How it works
Know Sure Thing rests on a premise its author argued at length: markets move as the sum of several cycles of different lengths, so any momentum indicator built on a single lookback is measuring one cycle and ignoring the rest. KST measures four horizons at once. Each is a rate of change over a different window, each is smoothed with its own moving average, and the four smoothed series are added together with progressively larger weights so the longer cycles dominate.
That weighting is the design's core decision. If the four components were weighted equally the result would be dominated by the noisiest, shortest one. By giving the longest window four times the weight of the shortest, KST becomes a slow, deliberate line whose direction reflects the longer cycles while the shorter ones supply timing detail at the margin. The output looks like a smoother, better-behaved MACD, and it is read in almost exactly the same way.
Three readings do the work. The signal-line crossover, where KST crosses a nine-period average of itself, is the primary trigger. The zero-line crossover marks the point where blended momentum across all four horizons flips sign, which is a slower and higher-conviction regime change. And divergence against price at a swing extreme is read conventionally, with the usual reliability caveats — though because KST is heavily smoothed, its divergences are less prone to appearing and vanishing than a single-period oscillator's.
Its author designed distinct parameter sets for short-term, intermediate and long-term analysis, and intended them to be used together: the long-term KST establishes the regime, the intermediate one identifies the swing, and the short-term one times the entry. Used that way it is a complete framework rather than a single line, and that layered use is what distinguishes it from simply running MACD on three timeframes.
Against its neighbours: it is slower than MACD and much slower than any single-horizon oscillator, which is the intended trade. It sits closest to the Coppock Curve in philosophy — both blend rates of change over long windows and smooth the result — but KST uses four horizons rather than two, works on any timeframe with the appropriate parameter set, and provides a signal line, so it produces regular tradable events rather than a signal every few years.
Calculation
The arithmetic in words, in the order it happens.
Build four smoothed rate-of-change components. Component one: a 10-period simple moving average of the 10-period ROC. Component two: a 10-period simple moving average of the 15-period ROC. Component three: a 10-period simple moving average of the 20-period ROC. Component four: a 15-period simple moving average of the 30-period ROC. Then KST = (component one x 1) + (component two x 2) + (component three x 3) + (component four x 4), so the longest horizon carries four times the influence of the shortest. The signal line is a 9-period simple moving average of KST. Each ROC is the standard 100 x (close - close N periods ago) / (close N periods ago).
Source
An AlgoBeamScript implementation of the formula above, written by us from the arithmetic so the code and the calculation agree line for line.
Runs unchanged on the platform and in the AlgoBeamTS runtime. The language reference is in the documentation.
Inputs
Defaults are the values most charting packages ship with. They are conventions, not optimal settings — the right length depends on your instrument and your holding period.
| Input | Default | What it changes |
|---|---|---|
| ROC Lengths | 10, 15, 20, 30 | The four momentum horizons being blended. Scaling all four up produces a slower line for higher-timeframe analysis; the author published separate sets for short, intermediate and long-term work rather than expecting one set to serve every horizon. |
| Smoothing Lengths | 10, 10, 10, 15 | Simple moving average applied to each rate of change before weighting. These are what turn four jumpy ROC series into components stable enough to add together; shortening them reintroduces the noise the design exists to remove. |
| Signal Length | 9 | Simple moving average of KST used as the trigger line. Longer settings reduce the number of crossovers and delay each one, in the usual trade-off. |
How to read it
What practitioners take from the plot. Read these as descriptions of market state, not as entry signals.
- KST crossing above its signal line
- Blended momentum is turning up relative to its own recent average. The primary trigger, and the same event a MACD signal cross represents but on a slower, multi-horizon input.
- KST crossing below its signal line
- The mirror short trigger. On the long-term parameter set these crossings are infrequent enough to be treated as position-level decisions.
- Crossing above zero
- Momentum across all four horizons has turned net positive. Slower and higher-conviction than the signal cross, and commonly used purely as a regime filter for faster rules.
- Long-term KST positive while short-term KST turns up
- The layered use the indicator was designed for: regime from the slow set, timing from the fast one. This alignment is the framework's actual signal.
- Divergence against price at a swing extreme
- Blended momentum failed to confirm a new price extreme. More stable than single-period divergence because four horizons must all fail to confirm.
Limitations
Where this indicator misleads. None of these are fixed by a better parameter.
- Four rates of change plus four smoothings means substantial lag. Signals arrive well after the turn, which is the deliberate price of the smoothing but a real cost in fast markets.
- It has eight parameters before the signal line, which makes it easy to overfit and hard to diagnose. Changing one length shifts the blend in ways that are not intuitive.
- In sideways markets the line still crosses its signal repeatedly, and because it is slow those whipsaws are expensive — the position is held through most of the adverse move before the exit triggers.
- Its output is in percentage-momentum units summed with weights, so the absolute value has no natural interpretation and cannot be compared meaningfully across instruments.
Educational reference. This page explains how an indicator is built and how it is commonly read. It is not investment advice, not a recommendation and not a signal service. No indicator is profitable on its own — each is a way of describing a market, and any rule built on one has to be tested with realistic costs before it is traded.