Settings

Written By Axiom Admin

Last updated 22 days ago

Settings

This page is the working reference for every knob Axiom Stoch Osc CTX exposes. It is the longest page in the pack, and on purpose — the wider workbench is the entire reason this trim exists, and a too-short settings page would be a betrayal of what you came here for. Use it two ways. On a first read, follow the order below to understand the shape of a single slot, then fan out into the sections that govern the pack as a whole. On later visits, jump to the section that matches the knob you have open in the dialog.

A principle that sits behind every recommendation here: there are no best settings. The defaults ship restrained on purpose, as a legal starting point. Every adjustment is a tradeoff. This page names, beside each knob, what you are giving up when you move it — not which direction to move it.

How to read this page

The indicator groups its inputs into five regions:

A per-slot region, repeated ten times — once for each of Stoch 01 through Stoch 10. Each slot carries the same shape of inputs. The values differ; the shape does not.
A per-slot Power User region, also repeated ten times. Each slot has its own On Bar Close switch and its own MA-family tuning knobs.
A General Settings region with the overbought and oversold guides.
A Display Settings region with the blended plot switches.
A Master Smoothing region, optional, that applies one additional pass to the blended pair after everything else.

This page teaches the per-slot shape once — in depth — and then fans out. Slots 02 through 10 carry the same semantics as slot 01; their defaults differ by design and are listed under "Per-slot defaults at a glance" below so you do not have to read ten near-identical sections.

The shape of a single slot

Every slot is a complete, self-contained stochastic calculator. It computes one K line, one D line (drawn indirectly, used for coloring the slot and feeding the blend), and contributes its weighted K and D to the composite.

Enable Stoch NN

Type: boolean
Defaults: slots 01–03 are true. Slots 04–10 are false.
What it does: turns the entire slot on or off. A disabled slot runs no computation, draws nothing, contributes nothing to the blend, and fires no alerts. It is as if the slot were not in the product.
Tradeoff: more slots enabled means more working memory and more correlated observations. An unused slot should be disabled rather than merely zero-weighted; the disabled state is cheaper in every sense.
Misuse to name out loud: enabling all ten slots at equal weight on a single symbol and treating the blend as "ten timeframes agreeing." That is not what the tool produces; the blend is a weighted average of correlated smoothings of the same price series, and agreement among those smoothings is mostly a statement about the smoothings, not about the price. See Limitations and Trust Boundaries for the full framing.

Hide Stoch NN Plot

Type: boolean
Default: false.
What it does: removes the slot's K line from the pane while keeping the slot fully active — it still computes, still contributes its weight to the blend, and still fires its own alerts.
Tradeoff: less visual clutter at the cost of losing that slot's bright/faded state cue. Useful for a slot you want in the composite for steering but not on the pane for viewing.
When to reach for it: a visually heavy pane where one slot's color is distracting but its contribution to the weighting matters. Or a deliberate "alert-only slot" configuration — see Workflows.

Source

Type: price series input
Default: close.
What it does: the series the slot feeds into its stochastic calculation. Any of the built-in TradingView series (close, open, high, low, hl2, hlc3, ohlc4, and so on) is valid, as is any plotted series from another indicator on the chart.
Cross-asset interaction: when Optional Ticker is set, the slot reads the source from that other symbol, not from the chart symbol. Along with the source, the slot's stochastic also reads the high and low from the other symbol. See the next section.
Tradeoff: non-close sources can be more or less noisy depending on the series. Changing the source changes the slot's stochastic meaningfully; do not treat source as a cosmetic setting.

TimeFrame

Type: string (TradingView timeframe format)
Defaults: slot 01 "5", slot 02 "15", slot 03 "60", slots 04–10 empty.
What it does: the higher timeframe the slot evaluates its stochastic on. An empty string means the slot inherits the chart's current timeframe; that is fine for a "same-timeframe" slot.
Hard rule: the slot timeframe must be greater than or equal to the chart timeframe. If it is lower — for example, slot 01 set to "1" while the chart is on 5-minute — the script raises a runtime error that names the offending slot number. The error is deliberate; a lower-than-chart slot cannot meaningfully return a higher-timeframe value from below the chart bar. Either raise the slot timeframe or drop the chart.
Timing consequence: higher-timeframe slots are subject to the On Bar Close? choice under Power User. See MTF and Repainting for the full treatment.
Tradeoff: a slower slot is calmer and less noisy, at the cost of latency against the chart. A faster slot is responsive to the chart's own moves, at the cost of adding little beyond what the chart itself already shows.

K Length

Type: integer, minimum 1
Default: 14.
What it does: the lookback length for the stochastic's %K computation — how far back the slot measures where its source sits inside the high-low range of the chosen timeframe and symbol.
Tradeoff: shorter lengths are more reactive and produce more K/D crosses. Longer lengths are smoother and hold their reading longer, at the cost of lag against turns. A K Length of 14 is the textbook reference point and a reasonable place to start.
Edge case: pushing K Length to an extreme can produce a slot that hits the 0..100 frame frequently, or a slot whose K rarely leaves the middle of the range. Use the extreme as a verification tool, not as a working configuration; see the clamp verification under Visuals and Logic.

K Smoothing

Type: integer, minimum 1
Default: 3.
What it does: the length used by the K MA (set by K Type below) to smooth raw %K into the slot's plotted K line.
Tradeoff: more smoothing on K reduces high-frequency whip at the cost of making K/D crosses later. K Smoothing of 3 is the textbook "slow stochastic" setting; 1 gives you "fast stochastic" behavior.

K Type

Type: enumerated MA family
Default: SMA.
Available types: SMA, EMA, ALMA, KAMA, FRAMA, Jurik, Laguerre, VAMA, and the related variants the shared Axiom MA library exposes.
What it does: picks the moving-average family used for the K smoothing pass.
Tradeoff: different MA families have different latency and overshoot profiles. SMA is the textbook reference. EMA is faster and more common. ALMA, KAMA, FRAMA, Jurik, Laguerre, and VAMA trade responsiveness for noise rejection in different ways, and each exposes its own tuning under Power User.
Important framing: changing K Type without a reason makes the slot less interpretable, not more. Stay on SMA until you can say out loud what you are trying to trade off by moving.

D Length

Type: integer, minimum 1
Default: 3.
What it does: the length used by the D MA (set by D Type below) to smooth the slot's K line into the slot's D line. The slot's D is not plotted directly; it drives the slot's bright-versus-faded color state and feeds the blended D.
Tradeoff: same as K Smoothing. A longer D smooths the slot's color transitions and reduces the cosmetic frequency of K/D crosses.

D Type

Type: enumerated MA family — same list as K Type
Default: SMA.
What it does: picks the moving-average family used for the D smoothing pass, independent of K Type.
Why two MAs per slot: separating the K and D MAs lets you study how different smoothing families interact inside one slot. A fast K with a slow D will produce more K/D crosses than an MA-matched pair, but those crosses do not represent "more signal" — they represent the smoothing mismatch. Name that out loud if you change the defaults.

Line Width

Type: integer, minimum 1
Default: 2.
What it does: the pixel width of the slot's K plot. Cosmetic, but not trivial — a pane with ten slots drawn at width 4 is hard to read.
Practical guidance: leave at 2 unless a slot is the subject of a specific routine and you want it visually dominant.

Blended Weight

Type: float, minimum 0, two-decimal step
Defaults: slot 01 33.3, slot 02 33.3, slot 03 33.3, slots 04–10 0.
What it does: the slot's contribution to the weighted average that produces the blended K and blended D. The weights do not need to sum to 100; they are treated as relative weights. A slot with weight 0 is skipped entirely in the average.
Tradeoff: a heavier-weighted slot dominates the blend. A slot at weight zero is invisible to the blend but still drawn (unless Hide Plot is on) and still alerting. That asymmetry is useful.
Design pattern: "slot present for viewing, absent from blend." Enable the slot, set weight to zero. The slot K line draws and the slot's state alerts fire, while the blend ignores the slot entirely. See Workflows.
Misuse to name out loud: splitting weights evenly across many slots on the same symbol and reading the blend as a "consensus." It is not a consensus; it is an autocorrelated average. See Limitations and Trust Boundaries.

Optional Ticker

Type: symbol input (string)
Default: empty.
What it does: reroutes the slot to a different symbol entirely. When this is set, the slot pulls all of its stochastic context — source, high, and low — from the chosen symbol, not from the chart symbol.
Honest framing this page will repeat on purpose: an Optional Ticker turns the slot into a stochastic of another market. The blend then contains a mixture of stochastic reads across different markets. This is a category change, not a tuning change. The first time you set one, write down what you are trying to learn from it, because the pane is no longer "a stochastic of this chart" — it is "a stochastic of this chart, blended with a stochastic of that other market." The full treatment lives on MTF and Repainting.
Tradeoff: context from a leading or correlated instrument (an index, a sector, a paired asset) versus the complexity of reasoning about two markets at once.

Per-slot defaults at a glance

Slots 02 through 10 share the per-slot shape above. Their defaults differ as follows.

Slot 01: enabled, timeframe "5", weight 33.3.
Slot 02: enabled, timeframe "15", weight 33.3.
Slot 03: enabled, timeframe "60", weight 33.3.
Slots 04 through 10: disabled, timeframe empty, weight 0.

Every other per-slot input takes the same default regardless of slot: source close, K Length 14, K Smoothing 3, K Type SMA, D Length 3, D Type SMA, Line Width 2, Optional Ticker empty, On Bar Close? on.

Slot 10 has one intentional visual exception noted below.

The per-slot Power User region

Each slot exposes a Power User region that holds the repaint switch and the MA-family tuning knobs. Treat the Power User region as per-slot, not global — two slots can have different On Bar Close? choices and different tunings, and that is normal for this trim.

On Bar Close?

Type: boolean
Default: true.
What it does: when true, the slot returns the previous confirmed higher-timeframe bar's K and D. When false, the slot returns the live higher-timeframe bar's K and D.
Scope: the switch is per slot and is shared by the slot's K and D. They cannot disagree on it.
Tradeoff: true gives you a non-repainting slot at the cost of latency; the slot is "behind" the live bar. false gives you responsiveness at the cost of repaint — the slot's value within a live higher-timeframe bar will change as that bar forms.
Framing to keep in mind: this switch is not a fix for repainting; it relocates the repainting tradeoff to a place where you can see it and choose sides per slot. MTF and Repainting is the page for the full treatment.
Misuse to name: turning the switch off across all slots to chase responsiveness, then acting on slot values that are still moving because their higher-timeframe bars have not closed.

MA-family tuning knobs (K side, repeated symmetrically on the D side)

K ALMA Floor Offset?, K ALMA Offset, K ALMA Sigma: tune ALMA when K Type is set to ALMA. Ignored otherwise.
K KAMA/FRAMA Fast, K KAMA/FRAMA Slow: tune KAMA and FRAMA when K Type is set to one of those families. Ignored otherwise.
K Jurik Phase, K Jurik Power: tune Jurik when K Type is set to Jurik. Ignored otherwise.
K Laguerre Alpha: tunes Laguerre when K Type is set to Laguerre. Ignored otherwise.
K VAMA Vol Length: tunes VAMA when K Type is set to VAMA. Ignored otherwise.

Each of the above reappears with a D prefix for the D line's MA. The K and D tunings are independent of each other even when K Type and D Type are set to the same family.

This page intentionally does not disclose the parameter semantics of any MA family. That detail lives inside the shared Axiom MA library and is not part of this product's publishable surface. If you are tuning an MA family here, you are telling the slot how to smooth — the workbench is yours to calibrate — but a parameter cheat sheet would do more harm than good, because the right value for any of these knobs is a function of your chart, your timeframe, and what you are trying to read, not a number someone else can hand you. Change one knob at a time and watch what the slot's K line actually does on your chart. If that feels slow, it is because tuning an MA responsibly is slow work.

General Settings

Overbought Level

Type: float, step 0.1
Default: 80.
What it does: draws a dashed gray horizontal guide at this level on the pane.
What it does not do: feed any alert condition, color any line, or otherwise affect the indicator's output. The 80 line is a guide — a visual landmark — not a trigger.
Tradeoff: the textbook 80 is sensible for many instruments and timeframes, but not all. If you work a highly trending instrument where stochastic rarely leaves the 60–100 range, raising the overbought guide clarifies the readable zone. If you work a heavily compressed instrument, lowering the overbought guide does the same.

Oversold Level

Type: float, step 0.1
Default: 20.
What it does / does not do: symmetric to Overbought. Visual guide only, no alert.

Display Settings

Plot Blended K/D

Type: boolean
Default: true.
What it does: shows the blended K and blended D on the pane. When off, the blended pair is still computed (and still affects the blended alerts) but is not drawn.
When to turn it off: a pane with many enabled slots where the blended pair is adding visual weight you do not need. The alerts still work.

Blended Line Width

Type: integer, minimum 1
Default: 3.
What it does: pixel width of the blended K and blended D.
Practical guidance: the blended pair is the composite you authored; a slightly heavier line than the slots is appropriate. Dropping to 2 makes the pane quieter when ten slots are active.

Master Smoothing

This region is off by default. Turn it on when — and only when — you have formed a reliable expectation of the raw blend's behavior, and you want a calmer composite for regime-style reading.

Enable Master Smoothing

Type: boolean
Default: false.
What it does: when on, applies one additional smoothing pass to the blended K and blended D after the weighted average and the clamp, then clamps again.
Tradeoff: calmer composite at the cost of responsiveness. The next subsection lays out the real risk.

Master MA Type

Type: enumerated MA family
Default: EMA.
What it does: picks the family used for the master-smoothing pass. Full list identical to K Type and D Type.

Master Length

Type: integer, minimum 1
Default: 3.
What it does: length of the master-smoothing pass. At length 1 most MA families pass the input through unchanged; long lengths compress the blended pair's movement meaningfully.
Tradeoff worth naming twice: a long master length can hold the blended pair on one side of its midline for several bars after the underlying slot lines have already turned. During a regime change, that delay is not cosmetic — it can hide the event you care about. If you run a long master length, cross-check against the raw slot lines every time the blend feels sluggish.

Master MA-family tuning knobs

The same list of tuning knobs that each slot exposes, but applied once to the master pass: Master ALMA Floor Offset?, Master ALMA Offset, Master ALMA Sigma, Master KAMA/FRAMA Fast, Master KAMA/FRAMA Slow, Master Jurik Phase, Master Jurik Power, Master Laguerre Alpha, Master VAMA Vol Length. These behave exactly like the per-slot tunings; each is only consulted when Master MA Type is set to the family it tunes.

A compounding note worth holding: by the time master smoothing is active, the blended pair has already been through K smoothing and D smoothing inside every slot. Master smoothing adds a fourth operation on top. The informational distance between raw price movement and the master-smoothed blended K is larger than most readers realize. This is not a defect; it is what you chose by stacking smoothings. Treat a master-smoothed composite as a different series from the raw blend, not as "the same line, just cleaner."

Misuse prevention — the settings most likely to hurt you

The knobs below are not dangerous by themselves. They become dangerous in specific combinations, and the damage usually shows up as a trade taken on a reading the pane could not actually support. Name these combinations before you ship a configuration, and again when the configuration has been running a few weeks and you have stopped questioning it.

All ten slots enabled at equal weight on one symbol. The blend is an autocorrelated average, not a consensus. Ten overlapping windows on the same price series agree most of the time by construction; the "all ten agree" moments will feel like confirmation and provide very little of it. If you want a multi-timeframe view, two or three well-chosen slots usually do more honest work than ten.
On Bar Close? off across every slot. Responsiveness has a real cost: every higher-timeframe slot now moves inside its own live bar, and acting on a value that will resettle before the bar closes is the exact behavior the switch exists to make visible. A reader who cannot tell you whether a specific slot's current value will still be that value in ten minutes has no business acting on it.
Optional Ticker set on a heavily-weighted slot, read as if the blend still describes the chart symbol. It does not. A cross-asset slot makes the blend a multi-market composite — usefully, if the mixture is deliberate, and dangerously if it is not. Either accept the multi-market reading and use it on purpose, or set the weight to zero so the slot contributes visually without contaminating the composite.
Long master smoothing, then alerts on blended state. The blended state can sit on one side of its midline for bars after the underlying slots have turned when the master pass is long. The alerts stay silent through exactly the transition you were waiting for. They are not malfunctioning; the smoothing is holding the state. If you run long master smoothing, wire alerts on the per-slot states and keep the blended pair as context rather than trigger.
Aggressive K MA with a short K Length. Some MA families (ALMA with an aggressive sigma, short Laguerre alpha) produce a K line that rarely reaches 80 or 20 even in strong moves. The 80/20 guides become decorative and any overbought/oversold framing you mentally carry stops applying. The slot is still honest; the reader's mental map of it is now miscalibrated.
Slot timeframe below chart timeframe. The runtime error names the slot. Either raise the slot or drop the chart. The error is the product refusing to draw a value it cannot defend; treat it that way and the fix becomes obvious.
Configurations you inherited or screenshotted from another trader. A settings file that works for one reader on one market with one trading horizon is not a default for yours. Before running someone else's configuration live, read each non-default knob back through this page and re-earn the reason for the choice. If you cannot, either re-derive the reason or reset the knob.

Every combination above has a legitimate cousin. Enabling all ten slots across a cross-asset watchlist is a sensible use of the surface; enabling them on one symbol at equal weight is not. Running On Bar Close? off on one slot for responsive reading is a sensible choice; running it off on every slot is usually not. Master smoothing at a short length alongside cross-checking the raw slots is a sensible calibration; master smoothing at a long length with alerts wired to the blended state is one of the cleanest ways to miss the event you were watching for. The misuse is not in any single setting. The misuse is in the combination, and the combination is what the reader has to audit.

A note about Slot 10

Slot 10 in particular uses a heavier faded-state transparency than slots 01 through 09. This is an intentional choice — the slot's base color is white, and at the standard faded-state transparency it would be hard to see against a light theme or hard to distinguish from the background against a dark one. Slot 10's heavier fade keeps the white line legible in both environments. Visuals and Logic names this as intentional demphasis so you do not go looking for a setting to "fix" it.

How to calibrate a knob without guessing

Most calibration mistakes on this page come from changing more than one thing at a time. The discipline that prevents them is unglamorous and reliable:

Isolate the change. Move one knob. Leave everything else alone, including other slots.
Name what you expect to see before you look. "K Length from 14 to 9 should make slot 02 cross its D more often, and the crosses should occur a bar or two earlier." If you cannot state the expectation, you are not ready to change the knob.
Watch through at least one state transition on the live chart. A full K/D cycle if possible. Stochastic on a quiet bar tells you nothing; stochastic during a real turn tells you whether the change did what you expected.
Compare against the prior configuration. If the change delivered the expected difference and you can still defend the reading, keep it. If the difference is smaller than you expected, the setting probably was not doing much for you before either. If the difference is larger or in the opposite direction, you have learned something about the knob that the manual could not have taught you.
Only then stack another change. Two changes at once is twice the reading work and four times the chance of drawing a wrong conclusion about which change did what.

This discipline is slow. Slow is the cost of learning a workbench well enough to trust it under pressure. Skipping it is how readers end up with configurations they cannot defend a month later.

If a setting did not get the treatment you needed on this page, the most likely destinations for the gap are Visuals and Logic for what a change looks like on the pane, MTF and Repainting for anything involving timeframe or repaint, Limitations and Trust Boundaries for the trust framing of a risky combination, and For the Geeks for the mental model of the pipeline as a whole.