2 bars, and write that down too. Otherwise you will find an SMT on every chart you look at.
Bullish and bearish SMT
Read each pair of cells side by side. Same bars, same swing. The top row disagrees; the bottom row agrees, and that is most days.
SMT divergence
Two correlated markets at the same swing
- The old swing
EURUSD
takes the old low
GBPUSD
holds above it
US500
takes the old high
NAS100
holds below it
SMT
one market takes the old extreme, the other does not
No SMT
both take it together
SMT on the chart: at the lows
A sample case, with both markets stacked over the same bars, New York time.
Worked example
Example 01: EURUSD and GBPUSD M15: a bullish SMT at the lows
The worked example, with every number walked through, is in the app.
SMT on the chart: at the highs
The bearish version, on a sample index pair, New York time.
Worked example
Example 02: US500 and NAS100 M5: a bearish SMT at the highs
The worked example, with every number walked through, is in the app.
A second opinion, not a trigger
Correlated markets drift apart briefly all the time. On its own an SMT is common and cheap, so the mistake is to trade every one. Treat it as a confirmation layer at a level and a time that already matter: a known pool of liquidity, inside a window you trade.
Written as a rule: at a marked old low, inside your session window, the second market must hold above its matching low on the same bar. If your model has no level and no window yet, build those first.
Test whether SMT adds anything
Before SMT goes into your rules, measure it on your own markets: record every case, then compare.
Replay both markets in one backtester session
EURUSD and GBPUSD to the same backtester session and replay a month on M15. Each time one pair takes a marked low, check the other on the same bar and write SMT or no SMT before you play on.Tag it, then compare
SMT. After a few dozen, filter the Trade Log by the tag and compare those trades with the rest of the same setup, in R.Check yourself
Quick review · 0 of 0