Workflows

Workflows

This page teaches concrete ways to use the indicator and concrete ways to misuse it. Each pattern comes with enough context that you can judge whether it fits your situation, not just copy the setup. The anti-patterns are just as important as the validated patterns — knowing what not to do with this tool will save you more trouble than knowing the optimal configuration.

Validated patterns

Three-timeframe confluence check

What it is: The default setup — three slots at different timeframes (5m, 15m, 60m by default) on the same instrument, checking whether short-term, medium-term, and longer-term baseline-distance readings agree.

How it works: Each slot measures how far price is from its baseline MA on its assigned timeframe. When all three show Fast above Slow (teal/aqua/blue — all in bullish regime), the setup has multi-timeframe support. When they disagree, the setup is in conflict.

When to use it: When you want to check whether a potential trade has agreement across the timeframes that matter to your process. Swing traders might use 15m/60m/D. Day traders might use 1m/5m/15m. The specific timeframes are less important than choosing ones that represent genuinely different positioning scales for your holding period.

How to configure it:

• Enable three slots with your chosen timeframes.

• Keep MA Type, Length, and Source consistent across the three slots (e.g., EMA 20 on close for all three). This isolates the timeframe as the variable.

• Set equal blend weights (or weight the timeframe you care most about slightly higher).

• Keep On Bar Close enabled on all three slots for trustworthy readings.

What to watch for:

• Progressive alignment — short-timeframe slot flips first, then medium, then long. This is a trend building across timeframes and is a stronger signal of directional commitment than all three flipping simultaneously.

• Short-vs-long disagreement — the 5-minute slot is bullish but the 60-minute slot is bearish. This usually means price is rallying within a broader downtrend, or the broader trend is in the early stages of turning. The oscillator does not tell you which — it shows you the disagreement. This is the moment that tests your discipline. If you are a trend-follower, the long slot's bearish reading is the one that governs your entries, and the short slot's bullishness is a counter-trend move you wait out. If you are a mean-reversion trader, the short slot's divergence from the long slot might be exactly what you are looking for. The oscillator serves both approaches. It does not pick one for you.

• All three at extremes — when all slots read above +80 or below -80, the multi-timeframe stretch is high. This does not mean a reversal is imminent. It means the position is extended relative to all three baselines. In a strong trend, this condition can persist for a long time.

What makes this pattern useful: It gives you a structured way to check whether the timeframes you care about agree before you commit. It does not tell you whether to trade. It tells you whether the baseline-distance picture is consistent or conflicted.

Cross-market correlation monitor

What it is: Two or more slots set to different tickers on the same timeframe, watching for agreement or divergence between related instruments.

How it works: Each slot starts from an ATR-normalized distance and then maps that reading onto the same bounded scale, so a reading of +40 on SPY and +40 on QQQ reflects a similar kind of volatility-adjusted stretch under the current sensitivity setting. When they move together, the correlation is holding. When they diverge — one rising while the other falls — something is shifting in the relationship.

When to use it: When you trade instruments that are typically correlated and you want to catch the moments when that correlation breaks down. Examples: SPY vs. QQQ, BTC vs. ETH, a sector ETF vs. its largest component, crude oil vs. an energy stock.

How to configure it:

• Set each slot to the same timeframe (so the comparison is apples-to-apples on the time dimension).

• Use the Optional Ticker field to assign different instruments to each slot.

• Keep the MA Type, Length, and Source consistent across the comparison slots.

• Consider setting blend weights to 0 for these slots and watching the individual lines rather than the composite. The blend of two diverging instruments will average the divergence away — the individual lines are where you see it.

What to watch for:

• Divergence onset — when two instruments that normally move together start pulling apart in oscillator space. This can precede a catch-up move (the lagging instrument follows the leader) or a breakdown in correlation.

• Magnitude difference — one instrument reading +60 while the other reads +20 means both are above their baselines, but one is stretched much further. This can indicate relative strength or relative overextension.

What this pattern cannot tell you: ATR normalization makes the scale comparable. It does not make the instruments comparable in any deeper sense. A +50 reading on Bitcoin and a +50 reading on a treasury bond can represent a similar place on the transformed, sensitivity-adjusted scale while still carrying completely different trading implications, catalysts, and behavior. The normalization solves the math problem of comparing distances. It does not solve the judgment problem of deciding whether two instruments should be compared.

Weighted priority stack

What it is: A multi-slot configuration where you deliberately assign higher blend weights to the timeframes or tickers you trust most for the current market condition.

How it works: The blend is a weighted average. By adjusting the weights, you shift the composite toward the readings that matter most to your current thesis. In a trending environment, you might weight the longer-timeframe slot more heavily because the trend is the dominant force. In a ranging environment, you might weight shorter timeframes because the range-bound dynamics play out faster.

How to configure it:

• Start from your confluence setup (three or more timeframes).

• Increase the weight on the timeframe that reflects your current market thesis.

• You do not need to reduce the other weights to compensate — weights auto-normalize. Just increase the one you want to emphasize.

When to adjust the weights:

• When market conditions change. A trending market rewards heavier weighting on the timeframe where the trend is most visible. A choppy market rewards lighter weighting on the longer timeframe (which may be lagging through chop) and heavier weighting on shorter timeframes that track the range dynamics.

• When you have a conviction about which timeframe is most relevant right now. This is a judgment call, not a formula.

The tradeoff: Heavier weighting on one slot means the blend is less diversified across timeframes. If that slot is wrong (e.g., the longer-timeframe reading is lagging through a genuine trend change), the blend will be slow to catch up. The more you concentrate weight, the more you are betting on one timeframe's reading being the right one.

Hidden-slot alignment filter

What it is: A slot that is enabled but has its plot hidden and its weight set to 0. It computes, fires alerts, and participates in the alignment check — but it does not draw a line and does not affect the blended composite.

How it works: You get the slot's per-slot alerts (bullish, bearish, regime flip) and the slot counts in the "All Slots Bullish/Bearish" alignment alerts. But the blended lines are not influenced by this slot's reading. It is a filter that runs alongside the main view without distorting it.

When to use it: When you want to be notified about a condition on a specific timeframe or ticker without that condition changing your primary composite. For example: your main analysis uses three slots on your primary instrument, but you also want an alert when a market benchmark (SPY, BTC) flips regime on the daily timeframe. Adding that as a hidden, zero-weight slot gives you the alert without pulling the blend toward that instrument's reading.

How to configure it:

• Enable the slot.

• Check "Hide Plot."

• Set Blended Weight to 0.

• Configure the timeframe, ticker, MA type, and length to match the condition you want to monitor.

What to watch for: If you later want this slot to influence the blend, you need to both increase the weight and (optionally) unhide the plot. Just unhiding the plot will draw the line but will not add it to the blend if the weight is still 0.

Progressive build-up

What it is: A configuration approach, not a trading pattern. Start with one or two slots. Get comfortable with how the oscillator reads for those specific configurations. Add slots only when you have a specific reason.

Why it matters: The ten-slot architecture is there for people who need it. It is not there to encourage you to use all ten slots from day one. Each additional slot adds a reading, contributes to the blend (if weighted), and potentially changes the composite character. If you do not know why a slot is there, it is adding noise, not signal.

How to do it:

1. Start with one slot on your chart timeframe. Understand how the oscillator reads for your primary instrument and MA choice.

2. Add a second slot on a higher timeframe. Watch how the two interact — when they agree, when they disagree, what happens when the shorter one flips first.

3. Add additional slots only when you have a specific question they answer: "What is the daily distance-from-baseline reading doing?" or "How does QQQ compare to SPY right now?"

4. If you find yourself enabling a slot without being able to articulate what it adds, leave it disabled.

Anti-patterns

These are the configurations and behaviors most likely to produce confusion or false confidence. They are not edge cases — they are the things people actually do with multi-slot oscillators.

Slot overload without intent

What it looks like: Five or more slots enabled with random timeframes and tickers, hoping the blend "figures it out." The settings panel is full of configurations that the user could not explain if asked. It usually starts innocently — you add one more slot because it seems like more data should help. Then another. Then you adjust the timeframes on two of them without a clear reason.

Why it fails: More inputs do not improve the reading if the inputs are not chosen for a reason. A blend of six unrelated readings produces a composite that means nothing in particular — it is the average of six different questions, none of which the user actually asked. The oscillator will show numbers. Those numbers will look like information. They are not. And the worst part: the blended line still moves smoothly, still changes color at regime flips, still fires alerts. Nothing about its behavior signals that the underlying configuration is incoherent. The tool runs exactly the same whether your slot choices are brilliant or arbitrary.

The fix: Disable every slot. Then enable them one at a time, with a specific purpose for each one. If you cannot name the purpose, the slot stays off.

Repaint roulette

What it looks like: Disabling On Bar Close on multiple slots to get "faster" readings without keeping track of which slots are provisional. The user watches the blended line and makes decisions based on it, not realizing that 60% of the blend's data can change when the higher-timeframe bars close. It feels like you are getting a performance upgrade — faster updates, a more responsive composite. The chart certainly looks that way.

Why it fails: The blended line does not distinguish between confirmed and provisional data. It just shows the weighted average. If half your slots are live and half are confirmed, the blend is a mix of things-that-are-true and things-that-might-become-true. Decisions made on this mix carry a risk the user cannot see. The most common version of this failure: you see a strong blended reading, feel confident, take a position, and then watch the reading shift when the HTF bars close. The blend did not lie. You just could not tell which part of it was finished and which part was still a draft.

The fix: Either keep all slots on confirmed data (On Bar Close ON), or maintain a clear mental model of which slots are provisional and how much weight they carry in the blend. If you cannot remember which slots are live, turn them all back to confirmed. The few seconds of latency you gain from live mode are almost never worth the confusion of not knowing what you are looking at.

Overbought/oversold as entry triggers

What it looks like: Treating the overbought or oversold level crossings as buy/sell signals. "The oscillator crossed above 70, so I should sell." Or the inverse: "It crossed below -70, so I should buy."

Why it fails: The overbought and oversold levels indicate how far the composite sits from baseline in ATR terms. They are position markers, not reversal signals. In a strong trend, the oscillator can stay above the overbought level for hours, days, or weeks. Acting on the level crossing alone — without confirming a regime shift, a return toward zero, or supporting evidence from other analysis — is a classic oscillator misuse pattern that produces losses during trending conditions.

The fix: Use overbought/oversold levels as context, not as triggers. "The oscillator is at +85" tells you the composite is stretched. What you do with that information depends on whether the trend is exhausting, accelerating, or ranging. The level alone does not answer that question.

Ignoring blend-vs-slot divergence

What it looks like: Watching only the blended Fast and Slow lines while individual slots are fighting each other. The blend shows +10 and looks calm. The user reads "slightly bullish" and moves on. They do not notice that Slot 01 is at +80 and Slot 03 is at -60.

Why it fails: The blend compresses disagreement into a single number. A calm blend can mean either "all slots are calm" or "some slots are strongly bullish and others are strongly bearish and the average happens to be near zero." These are fundamentally different market conditions that look identical in the blended reading.

The fix: When the blended reading looks unremarkable — especially near zero or near the midline — check the individual slot lines. If they are clustered near the same value, the composite is genuine. If they are spread far apart, the composite is masking a conflict that is worth paying attention to.

Ultra-short Slow with frequent regime flips

What it looks like: Setting Slow Length to 1 or 2 on multiple slots. Every minor oscillation in the Fast line triggers a regime flip. The regime colors flicker constantly. Per-slot regime alerts fire many times per session.

Why it fails: With a Slow Length of 1, the Slow line is nearly identical to the Fast line. Every small wiggle in Fast crosses Slow and triggers a regime change. The regime concept — which is meant to capture meaningful shifts in the Fast/Slow relationship — becomes noise. The alerts become a stream of notifications that carry no useful information.

The fix: The default Slow Length of 3 (with EMA) provides a reasonable balance between responsiveness and stability. If you find regime flips too frequent, increase the Slow Length. If you need faster regime detection, consider 2 rather than 1 — even one bar of smoothing makes a noticeable difference in flip frequency.

Thinking about slot configuration

The patterns above give you concrete starting points. But the deeper skill is learning to think about slot configuration as a set of deliberate choices rather than a form to fill out.

Each slot answers one question: "How far is [this price source] from [this MA baseline] on [this timeframe] for [this instrument]?" The blend answers: "What is the weighted average of those answers?"

Good configurations start with good questions. Before enabling a slot, ask yourself:

1. What am I trying to see? A specific timeframe's baseline-distance reading? A comparison between instruments? A structural baseline?

2. Why does this slot need to be separate from the others? If two slots use the same timeframe, ticker, and MA type, one of them is redundant.

3. How should this slot relate to the blend? Should it contribute (give it weight) or just stand alone (weight = 0)?

4. Does this slot need to be confirmed or can it be live? Most of the time, confirmed is the right default. Only allow live data when you have a specific reason.

If you can answer these four questions for every enabled slot, your configuration is deliberate. If you cannot, you are guessing — and the oscillator will look like it is telling you something when it is really just averaging your guesses.

The temptation with a ten-slot tool is to fill the slots. Resist that. The best configurations we have seen are often three or four slots with clear roles and weights that reflect actual conviction. The unused slots are not wasted capacity — they are discipline. Every empty slot is a question you decided not to ask, and that restraint is part of what makes the questions you did ask worth answering.