SMC Setup 4 finds three-drive patterns on MT4 and MT5, and then refuses to signal on most of them unless the third drive lands inside a level that already existed before the first drive started.
Two independent filters, and the first one is a test almost nothing else in this space applies.
The drives have to be symmetric in time
The three pivots are labelled 1, 2 and 3 on the chart. Getting labelled at all requires two things.
Price extension. Each leg has to be longer than the one before it, measured as a ratio. Drive two has to extend past drive one by more than the leg it came from, and drive three past drive two the same way. Three pushes of decreasing size don't qualify; the pattern has to be accelerating.
Time symmetry. This is the unusual one. The tool measures the full span from drive one to drive three and requires the middle drive to fall between 37.5% and 62.5% of that span. The middle quarter, near enough.
Think about what that throws out. A fast first leg followed by a long grinding second leg produces three pivots at three prices that satisfy every price rule and look like a three-drive pattern in a screenshot. It isn't one. The market that made it was doing two different things, and the rhythm you can see by eye is exactly what this test encodes.
I've not seen another indicator measure it, and it costs one line.
The PD array gate
By default the pattern alone isn't enough. The range of the final drive has to overlap a premium or discount array, and you choose which kind:
| Setting | What it requires |
|---|---|
OB (default) |
The final drive overlaps an order block |
FVG |
It overlaps a fair value gap |
OB and FVG |
Either one |
NONE |
No array needed, the pattern alone signals |
Order blocks come from a slow 50-bar pivot: at each confirmed pivot the tool walks back up to three bars for the last opposite-colour candle and stores its high and low. That block is live until price closes through it, and it's marked as tested once price has entered it and closed back out. Fair value gaps are ordinary three-bar gaps, filtered by ATR.
The gap filter has one condition worth knowing: the gap has to have formed before drive one. A gap created during the pattern doesn't count. The point of the array is that it's a level the market already respected, not something the pattern made on its way past.
What you get
Small 1, 2, 3 labels at the drives and a Wingdings arrow at the third, on the bar it confirmed. The alert names the reason, so you get bear setup inside bullish FVG rather than a generic signal, and you know which filter passed.
Order blocks and gaps can be drawn with Show all OBs and Show all FVGs, both off by default. Turn them on once to see what the tool is matching against, then turn them off; the chart gets busy quickly with a 50-bar pivot behind the blocks.
What it doesn't do
No entry, stop or target. It marks where a three-drive completed inside a level. Where you'd trade that is your business.
No indicator buffers. Everything is a chart object, so iCustom gets you nothing. Alerts are the only route out.
No higher-timeframe read. Drives and arrays are all measured on the chart timeframe.
Where it's weak
The default pivot period is 2. That's much shorter than the other tools in this family and it means the drives are found from very local swings. It's the right default for finding harmonic-style patterns and it's the wrong one if you're expecting major structure. Raise it and you'll get far fewer, much larger patterns.
Order blocks lag by 50 bars. A 50-bar pivot can't be confirmed until 50 bars have passed, so a block only enters the pool long after the candle that made it. Blocks are correctly placed in history; they're just not available early.
Time symmetry is measured on bar timestamps. Over a weekend gap or a session break the span includes hours the market wasn't open, so patterns spanning a weekend are judged against a stretched clock.
The ratio tests use raw division. Legs of nearly equal size produce ratios hovering around 1, so pattern detection near the threshold is sensitive in a way that isn't tunable.
It's quiet with the default array filter on. Three drives, symmetric in time, accelerating in price, landing inside a pre-existing block. Set PDArray type to NONE if you want to see how many raw patterns are being filtered out. It's usually most of them.
Questions you'll have
Does SMC Setup 4 repaint? No. Patterns are confirmed on closed bars behind a new-bar check, and the arrow is placed on the bar that confirmed. Order blocks are removed when price closes through them, which is removal rather than redrawing.
What do the 1, 2 and 3 labels mean? The three drives, oldest to newest. The signal arrow sits above or below drive three.
Why do I get so few signals? The PD array filter. The final drive has to overlap an order block or a gap that already existed. Try NONE to see the raw pattern count first, then decide.
What is the time symmetry test? The middle drive has to fall between 37.5% and 62.5% of the total span from drive one to drive three. It rejects patterns whose legs took very different amounts of time.
Why don't I see any order blocks on the chart? Show all OBs is off by default. Turn it on to see the pool the filter is matching against.
Can an EA read the signals? No. The indicator publishes no buffers; everything is a chart object. Use the terminal, email or Telegram alerts.
Best timeframe? H1 and H4 with the default pivot period of 2. On M5 raise it, or the drives are found inside noise.
MT4? Both platforms are included, one download each. Same symmetry test, same array filter, same alerts.
Support
Message me through the site with a screenshot, the symbol and the timeframe, and I can tell you which of the two filters a candidate failed. Updates are included for as long as your subscription runs.
Analysis tool. It places no trades and makes no profitability claim. Past price behaviour is not a guide to future price behaviour.