🏛️ Giza Gold EA launch: $199 for the first 10 copies, then $499. Get it on MQL5 | Launch post | Watch Giza trade in public, every trade on the gold chart: thegoldheist.com
The short answer
SMT divergence is not a stand-alone signal on gold and silver: in our tests it did about as well as a normal breakout. The most promising place was Bitcoin and Ethereum on the 4-hour chart, where the bearish SMT was right 54.6% of the time (861 signals).
The idea
Gold and silver usually move together, and so do Bitcoin and Ethereum. An SMT divergence (an ICT term) is when one of the two makes a new high and the other one doesn't. ICT traders read this as a warning that the price is about to turn down. The same works at the lows: one makes a new low, the other doesn't, so the price may turn up.
How we tested it
We used gold and silver (from mid-2014) and Bitcoin and Ethereum (2018 to 2026), on 1-hour, 4-hour and daily charts, with different settings: 216 tests in total. For every SMT signal we measured what the price did next, and compared it with normal breakouts, where both markets made the new high together. If SMT works, the price must move more in the predicted direction after an SMT signal than after a normal breakout.
What happened
- Gold and silver: on average the price moved 0.04% better after an SMT signal than after a normal breakout. Not one of the 108 tests showed more than luck would explain (luck alone would give about 5)
- Bitcoin and Ethereum: on average 0.01% worse than a normal breakout. 9 of the 108 tests looked better than luck, against about 5 you would expect from luck alone; 7 of those 9 were "bearish" signals on the 4-hour chart
- Overall, the signals were right 44% to 54% of the time
The best single result was the bearish SMT on Bitcoin and Ethereum on the 4-hour chart. After it, the price moved about 0.5% more in the predicted direction than after a normal breakout over the next 3 candles, and it was right 54.6% of the time (861 signals). That one deserves a closer look on crypto.
What it means for you
- On gold and silver, use SMT together with other context (a key level, a killzone, the higher-timeframe direction), not alone
- A signal that is right 50% to 54% of the time is not a reason to enter a trade on its own. It needs good exits and low costs to make money
- When someone shows you a few winning results out of many tests, ask how many would win by luck alone. 9 out of 108 is not far from 5
- When you test an idea yourself, keep a part of the history aside and look at it only once, at the very end
Our free indicator ICT SMT Divergence Honest Stats marks every SMT on your chart and runs this exact comparison against normal breakouts, for any pair you choose: https://www.mql5.com/en/market/product/201470 (MT4 version: https://www.mql5.com/en/market/product/201471)
Details for those who want them
- Data: Dukascopy gold and silver hourly from mid-2014; BTC and ETH 2018 to 2026
- New highs and lows over the last 10, 20 or 50 candles; results measured over several following periods; each test also split into two halves of the history
- Gold and silver: the average difference was +0.038%, positive in 52% of tests, right in both halves in 25%. BTC and ETH: -0.010%, positive in 46% of tests
Have a strategy, an indicator or an EA you want tested like this? Write it in the comments.
If one of our free indicators or tools helped you, a short review on its Market page helps other traders find it.
Next week: trend following on 27 CFD markets, and the cost nobody puts in the backtest.
