Workflows

This page walks four named workflows and four named anti-patterns end to end, as concrete routines a reader can follow and verify. Each workflow names why a reader would reach for it, how to configure it in a handful of steps, how to confirm the configuration is behaving the way you intended, and one common misuse to avoid. The anti-patterns at the end are configurations that look reasonable and quietly teach the wrong lesson.

Pick the workflow that matches the job you are actually trying to do on the chart — not the one that uses the most features. Narrow is a feature. A two-slot focus that answers the question you are actually asking beats a three-slot stack you are half-reading.

Workflow 1. The three-slot default read

Why a reader reaches for this. You want a working multi-timeframe Donchian picture on one chart without spending time on configuration yet. You want to confirm the indicator renders, see the three timeframes step at their own cadences, and start building an intuition for how agreement and cancellation show up in the blended channel.

Setup.

1. Add Axiom DC to a 1-minute chart of a liquid symbol.

2. Leave every input at its shipped default. Three slots: Slot 01 at 5m weight 40, Slot 02 at 15m weight 35, Slot 03 at 60m weight 25. On Bar Close? on.

3. Confirm the picture: three colored slots (teal, blue, purple), each with its three lines, and a red blended channel with a translucent red fill and a lime basis.

Verify.

• Watch one full 5-minute bar close. The teal slot's upper and lower lines step at that close, not before. Under ON, this is the expected behavior.

• Hover a slot line in TradingView's data window and check the title — "DC 01 Upper," "DC 02 Basis," etc. Confirm each line is the slot and role you think it is.

• Toggle Hide DC 02 Plot briefly. The blue slot's three lines disappear; the blended channel does not change. Toggle back.

Common misuse.

Reading the red blended channel as "the answer" and ignoring the three slots underneath. The blend is a weighted composite biased 40 / 35 / 25 toward the faster timeframes. A read that only looks at the blend will miss cases where two slots disagree and the third is carrying the vote.

Workflow 2. A custom HTF ladder that fits your cadence

Why a reader reaches for this. The shipped 5m / 15m / 60m is not your cadence. You trade on a different timeframe, your read depends on different structure, or you want the stack to reflect the timeframes you already look at manually. A custom ladder is a three-slot stack you author yourself, at your own weights, without apologizing for stepping outside the defaults.

Setup.

1. Choose three timeframes that match your read. A common pattern is "near timeframe, structure timeframe, context timeframe" — for example 15m / 1H / 4H for a swing read, or 1m / 5m / 15m for a fast scalp read. Constraint: every slot's timeframe must be equal to or higher than the chart's.

2. In the inputs dialog, set Slot 01's TimeFrame: to your near timeframe, Slot 02's to your structure timeframe, and Slot 03's to your context timeframe.

3. Set weights deliberately. If the structure timeframe is the one you trust most for the question you are asking, give it the largest weight. If you want context to dominate, weight the context slot heaviest. Equal weights are a valid statement that all three timeframes are equally informative — treat that as a claim you are making, not a neutral starting point.

4. Leave On Bar Close? on. If the stack is for context rather than for micro-timing, ON is the posture you want.

5. If Length: 20 does not fit the timeframes you chose — for example a 4H slot may want a different lookback than a 15m slot — adjust per slot.

Verify.

• Watch each slot step on its own timeframe. If Slot 03 is at 4H, its lines hold still for four hours and step at each 4H close.

• Confirm the blended channel is moving where your weighting suggests it should. If you weighted the context slot heaviest and the blend sits close to the context slot's lines, the composite is doing what you asked.

• Confirm no runtime-error banner. If you see one, a slot's timeframe is below the chart's; fix that slot.

Common misuse.

Treating "a custom ladder" as "upgrading" the defaults. The defaults are not a floor, and a custom ladder is not automatically more correct. The right ladder is the one that matches the question you are asking; if the defaults already do that, leave them alone.

Workflow 3. Two-slot focus

Why a reader reaches for this. You want less density on the chart. A three-slot stack is adding a timeframe you are not really using, and the blended channel is either redundant with the two slots you care about or is being pulled around by the third slot in a way that does not help your read. Disable one slot. Work with two.

Setup.

1. Decide which two slots you want. A common pair is "near structure plus context": keep Slot 01 and Slot 03, disable Slot 02.

2. Turn off Enable DC 02 (or whichever slot you are dropping). The disabled slot's three lines disappear and the slot stops contributing to the blend and to alignment alerts.

3. Set the remaining two slots' weights to reflect the relative importance of near structure versus context. If you want the near slot to dominate, weight it heavier; if you want the context slot to anchor the read, weight it heavier. The numbers do not have to sum to 100 — the blend normalizes by the total contributing weight.

4. Leave On Bar Close? on.

Verify.

• The disabled slot's lines are gone and stay gone.

• The blended channel is now a composite of only the two remaining slots. Zero one of the two weights briefly and confirm the blended lines snap to the other slot's values; restore the weight.

• Alignment alerts now require agreement between the two remaining slots only — a weaker condition than 3-of-3 alignment. If alignment alerts are part of your setup, adjust how much weight you give them.

Common misuse.

Disabling a slot you have not yet looked at, because "three is too many." The slot you are not looking at may be the one carrying the context your read actually depends on. Take one pass through the three default slots before deciding which one to drop.

Workflow 4. Alerts-only silent slot

Why a reader reaches for this. You want an alignment alert that factors in a timeframe you do not want to plot. The classic case: you want "all three timeframes agree" as an alert, but you only want to look at two of them on the chart. Another case: you want a specific slot's basis-trend alert without drawing the slot's lines.

This is a power-user pattern. Before reaching for it, make sure the lines you are hiding are genuinely not worth seeing — a hidden slot is a slot whose contribution the reader cannot directly look at.

Setup.

1. Enable the slot you want to hide: Enable DC NN on.

2. Turn on Hide DC NN Plot for that slot. The three lines disappear from the chart.

3. Decide whether the hidden slot should steer the blend. If not, set its Blended Weight to 0. If yes, leave the weight non-zero and accept that the blended channel will shift based on a slot whose lines are not on the chart.

4. Configure the alert you want in the usual way. A per-slot basis-trend alert on the hidden slot will fire based on that slot's state; an alignment alert will require the hidden slot to agree along with every other enabled slot.

Verify.

• The hidden slot's three lines are not on the chart.

• With Blended Weight = 0, the blended channel behaves as though that slot were not there.

• With Blended Weight non-zero, the blended channel moves in response to the hidden slot — watch the blend while the hidden slot's basis trend flips and confirm the movement.

• Set up the alignment alert and confirm it only fires when the visible slots agree and the hidden slot also agrees. When they disagree, the alignment alert stays silent. This is the alert working correctly — alignment counts hidden slots.

Common misuse.

Forgetting the slot is hidden. Three months later the chart has a hidden slot with a non-zero weight, the reader cannot explain why the blended channel is moving where it is, and the cause is an invisible configuration they set up and forgot. Keep a note somewhere, or leave the slot's weight at zero so the blend ignores it.

Anti-patterns — named traps worth avoiding

Anti-pattern 1. "On Bar Close? OFF reacts faster" as the whole thought

OFF is faster in the sense that the slot lines reflect the live higher-timeframe bar. What you are really buying is redraws on the current chart bar until each slot's higher-timeframe bar closes. Because the switch is global on this trim, one flip changes the posture of every slot at once — the 60-minute slot is now reporting a value that can move for the next 59 minutes, not just the fast slot.

This is an anti-pattern when it shows up as reflex. Read MTF & Repainting and decide on the posture with the tradeoff in hand.

Anti-pattern 2. Equal weights without a reason

Setting Slots 01, 02, and 03 to equal weights (for example 33 / 33 / 33) because "equal feels neutral." Equal weighting is a claim: each of the three timeframes is equally informative for the question you are asking. Sometimes that claim is right. Usually it is worth examining.

If you want the blend to lean toward a specific timeframe, weight that slot heavier. If you genuinely think the three are equally informative, set them equal — but do it as a statement, not a default.

Anti-pattern 3. Basis Trend Length = 1 as "more responsive"

A Basis Trend Length of 1 compares the current basis to its value one bar ago. On a slot's own timeframe, that is effectively "is the basis higher than it was on the last bar." The resulting trend boolean flickers on small moves, and every per-slot basis-trend alert becomes noisier for it.

If you want more responsiveness, lower Basis Trend Length to 2 or 3, not 1. The shipped default of 3 is a middle ground. If the alerts coming out of a length-1 configuration feel too noisy, the length is the cause.

Anti-pattern 4. Chasing Length: extremes

Pushing Length: to very small values (like 3 or 5) on a fast timeframe often produces an outer channel that tracks micro-noise — the highs and lows move on every few bars, the bounds wobble, and the basis does not save you because the basis smoothing only smooths the midpoint, not the bounds. The result is a ragged outer channel that looks busy but is not carrying range information.

Pushing Length: to very large values smears the range across a long window. The outer channel becomes a slow wide band that does not reflect current structure.

Both extremes are legitimate for specific reads and both are anti-patterns when they happen by accident. If the outer channel looks wrong for the instrument and the timeframe, Length: is the first place to look — not Basis MA Length.

One last note

All four workflows above live inside the Base trim's surface — three slots, one global repaint switch, the chart symbol only, the MA Lite smoother catalog. If your workflow requires behavior that sits outside that surface — a cross-symbol slot, a per-slot repaint switch, a Pro-catalog smoother, more than three slots — you are not asking the wrong question. You are asking for a feature that lives on the CTX trim (axiom-mtf-dc-ctx). Reach for that trim instead of bending this one. Limitations & Trust Boundaries has the scope picture in one place.

A rule of thumb that shortens a lot of configuration dead ends: if you find yourself trying to fake a feature that this trim does not carry — a hidden slot on the chart symbol standing in for a cross-symbol read, three slots at near-identical timeframes standing in for a fourth slot — stop. You are paying the price of a workaround and still not getting the feature. Use the trim that carries the feature you actually need. The right tool is almost always cheaper than the clever configuration.