Workflows
This page walks through repeatable use patterns and named anti-patterns grounded in how the script behaves. Each workflow includes enough context for you to evaluate whether it fits your trading. The anti-patterns inc...
Written By Axiom Admin
Last updated About 1 month ago
Workflows
This page walks through repeatable use patterns and named anti-patterns grounded in how the script behaves. Each workflow includes enough context for you to evaluate whether it fits your trading. The anti-patterns include enough explanation that you can recognize them before they cost you.
Workflow examples
Three-timeframe structural stack
What it is: Three slots at staggered timeframes providing short, medium, and longer structural context on one chart. The blend summarizes the three into a single composite.
Example setup:
How to use it: Read the chart from the widest timeframe inward. Start with DC 03 (4H) to get the broadest structural picture β is price in the upper or lower half of the 4-hour range? Then check DC 02 (1H) for the medium picture. Then DC 01 (15m) for the nearest structure. The blend gives you a composite glance, but the individual slots tell you where the agreement and disagreement live.
When the slots agree: All three read bullish or bearish. This is meaningful structural alignment across genuinely different time horizons. The blend will confirm the direction and sit comfortably among the individual channels.
When the slots disagree: The short-term slot flips while the longer-term slots have not. This is where the useful structural information lives β and where the urge to simplify is strongest. You will want to pick a side. Resist that for a moment.
Look at the blend first. If it is weighted toward the short-term slot, the blend will read bullish even though the longer-term slots have not confirmed. That can make the disagreement feel more resolved than it is. Then look at the individual channels. How far is price from the longer-term basis? Is it testing the boundary or nowhere near it? Is the short-term flip happening on increasing range (widening channel) or compressing range?
The indicator shows you the question: something is shifting at one structural scale that has not propagated to the others. The answer β whether the short-term move leads or fades β requires context the indicator cannot provide. But seeing the question early, before the longer-term structure catches up, is the specific advantage of a multi-timeframe stack.
Weight guidance: Equal or near-equal weights (like 35/35/30) treat all three timeframes as roughly equivalent contributions to the blend. If you want the blend to lean toward longer-term structure, increase the longer-timeframe slot's weight. If you want it more reactive to short-term shifts, increase the short-timeframe slot's weight. There is no correct distribution β it depends on what you want the blend to emphasize.
Why this works: Three timeframes that are meaningfully separated give you three structural lenses. Each one captures a different market rhythm. Together they produce a picture that any single timeframe misses: whether short-term and long-term structure are telling the same story.
Cross-market context check
What it is: One slot configured to pull data from a different instrument, showing that instrument's Donchian structure scaled into your chart's price space.
Example setup: Trading ES (E-mini S&P 500 futures) and wanting to see SPY's structural context on the same chart.
How to use it: Set the weight to 0 so the cross-ticker slot does not pull the blended channel. This keeps it as a standalone visual reference. Watch where SPY's scaled Donchian structure sits relative to your ES chart. When both instruments' structures align β both in the upper half of their respective channels, or both breaking above their upper bounds β the structural picture is reinforcing. When they diverge, you have a question worth investigating.
Why weight zero matters here: Cross-ticker scaling is approximate. The ratio between the two instruments' closes updates each bar, and the scaling drifts when the instruments diverge. Feeding an approximate scaled channel into the blend would introduce imprecision into the composite without the user necessarily knowing how much. Keeping the cross-ticker slot at weight zero lets you see the structural reference without contaminating the blend.
When this is useful: When you trade an instrument that has a strong structural relationship to another instrument and you want a quick visual check on whether the related market's structure confirms or contradicts what you see on your primary chart.
When this is not useful: When the two instruments have no meaningful structural relationship. Scaling corn into a Bitcoin chart will produce lines, but those lines do not carry structural information. The scaling is mechanical β it does not know or care whether the relationship is meaningful.
Hidden structural influencer
What it is: A wide-timeframe slot that is enabled, hidden from the chart, and given blend weight. The blended channel is "anchored" toward longer-term structure without adding visual noise.
Example setup:
How to use it: DC 03 does not appear on the chart, but its 40% weight pulls the blended channel toward the daily structural range. The blend moves more slowly and reflects a longer-term anchor than the visible channels alone would suggest. This is useful when you want the blend to represent a broader structural context while keeping the chart visually clean.
The critical thing to remember: The blend will not match the visible channels. If you forget that DC 03 is hidden and weighted, the blend will appear to be in a position you cannot explain from the visible data. A week later, you will wonder why the blend was "wrong" during a session β it was not wrong, you just could not see what was pulling it. This workflow only works if you keep clear notes about what is hidden and why. Mental notes are not enough if you revisit the chart days later.
When this makes sense: When your primary chart work uses shorter timeframes but you want the blend to carry a longer-term bias β pulling the composite toward a structural reference that would otherwise require adding another visible channel to an already busy chart.
Weight-zero reference channel
What it is: A slot that is enabled and visible on the chart but has its blend weight set to zero. It is a pure visual reference that does not influence the blended channel.
Example setup:
How to use it: The weekly channel appears on the chart so you can see where price sits relative to the weekly range. But because its weight is zero, it has no effect on the blended channel. The blend only reflects the other slots.
When this makes sense: When you want to monitor a specific structural timeframe independently β to watch for weekly-level breakouts, for example β without letting it distort a blend that is designed around shorter-term timeframes. It is also useful when you are not sure whether a slot belongs in the blend yet and want to observe its behavior before assigning it weight.
Anti-patterns
All ten slots enabled
What happens: The chart becomes a mess of overlapping channels. Visual interpretation breaks down because the individual channels are difficult to distinguish. The blend becomes a composite of so many inputs that it smooths out most structural detail.
Why traders do it: The feeling that more data should produce better decisions. Ten slots exist, so using all of them seems like using the tool fully.
Why it fails: Structural insight comes from well-chosen timeframes that capture meaningfully different market rhythms. Past four or five slots, additional timeframes usually overlap with ones you already have. The extra channels add visual noise and blend complexity without adding structural information. The blend degrades into a general average that is too diffuse to reveal specific structural tensions.
What to do instead: Choose three to five slots that cover genuinely different time horizons. Leave the rest disabled until you have a specific structural question they answer.
All slots on the same timeframe with different lengths
What happens: Multiple Donchian Channels on the same data series with different lookback windows. The channels show the same timeframe's structure at different scales.
Why it is not multi-timeframe analysis: The slots are all looking at the same bars. Their agreement is a function of arithmetic overlap, not structural diversity. The blend averages different lookback windows on the same data, which is just a smoothed version of one timeframe.
When it might still be useful: If you specifically want to study how different lookback lengths capture structure on a single timeframe β that is a valid exercise, but it is not what the multi-timeframe architecture is designed for. Know the difference.
Backtesting with On Bar Close off
What happens: The channels look smooth and responsive in history. Breakouts appear clean. The structural readings seem more timely than they would with the staircase pattern.
Why it is dangerous: Those historical values were not available at the time. TradingView shows the final state of each building bar as if it were known from the beginning. Any backtest using this data is evaluating decisions against information that did not exist when the decisions would have been made. The results will look better than reality because the chart is giving you the answers before the questions were asked. And the problem is invisible β the chart does not look wrong. It looks clean. That cleanliness is the lie.
What to do instead: Leave On Bar Close on for any slot that feeds a backtest, strategy evaluation, or historical analysis. If you want fast updates for live trading, create a separate slot for that purpose and set its weight to 0 so it does not contaminate the blend.
Cross-ticker with unrelated instruments
What happens: The scaling produces a channel on the chart, but the channel does not represent meaningful structural information. The ratio between unrelated instruments wanders without structural logic, so the scaled channel drifts in ways that have nothing to do with either market's actual structure.
Why traders try it: Curiosity, or a hope that overlaying one market's structure on another will reveal hidden relationships. Sometimes it does. Usually it does not.
How to tell the difference: If the scaled channel's movements correspond to structural shifts you can also see on the alternate instrument's own chart, the relationship is plausible. If the scaled channel just wanders without any correspondence to either instrument's behavior, the relationship is not there.
Treating the blend as consensus
What happens: A trader sees the blended channel reading bullish and interprets it as "all timeframes agree." They do not check the individual slots.
Why it misleads: The blend is a weighted average, not a vote. A single slot with 80% weight and nine slots with a combined 20% weight will produce a blend that is effectively one slot's opinion. The blend does not measure agreement β it measures a weighted midpoint. Two slots pointing in opposite directions produce a blend in the middle, which looks neutral but does not mean the market is neutral. It means two structural timeframes disagree and the blend is splitting the difference.
How to check whether the blend is telling you something real: Open the data window and compare the blended basis to each individual slot's basis. If the blended basis is within a few ticks of one slot and far from the others, that slot is dominating the composite. If the blended basis sits roughly between the individual slots, the blend is actually aggregating. Then check the weights β do they reflect a deliberate hypothesis about which timeframes matter more, or did you set them once and forget about them? The blend is only as thoughtful as the weights you assigned it.
What to do instead: Use the blend for quick visual orientation β "where does the composite structural picture sit right now?" β and then check the individual slots before making any decision. The individual channels tell you whether the structural picture is unified or divided. The blend tells you a weighted average. When those two stories differ, trust the individual channels.