Workflows

This page is for the reader who already knows the pane and wants to know how to use it. Three patterns are described below. Each is a way a committed trader can actually run the indicator — not a trade setup and not a rule set. Each pattern ends with the questions the reader should be asking themselves while they watch the pane, not the actions they should take.

Three anti-patterns follow the patterns, named plainly. The reason anti-patterns live on the same page as patterns is that the anti-patterns are the shortest path from a clean configuration to a broken trust boundary, and naming them beside the patterns keeps that risk visible.

Nothing on this page promises a win rate, tells you what to trade, or recommends "optimal" settings. The point is the read. What you do with the read is your work.

Pattern A — Intraday trend posture check

The configuration

Defaults. Slot 01 at 5m, slot 02 at 15m, slot 03 at 60m. Slots 04 and 05 off. Weights 33.3 on the three active slots. On Bar Close? = true on all three. BBWP on. Donchian on. Keltner off. Master smoothing off. Divergence on with Pivot Len 20 and Plot On Pivot off.

The read

You are watching an intraday chart — 5m or lower — and you want a sanity check on the blended stochastic posture across three intraday timeframes. The blend is the hero. You read it in five moves.

1. Blend fill colour. Lime means the weighted mean of 5m, 15m, and 60m K lines is above the weighted mean of the corresponding D lines. Red means the inverse. That is your starting regime read.

2. Blend K position in the 0..100 range. Is it near 50? Near 80? Pinned above 80? A pinned read is not the same as an exhausted read. A blend pinned above 80 in a real intraday trend can stay pinned for the whole session.

3. BBWP columns. Short blue columns mean the blend has been compressed against its own history — a low-width regime in oscillator space. Tall aqua columns mean the blend's band width has expanded above the configured threshold. When BBWP is contracting, expect quieter blend movement; when BBWP is expanding and the blend is lifting past 50, that is the mature confirmation of a developing trend posture.

4. Donchian edges. Is the blend K pressing the upper stepline? The lower? Sitting in the middle of the channel? A blend pressing an upper stepline while BBWP expands is the blend saying "I am making new local highs and my range is opening up" — one kind of trend signature.

5. Divergence triangles, if any. A recent bullish triangle near the floor in an already-bearish blend is the pane asking you a question: chart price just made a lower low while the blended K did not. That is a question worth holding open; it is not a trade.

The questions to ask yourself

• Is the blend regime (lime or red) matching what I see on the chart itself?

• If not, which slot is driving the disagreement? (Unhide the slot plots temporarily to find out.)

• Is BBWP telling me the blend is quiet or loud right now?

• Is the Donchian channel compressing or widening?

• If a divergence just confirmed, how confident am I that chart price has stopped doing what the oscillator is saying it stopped doing?

The cost this pattern buys

Pattern A is the simplest configuration with the most surface. You are asking a three-timeframe stochastic stack to tell you about intraday posture, using two structure features for context. The cost is that three slots at the same default settings (K 14, SMA smoothing 3, D 3 SMA) across three timeframes sample one measurement at three cadences — not three independent observations. Aligned stochastic reads across your three intraday timeframes are not "three votes" in any meaningful sense; they are one method sampled at three speeds. Hold that fact in mind when the blend looks decisive.

When to expand the pattern

• You want a cross-asset leg in the blend. Move to Pattern B.

• Three slots are not enough. Move to Pattern C, or consider whether the fourth slot will actually add information or just more of the same.

Pattern B — Cross-asset context

The configuration

Slot 01 on chart symbol at chart TF (or at the chart TF explicitly), On Bar Close? = true. Slot 02 on chart symbol at 60m, On Bar Close? = true. Slot 03 on a correlated or benchmark symbol via Optional Ticker, same timeframe as slot 02 or your preferred HTF, On Bar Close? = true. Slots 04 and 05 off. Weights 33.3 on all three, or biased toward the chart symbol if you want the chart to lead the blend. BBWP on. Donchian on. Keltner optional. Master smoothing off. Divergence on.

The read

The blend is now a two-thirds-self, one-third-context read. You are asking: "what does my instrument look like on the blended stochastic when I fold in a cross-asset leg?" The pane still shows one blend line, but that line is now a mix of chart-symbol stochastic and cross-symbol stochastic.

The five-move read from Pattern A still applies, with two additions:

• Watch how the blend behaves around session boundaries on the cross-symbol slot. If slot 03 points at an equities symbol and your chart is on a 24-hour futures contract, the cross-symbol slot will have data only during the equities session. Outside that session, slot 03 may flatten or show stale values, and the blend will effectively rebalance onto slots 01 and 02. That rebalance is not a new signal; it is a mechanical artefact of the session mismatch.

• Treat divergence readings with extra care. The divergence module compares chart-price pivots against the blended K. With a cross-symbol slot in the blend, the blended K now includes a stochastic reading from a different instrument. A divergence confirming here is still a legitimate geometric pairing, but its interpretation is murkier than Pattern A's same-symbol divergences. Use it as a question, as always, but a question with an extra caveat attached.

The questions to ask yourself

• Is the cross-symbol slot actually adding information I could not get from just reading the cross-symbol's chart?

• Is the cross-symbol slot's session matching my chart well enough that the blend behaves sensibly across session boundaries?

• When the chart symbol and the cross symbol disagree, what does that disagreement look like on the pane? (A blend pulled toward 50 by disagreement is the pane showing me indecision across the two symbols.)

• Is the divergence I just saw on the blended K a chart-symbol divergence or a cross-symbol-driven one? (Unhide slot 03's plot temporarily to check.)

The cost this pattern buys

You gain a cross-asset leg inside the blend. You lose some interpretability because the blend is now a hybrid. The divergence module still works but its output is about a series that is not purely the chart symbol's. Some readers find this useful when they are already running a cross-asset read in their head and want an instrument for it. Other readers find it confusing and prefer to run the cross-symbol's chart separately. Both responses are legitimate.

When this pattern is wrong

If you point slot 03 at a symbol whose liquidity regime is very different from the chart symbol's — a large-cap benchmark paired with a tiny-cap name, for example — the blend becomes a hybrid across liquidity regimes. The stochastic on a tiny-cap with wide spreads and gappy bars is a different creature from a stochastic on a liquid benchmark. The blend does not know that. Your read has to.

Pattern C — Fuller stack with a deliberate weight shift

The configuration

All five slots on. Slots 01, 02, 03 at three intraday timeframes with weights 33.3 each. Slot 04 at a higher timeframe (240m or D) with weight 20 or 15. Slot 05 at a different source (HLC3 or hl2) or a different symbol, with weight 10 or 15. On Bar Close? = true on every enabled slot. Master smoothing optional with a short EMA length (2 or 3) to take the edge off the five-slot noise. BBWP on. Donchian on. Keltner optional. Divergence on.

Important: if you enable slot 05, the All Stoch Slots Bullish and All Stoch Slots Bearish alignment alerts will not fire. This is documented on the alerts page and on the limitations page. Pattern C assumes you are reading the blend visually and using per-slot alerts, not depending on the alignment alerts.

The read

The five-move read still applies. The difference is that your blend is now a deliberately differentiated stack: three slots sampling the chart symbol at intraday cadences, one slot on a higher timeframe for slow context, one slot running a different source or a different symbol for a specific reason.

The goal is not "more slots equals more information." The goal is that each slot is doing a specific job the others cannot do:

• Slot 01, 02, 03: intraday cadence on the chart symbol.

• Slot 04: higher-timeframe context on the chart symbol.

• Slot 05: differentiation — a different source, a different symbol, or a different family.

With the weights biased toward the intraday slots (33.3 each), the higher-timeframe and differentiation slots are moderators, not drivers. The blend still leads with intraday posture and the slow-context and differentiation slots tug it toward or away from the intraday read.

If you want slot 04 and slot 05 to lead, raise their weights. The weights are relative; you do not have to sum to 100. Scaling all weights by a constant does not change the blend.

The questions to ask yourself

• Is slot 04 actually adding HTF information, or is it just a slower version of what the intraday slots are already saying?

• Is slot 05's differentiation giving me something different, or am I running five versions of the same measurement?

• Is the blend saying something coherent, or is it sitting near 50 most of the time because the slots are disagreeing?

• When a divergence confirms, which slot's K was the one that pulled the blend into the divergent geometry?

• Has master smoothing helped me read the blend, or has it just added lag without clarifying anything?

The cost this pattern buys

You gain a richer read with more perspective. You lose some speed because the pane is busier and because five slots running through master smoothing can feel slower than three. You also take on the complexity of verifying that each slot is doing a specific job — and the quiet risk that you will one day reconfigure slot 05 and stop noticing that alignment alerts are silent.

When this pattern is wrong

If you cannot name what slot 05 is for, slot 05 should not be enabled. If the five slots all run the same source, the same lookback, the same MA family, and differ only in timeframe, the blend is effectively the same measurement sampled at five cadences — and turning that into five votes is exactly the over-trust risk the Limitations page names. Enable slots deliberately or not at all.

Anti-pattern A — Divergence as entry

The most magnetic visual on this pane is the divergence triangle. A freshly printed lime triangle near the floor can look like an invitation. It is not.

A confirmed bullish divergence is a geometric description of a pivot that already completed. The pivot itself is Pivot Len bars to the left of the confirmation bar. Chart price has had Pivot Len bars to do whatever it was going to do by the time the triangle appears. A trade placed on the confirmation bar is trading on a lagged descriptor, not on the pivot.

The correct use of a divergence triangle is as a question, not as a trigger:

• "Chart price made a lower low. The blended K did not. What is going on in this market that produced that mismatch?"

• "Is there another reason I would take this side of the market — price action, volume, context, session — independent of the divergence?"

• "If I would not take this side without the divergence, the divergence is not giving me conviction; it is giving me an excuse."

The divergence triangle is a place to slow down and look at everything else on your screen. It is not a place to press the button.

Anti-pattern B — Five-slot alignment as breadth proof

Five enabled slots all bullish, the All Stoch Slots Bullish alert firing (or rather, it would be firing, if slot 05 were not enabled — see the alerts page), feels like five independent witnesses agreeing. It is not.

Two cases:

1. Five slots at similar defaults, differing only in timeframe. Same source, same lookback, same MA family, same source smoothing. The five slots are one measurement sampled at five cadences. When they all agree, you are looking at one thing holding true across five lookbacks — not five independent confirmations. The blend may be decisive, but the breadth is illusory.

2. Five slots deliberately differentiated. Different sources, different lookbacks, different MA families, different symbols. In this case the slots really are doing different jobs, and alignment across them is more meaningful. But even here, the four structure features still read off one blended K, so structure-feature co-motion does not add new breadth.

The useful posture: read five-slot alignment as "the stochastic pipeline is leaning this way under the specific assumptions I configured." Not as "five independent observers confirm." The difference is everything.

Anti-pattern C — BBWP as price volatility

The BBWP columns at the bottom of the pane measure the Bollinger width percentile of the blended K. Not of price. Short blue columns mean the blend's band width has been compressed by its own recent history. Tall aqua columns mean the blend's band width has expanded above the configured threshold.

Sometimes the oscillator and price are volatile together. Sometimes they are not. A trending session can produce a compressed oscillator (tight stochastic readings) with high price volatility (wide price bars). A choppy session can produce an expanded oscillator (volatile stochastic) with quiet price action (narrow bars mean stochastic swinging between extremes on small price moves).

If your read is "BBWP is low, price must be quiet, I should avoid this session" — you are substituting one measurement for another. The correct read is: "the blend's own width is compressed right now. If I want to know about price volatility, I need to look at price."

A short closing note on running these patterns

The three patterns are not ranked by sophistication. Pattern A is not a beginner's pattern. Pattern C is not an advanced pattern. They are different configurations for different jobs. A reader who has run the indicator for a year may spend most of their time in Pattern A because Pattern A answers the question they actually have on the instrument they actually trade; the same reader may move to Pattern C only on specific days when a cross-asset or higher-timeframe context is genuinely at play. Moving between patterns is a response to the question you are asking on that session, not proof that one configuration is more serious than another.

When to step between them:

• Step from A to B when your day-to-day read is already folding in a specific cross-asset relationship — DXY on an FX chart, a benchmark index on a single-name equity chart, a related future on a spot chart — and that cross-asset read has become load-bearing enough that you want it inside the blend rather than on a separate panel. The motion from A to B is a formalization of a habit you already have, not a new technique you are adopting.

• Step from A to C when three slots are not enough to cover a specific job — you want a higher timeframe context the intraday slots cannot supply, or a source variation that reveals a particular aspect of intra-bar behaviour — and you can name what slot 04 and slot 05 are doing that no rearrangement of slots 01–03 could do. If the answer is "slot 04 is another intraday slot with slightly different settings," slot 04 is noise dressed as breadth and Pattern A is still the right configuration.

• Step down from C to A when the pane is busy and the read is not cleaner for it. More slots do not mean more clarity. A pane with three confident slots tends to produce a blend that agrees with itself when agreement is warranted and disagrees with itself when disagreement is warranted. A pane with five slots running similar measurements at five cadences tends to produce a blend that always looks somewhere near consensus, which is exactly how you lose the informative disagreements that would have flagged a regime change.

A deliberate pattern is one where you can name, per slot, what that slot is for. If you cannot, you are running a configuration you inherited, and the pane is harder to read than it needs to be. The pack does not care which pattern you run. It cares that the pattern you are running is the one the reading you are doing actually needs.

Where to go next: Visuals and logic for the mental model these patterns rely on, Alerts for how to wire the patterns into notifications, Limitations and trust boundaries for the honest edges that the anti-patterns surface, Troubleshooting for symptom-to-cause rows if a pattern stops behaving the way you expected.