Settings

Settings

This page documents every configurable setting in Axiom MA Osc Lite, organized in the same order they appear in TradingView's settings panel.

The point is not just to tell you what each setting does. It is to help you understand the tradeoffs well enough that you change settings on purpose rather than by trial and error. A setting changed without understanding its consequence is not configuration — it is a variable you introduced without knowing what it controls.

Per-slot settings

The indicator has three independent MA slots: MA 01, MA 02, and MA 03. Each slot has an identical set of settings. The defaults differ only in timeframe (5m / 15m / 60m) and color. Everything explained below applies to all three slots.

Enable

Disabling a slot is the cleanest way to remove it from everything. Unlike setting the weight to zero (which removes it from the blend but keeps the plot and per-slot alerts active), disabling truly shuts the slot off.

If you disable all three slots while the blended plot is on, the blended lines will disappear because there is nothing to average. This is expected.

Hide Plot

This is the setting most likely to create confusion down the road. When you hide a slot's plot, the line disappears but the slot's influence on the blended oscillator does not. If the hidden slot is in strong bullish territory while your visible slots are bearish, the blended line will be pulled upward by an invisible force. The blend will not match what you see.

When to use it: When you want a cleaner pane — fewer lines, less visual clutter — but you still want all three slots contributing to the blended reading and all three slots available for alerts.

When not to use it: When you are trying to understand why the blended line is behaving unexpectedly. If you cannot explain the blend from the visible lines alone, un-hide all slots and check.

Source

Most users should leave this on Close. Changing to HLC3, OHLC4, or other composites will produce a slightly different distance measurement because the source smooths intrabar price action. The ATR normalization still uses the standard ATR calculation on the slot's symbol regardless of the source setting — so you are measuring a different numerator (composite price minus MA) against the same denominator (ATR).

Timeframe

Hard constraint: The slot's timeframe must be equal to or higher than the chart timeframe. If you set a slot to 5m and your chart is on 15m, the indicator throws a runtime error and will not load. This is enforced by the script, not by TradingView.

Choosing timeframes: Pick timeframes that represent structurally meaningful scales in your trading process. If you are an intraday trader working on the 1-minute chart, 5m / 15m / 60m gives you short-term, intermediate, and session-level context. If you are a swing trader on the 15-minute chart, you might use 15m / 60m / 240m (or Daily).

The important thing is that each slot's timeframe gives you a different perspective on distance from the average. If you set all three to the same timeframe (e.g., 15m / 15m / 15m) with different MA lengths, you get three measurements of the same timeframe's data — which is not multi-timeframe analysis. You lose the tool's main architectural advantage. See Workflows — Anti-patterns for more on this.

Setting timeframes very far apart (e.g., 1m / 15m / Weekly) can make the longest slot nearly unresponsive on an intraday chart. The weekly MA moves so slowly relative to the 1-minute chart that the Fast line for that slot may appear flat for most of the session.

Length

Longer lengths produce a smoother baseline MA that responds more slowly to price changes. The oscillator's Fast line will move more gradually because the MA it measures distance from is more stable. Shorter lengths produce a more reactive baseline — the Fast line will show more movement because the MA is chasing price more closely.

Very short lengths on high timeframes (e.g., Length 3 on the 60m slot) produce an erratic baseline MA that jumps with each bar. The oscillator Fast line becomes noisy because it is measuring distance from a baseline that itself is volatile. Very long lengths (e.g., Length 200 on a 5m slot) make the baseline MA nearly static over short periods, so the Fast line mostly tracks price movement directly (since the MA side of the distance equation barely moves).

MA Type

Available types include EMA, SMA, WMA, VWMA, ALMA, and others supported by the library. The choice affects how the baseline MA weights recent versus older data:

• SMA gives equal weight to all bars in the lookback period

• EMA gives exponentially more weight to recent bars, making the baseline more responsive

• WMA gives linearly more weight to recent bars

• ALMA uses a Gaussian distribution centered at a configurable offset — requires the PU ALMA settings (Offset, Sigma, Floor)

Changing the baseline MA type changes how the oscillator's Fast line responds to price. An EMA baseline reacts faster to recent price moves than an SMA baseline of the same length, which means the distance between price and the MA tends to be smaller with EMA — and the oscillator's readings tend to stay closer to zero. With SMA, the MA lags more, so the oscillator tends to show larger distance readings.

If you use ALMA, note that the ALMA parameters (Offset, Sigma, Floor) are global — they apply to every slot that uses ALMA, both for the baseline and for the Slow smoothing. You cannot set different ALMA parameters per slot in the Lite version.

Slow Length

The Slow line is not plotted individually per slot. Its only visible effect is the color of the Fast line (regime color) and its contribution to the Blended Slow line.

A Slow Length of 1 makes Slow nearly identical to Fast, which means regime flips happen on almost every bar — the color will flicker constantly, and the "Is Bullish" / "Is Bearish" alerts will toggle rapidly. This is rarely useful.

A Slow Length of 3 (the default) keeps Slow close to Fast but with enough separation to produce meaningful regime transitions. The regime flips when the distance score turns relative to its own recent average.

Longer Slow Lengths (10, 20, or more) create persistent regime states. The slot will stay in one regime for extended periods, ignoring short-term oscillations. This can be useful for capturing structural trend shifts, but it also means regime flips come late — the underlying momentum direction may have already changed before the Slow line catches up.

Slow Type

Same library and same type options as MA Type. The choice affects how the Slow line responds to changes in Fast: EMA is responsive, SMA is equal-weighted, ALMA uses the global ALMA parameters. If you use ALMA here, the global PU ALMA settings apply.

Line Width

No functional impact.

Blended Weight

The weights do not need to add up to 100 or any specific number. What matters is the ratio. Setting all three to 10 produces the same blend as setting all three to 100.

Setting weight to 0 is a special case. It removes the slot from the blend, but it does not remove the slot from alignment logic. If the slot stays enabled and has data, it still plots its Fast line (if not hidden), still fires its own per-slot alerts, and still participates in the "All Slots Bullish/Bearish" checks. This is useful when you want one slot to stop pulling the blended oscillator while still remaining part of the broader state of the script.

Extreme weight imbalance (e.g., 90 / 5 / 5) makes the blend effectively a single-slot reading with minor corrections from the other two. The blend will look and behave almost identically to the heavy slot's Fast line. If that is your intention, it is a valid setup. Just be aware that the blend label suggests a multi-timeframe composite, but the reality is a single-timeframe reading wearing a blend label.

Optional Ticker

This is what enables cross-ticker comparison. Because the oscillator normalizes distance by ATR and then bounds it with tanh, readings from different instruments sit on the same −100 to +100 scale.

Critical understanding: When you set an Optional Ticker, the slot's readings are about that ticker, not about the chart symbol. If your chart shows AAPL and you set slot 03's Optional Ticker to SPY, the slot 03 Fast line tells you how far SPY's price is from SPY's 60m EMA — it says nothing about AAPL's price directly.

When this slot is blended with chart-symbol slots, the blend combines distance readings from different instruments. This is arithmetically valid because the scores are unitless. Whether it is analytically meaningful depends on whether the instruments share some structural relationship. Blending AAPL with SPY can tell you something useful about how the stock's MA-distance compares to the broader market. Blending AAPL with Treasury yields produces a number that has no useful interpretation.

See Multi-Ticker Mixing for practical setup guidance, trust limits, and verification steps. See For the Geeks for a deeper explanation of why the normalization makes cross-ticker blending work mechanically.

General oscillator settings

These settings apply globally — they affect every slot simultaneously.

ATR Length

ATR is the denominator in the oscillator's normalization. A longer ATR length produces a more stable normalization base — the oscillator changes more gradually because the measure of "typical volatility" is smoother and less reactive to short-term spikes. A shorter ATR length makes the normalization more responsive to recent volatility, which can cause the oscillator to swing sharply when ATR contracts or expands.

The practical consequence: if ATR is computed over 14 bars and a sudden volatility event occurs, the ATR increases gradually over the next 14 bars, and the oscillator compression happens gradually. If ATR is computed over 3 bars, the spike reshapes ATR almost immediately, and the oscillator can drop sharply on a single bar as the denominator jumps.

Most users should leave this at 14 unless they have a specific reason to change it. If you want faster adaptation to volatility regime changes, try 7 or 10. If you want the normalization to be very stable and slow-moving, try 21 or higher.

ATR Sensitivity

This is the most consequential tuning parameter in the entire indicator. It does not make the oscillator "more sensitive" or "less sensitive" in a simple, one-dimensional way. It reshapes the entire mapping between actual distance and the bounded display.

How sensitivity works

The oscillator computes the distance between price and the MA, divides by ATR, and then multiplies by the sensitivity value before passing the result to the tanh function. Because tanh approaches ±1 asymptotically, the sensitivity determines how much of the input range gets mapped into the central, high-resolution zone of the oscillator versus how much gets pushed into the saturated, low-resolution zone near the bounds.

What happens at different sensitivity levels

Low sensitivity (0.3 – 0.5): The oscillator uses the center of the range. Readings rarely approach ±100 even during strong trends. The oscillator preserves fine-grained distance differences because most readings fall in the linear portion of the tanh curve where small input changes produce proportional output changes. The cost: the oscillator looks muted. Extreme moves that would pin a higher-sensitivity version at ±100 might only reach ±50 here. The OB/OS thresholds at ±70 may rarely be reached.

Default sensitivity (1.0): The normalized distance is passed directly to tanh without rescaling. For most instruments at moderate volatility, the oscillator uses a reasonable range — spending time in the ±30 to ±80 zone during trending periods and reaching toward ±100 during strong moves. The OB/OS levels at ±70 are reachable but not constantly triggered.

High sensitivity (2.0 – 3.0): The oscillator is pushed toward the bounds aggressively. Moderate distance readings already land near ±80 or ±90. During trends, the oscillator pins at ±100 quickly and stays there. The resolution near center is compressed — the oscillator flips through zero sharply and spends most of its time near the extremes. At very high sensitivity, the oscillator becomes nearly binary: it reads close to +100 when price is above the MA and close to −100 when below, with only brief transitions through the center.

Why this matters

If you are comparing oscillator readings across sessions, across instruments, or across different configurations, the sensitivity setting changes what every number means. A reading of +70 at sensitivity 0.5 represents a larger actual distance (in ATR units) than +70 at sensitivity 2.0. Two charts with different sensitivity settings are not directly comparable even if the numbers look the same.

There is no correct sensitivity value. Different instruments, different timeframes, and different analytical needs call for different settings. The question to ask yourself is: "How do I want the −100 to +100 range to map to the distance my instrument typically travels from its MA?" If the oscillator spends most of its time pinned at the extremes, your sensitivity is too high for this instrument — you are reading a binary switch, not an oscillator. If it never reaches the extremes and the OB/OS thresholds feel unreachable, your sensitivity is too low — you are reading a flat line with occasional bumps.

See For the Geeks — Sensitivity experiment for a hands-on way to see the effect on your own chart.

<!-- TEACHING ASSET: Sensitivity comparison visual — three panels showing the same instrument and time range at sensitivity 0.5, 1.0, and 2.5, annotated to show range usage, time near extremes, and zero-crossing sharpness -->

Overbought Level

This is a line you draw on the oscillator, not a property of the oscillator itself. The oscillator does not behave differently at this level. Moving it changes where the dashed line appears and when the OB alert fires — nothing else.

Do not import RSI threshold intuitions here. RSI's 70 has decades of empirical context in range-bound markets. This oscillator's +70 is a point on a tanh-bounded scale where the relationship between the number and the underlying distance is already nonlinear. Crossing it means the distance from the MA is large enough to reach that mark given your sensitivity setting — it says nothing about whether the distance will grow or shrink from here.

Oversold Level

Same considerations as the overbought level.

Display settings

Plot Blended Fast/Slow

Turning this off hides the blended visual. It does not disable blended alert conditions. The blended regime alerts, zero-cross alerts, OB/OS alerts, and alignment alerts still evaluate and can still fire even with the blended plot hidden. If you want to see only the individual slot lines but still receive blended alerts, you can turn this off safely. But if you turn it off intending to stop blended alerts, you need to manage the alerts in TradingView's alert dialog instead.

Blended Line Width

No functional impact.

Master smoothing settings

Enable Master Smoothing

This creates a double-smoothed composite. The per-slot Slow lines are already smoothed versions of their Fast lines. The blended lines average those. Master smoothing then smooths the blend. The result is a calmer, slower-moving oscillator — but one that can significantly delay regime detection, OB/OS crossings, and zero-line crossings.

When to use it: When the raw blended oscillator is too noisy for your purposes and you want a calmer composite reading that filters out short-term fluctuations. Some traders prefer this for a "big picture" MA-distance pane that only moves on sustained changes.

When to be cautious: When timing matters. The delay compounds: per-slot smoothing lag + blending averaging + master smoothing lag. By the time the master-smoothed blend flips regime, the underlying slots may have already flipped and partially recovered. If you are using alerts on the blended oscillator, master smoothing delays every alert.

Master MA Type

If you select ALMA, the global PU ALMA settings apply.

Master Length

Longer lengths add more smoothing and more lag. A Master Length of 3 with EMA provides mild smoothing — noticeable but not dramatic. A Master Length of 10 or higher starts to flatten the oscillator significantly, and combined with Slow Lengths of 3+ on individual slots, the regime transitions become very slow.

Be careful about stacking long Slow Lengths and long Master Length together. A Slow Length of 10 on each slot plus Master Smoothing with a length of 10 produces an oscillator that barely moves. The distance information is still in there, but by the time it surfaces through three layers of averaging, the moment that mattered has passed.

PU (Power User) settings

These settings are global and affect all slots simultaneously.

On Bar Close?

This is the most important safety setting in the indicator. See MTF & Repainting for a full explanation, verification routine, and residual risk analysis.

When on: The oscillator uses only the last confirmed higher-timeframe bar's values. Historical readings are stable — they match what was available at the time each chart bar closed. The trade-off is that recent bars show slightly lagged values (the previous HTF close, not the current building bar).

When off: The oscillator updates using the current building HTF bar. It feels faster and more responsive. But the lower-timeframe bars inside that HTF block are later backfilled with the finalized HTF value. The oscillator repaints. Any historical comparison — including visually overlaying the oscillator on a strategy's entries and exits — is unreliable because the oscillator values in history include data that was not available in real time.

In the Lite version, this is a single global toggle. There is no per-slot override.

ALMA Floor Offset?

This only matters when at least one slot uses ALMA for its baseline MA or Slow smoothing. Most users should leave this at the default unless they have a specific reason to quantize the ALMA kernel position.

ALMA Offset

Applied globally — every slot that uses ALMA, whether for the baseline MA or for the Slow smoothing, uses this same offset value. You cannot set different offsets per slot.

Higher offset values (closer to 1.0) make the ALMA more responsive to recent data, similar to EMA behavior. Lower offset values (closer to 0) make the ALMA weight older data more heavily, producing a laggier average.

ALMA Sigma

Same global scope as ALMA Offset. Affects all ALMA instances across all slots.

Setting dependency map

Several dependencies exist between settings that are not obvious from the UI. These are the ones most likely to produce behavior you did not expect and cannot explain from the setting you just changed.

Global scope dependencies

• On Bar Close applies to all three slots. You cannot have one slot using confirmed data and another using live data in the Lite version.

• ATR Length applies to all three slots. Every slot's normalization uses the same ATR lookback.

• ATR Sensitivity applies to all three slots. Changing sensitivity reshapes all three oscillator lines simultaneously.

• ALMA Offset, Sigma, and Floor apply globally to every instance of ALMA across all slots — both baseline and Slow MAs.

Hide Plot versus Disable

The critical row is the middle one. A hidden slot is invisible but fully active in the blend and alert system.

Weight of zero

Setting a slot's weight to 0 removes it from the blend. It does not remove it from the alignment count. If the slot is enabled and producing data, it still computes, still plots its Fast line (unless hidden), still fires its per-slot alerts (Is Bullish, Is Bearish, Regime Flip), and still participates in All Slots Bullish/Bearish.

Blended alerts versus blended plot

Turning off "Plot Blended Fast/Slow" removes the visual but does not suppress blended alert conditions. The blended regime, zero-cross, OB/OS, and alignment alerts still evaluate on every confirmed bar. If you want to stop a blended alert from firing, you must delete or disable it in TradingView's alert manager.

Master smoothing scope

Master smoothing applies to the Blended Fast and Blended Slow lines only. It does not affect individual slot Fast lines, their regime colors, or their per-slot alerts. If you enable master smoothing and then look at a per-slot alert firing alongside the blended oscillator, the blended regime may lag behind the individual slot's regime because the blending itself already averages, and the master pass adds further delay.