Troubleshooting

Use this page the way you would use a triage desk — walk in with a symptom, leave with a reasoned next step. The symptoms are grouped into three categories so you learn to recognize them by shape:

• Setup errors — something is wrong with how the indicator is loaded or configured; the tool is refusing to compute, or configured itself in a way that produces nonsense.

• Misreads — the tool is behaving correctly; the reader's interpretation is off.

• Genuine limits — the tool is behaving correctly; the reader has hit a boundary Base does not cross.

If a symptom could fit more than one category, the category it is filed under is the category you should check first. The larger habit the page is training is triage, not memorization — the specific symptoms change as you get more time on the pane, but the three categories hold, and knowing which one you are in is usually more than half of the fix.

Setup errors

Runtime error: Stoch 0N timeframe cannot be lower than the chart timeframe.

What you are seeing. TradingView shows a runtime error with a message naming a specific slot. The pane does not draw.

Likely cause. One of the slots has a TimeFrame: value below the current chart timeframe. For example, slot 02 set to 1 on a 5m chart.

Quick check. Read the slot name in the error message. Open the input panel. Find that slot's TimeFrame: and compare to the chart's timeframe.

Fix. Raise the slot's timeframe to the chart's timeframe or higher, or lower the chart's timeframe below the slot's. The rule is one-way: slot TF ≥ chart TF.

See MTF and Repainting for the rule in full.

All three slot lines read almost identically; the blend collapses onto one read

What you are seeing. On a particular chart timeframe, the three slot K lines track each other tightly and the blend sits on top of them. Alignment fires almost continuously.

Likely cause. Near-duplicate slot configuration. Multiple slots are asking almost the same question — same or very close timeframes, same source, same K length, same smoothing, same MA families — so their K lines naturally travel together.

Quick check. Open the input panel. Compare the three slot timeframes, sources, K lengths, smoothing lengths, and MA families. If you switched from a 1m chart to a 1H chart with the default 5 / 15 / 60 stack, expect a runtime error for slots 01 and 02 instead; the source does not silently collapse lower-timeframe slots into the chart.

Fix. Rebuild the stack so every slot timeframe is at or above the chart timeframe and the slots are meaningfully differentiated. On a 1H chart, a useful stack might be 1H / 4H / 1D. On a 5m chart, 5m / 15m / 60m works.

Slot line missing on a fresh load

What you are seeing. One or more slot lines are not drawing during the first minutes after load.

Likely cause. Warm-up. ta.stoch with K Length 14 plus two smoothing passes needs history to produce values. On a 60m slot, the required history can span several hours of 1m chart bars.

Quick check. Wait. Check after a few bars.

Fix. If the chart has less history than the slot needs, give it time or reload later. The slot is not broken. It is waiting for bars.

Blend not drawing

What you are seeing. The three slot K lines are visible. The blended K, blended D, and tinted fill are not.

Likely cause 1. Plot Blended K/D is off. Display only.

Quick check. Open the input panel. Check the Plot Blended K/D boolean in the Display group.

Fix. Turn it on. Note: the blend has been computing and blend-based alerts have been evaluating the entire time — the toggle controls display, not logic.

Likely cause 2. Every enabled slot is either warming up or has a weight of 0. The blend has nothing to blend.

Quick check. Check slot enable/weight/warm-up status.

Fix. Wait for warm-up, or restore a non-zero weight to at least one slot.

Blend moves for reasons the visible slots cannot explain

What you are seeing. The blend responds to something the slot lines you can see do not account for.

Likely cause. A hidden slot (Hide Stoch 0N Plot = true) is contributing to the blend.

Quick check. Open the input panel. Walk the three slots. Check Hide Stoch 0N Plot on each.

Fix. Unhide the slot to see its contribution, or disable it to remove it from every downstream behavior. Hide is display-only; it does not exclude a slot from the blend.

Alignment alert never fires on a clearly trending instrument

What you are seeing. The market is visibly trending in one direction, the slot colors seem to agree, and yet All Stoch Slots Bullish (or Bearish) does not fire.

Likely cause 1. Warm-up or missing data means there may not be a clean set of K/D states yet. The alignment tally only counts enabled slots with non-na K, and a slot whose D is not ready cannot honestly add a bullish or bearish vote.

Quick check. Look at all three slot lines. Is one missing or drawing a fallback?

Fix. Wait for warm-up, then re-check once all enabled slot lines are visibly reporting.

Likely cause 2. A slot's color disagrees with what you thought it was saying. A full-tone color means K above D, not K above midline.

Quick check. Confirm each slot's color state matches the alignment direction you expected.

Fix. Read colors, not inferences from values.

Misreads

Slot line at 55 is colored faded and it feels wrong

What you are seeing. The slot's K line is well above 50, but the line is faded instead of full tone.

Likely cause. This is the correct behavior. Slot color is driven by the K-vs-D relationship on the slot, not by K's position against the midline. A slot at 55 with K below its own D is exactly as described — value above midline, K-vs-D leaning down. The color is reporting the more informative of the two states.

Quick check. See Visuals and Logic. The section on the "slot at 55 in full tone is not the same event as a slot at 55 faded" is the full treatment.

Fix. Reset the habit: color before value.

Slot value looks different from a textbook stoch on a separate pane

What you are seeing. You compared one of the slot lines on Base to a standard single-timeframe stochastic indicator on the same timeframe, and the numbers do not match.

Likely cause. The Base slot K line is smoothed once — classical slow-%K, with your chosen length and MA family. A "standard" stochastic indicator showing raw %K and smoothed %K may be drawing one of them, not the same one you are comparing to. Also, the Source: input on Base can substitute hlc3 or ohlc4 for the close term, shifting the raw %K.

Quick check. Compare your slot's K Smoothing: length and family to the textbook stoch's settings. Check Source:.

Fix. For a strict apples-to-apples check, set Source: close, K Smoothing: 3 SMA, D Length: 3 SMA on the slot, and compare against a classical slow stoch indicator with matching K length, 3-bar slowing, and 3-bar D. The numbers should be close.

Per-slot alert fires on every confirmed bar for many bars in a row

What you are seeing. You wired Stoch 01 Is Bullish to fire on every match in TradingView, and it fires twenty or forty times in a stretch.

Likely cause. This is the intended behavior. Per-slot alerts are level conditions, not transition conditions. They fire on every confirmed chart bar the state holds.

Quick check. See Alerts. The state-not-transition framing is the core of the page.

Fix. If you want one alert per bar, use TradingView's once per bar close option. If you want a single alert at the moment state became true, you need external state tracking or a separate transition-focused script. The blended K and D are plotted, but Base does not plot per-slot D lines, so per-slot K-vs-D transition alerts cannot be built from the visible slot K line alone.

Blend near 50 but the slots look active

What you are seeing. The blend is sitting close to 50 and the blend bull/bear alerts are not firing, but the three slots appear to be doing something.

Likely cause. The slots are disagreeing with each other. A weighted average of slots pulling in opposite directions cancels near the middle of the range.

Quick check. Read the three slot colors. Are any two the same? Are any two opposite?

Fix. Read the slot spread, not just the blend. A quiet blend with tightly clustered slots is a different state from a quiet blend with slots spread widely across 0..100. See Visuals and Logic.

Slot pinned at 95 and the bull alert keeps firing

What you are seeing. A slot — commonly slot 03 on a trending session — is pinned near 95 for a long run, and the per-slot bull alert keeps firing.

Likely cause. Saturation. On a strong trend, the slot's K is sitting near the top of its K-length lookback window for an extended run. The per-slot bull alert is firing on every confirmed chart bar the state holds.

Quick check. See Visuals and Logic on saturation as context.

Fix. Do not fade saturation. Saturation is the state, not the setup. If you want an alert for saturation release — the moment the slot leaves the pinned region — you wire a crossunder alert externally on the slot line.

60m slot moves inside open bars

What you are seeing. Slot 03 (60m by default) drifts across chart bars within the same 60m bar.

Likely cause. On Bar Close? is off. With the switch off, the slot returns live higher-timeframe values that can revise until the 60m bar closes.

Quick check. Open the input panel. Check On Bar Close? in the PU Settings group.

Fix. Turn it on if you want stable, confirmed reads. Keep it off if you want live intra-bar reads and you accept the revision cost. See MTF and Repainting.

Master smoothing is on but the blend alerts fire later than expected

What you are seeing. You enabled master smoothing to calm the blend, and the blend alerts are now firing later than they used to.

Likely cause. Exactly as designed. Master smoothing adds a smoothing pass on top of the blend, which adds lag. Blend-based alerts evaluate on the smoothed blend and therefore fire later.

Quick check. See Settings on master smoothing.

Fix. Choose between the two: calm visuals with lagged alerts (master on), or responsive alerts with a choppier visual (master off). Both are honest. Neither is better in the abstract.

Genuine limits

You want per-slot On Bar Close?

The limit. Base exposes one global switch. All three slots move together.

Where to look. CTX exposes per-slot On Bar Close?. If you need slot 03 on ON for a stable HTF context display and slot 01 on OFF for a live intra-bar read simultaneously, that is a CTX configuration.

You want cross-asset or outside-symbol slots

The limit. Every slot on Base runs on the chart's symbol.

Where to look. CTX exposes outside-symbol slots.

You want more than three slots

The limit. Base is a three-slot workbench by design.

Where to look. CTX exposes ten slots.

You want inside-family MA tuning

The limit. Base imports the Lite MA library. In the current source, that means SMA, EMA, RMA, WMA, VWMA, and SWMA. HMA, ALMA, Jurik/JMA, KAMA, FRAMA, and inside-family controls are not exposed on Base.

Where to look. A wider trim or separate tool that actually imports the Extended MA surface. Do not go looking for those knobs in this Base input panel; they are not there.

You want a transition alert on K crossing D, or on alignment becoming true

The limit. Every alert on Base is state-based, not transition-based. The indicator does not expose transition alerts.

Where to look. TradingView's alert wiring layer for plotted-series crosses, and external flag tracking if you need "just became true" behavior from these state alerts. Base does not plot per-slot D lines and does not pretend to expose native transition alerts.

You want a divergence alert

The limit. Base does not implement divergence detection.

Where to look. A divergence indicator layered alongside the pane. Base does not pretend to do this work.

You want trade entry and exit signals

The limit. Base reports state. It does not produce trade instructions.

Where to look. Whatever trading methodology you use on top of context indicators. Base's job is the context, not the decision.

The larger pattern — what triage teaches you

Six months from now, most of the specific symptoms on this page will be easy. A new one will show up. The value of walking the three categories is that when a new symptom surfaces, you do not start from scratch — you ask the three triage questions:

1. Is the tool refusing to compute, drawing fallback values, or producing output that does not match its inputs? That is a setup error.

2. Is the tool behaving exactly as its inputs dictate, and your interpretation is reading something into it that is not there? That is a misread.

3. Is the tool behaving exactly as its inputs dictate, and what you need next is something Base does not do? That is a genuine limit.

A reader who can place a symptom inside one of those three categories quickly has already done the expensive part of the work. The specific fix is usually a sentence or two after that. The cost of not having the triage habit is re-learning each symptom from scratch, and that cost compounds across months of use.

Cross-references

• Symptom categories with more depth on misread-prevention: Limitations and Trust Boundaries.

• Inputs behind each setup error: Settings.

• Alert behavior in detail: Alerts.

• MTF rules and repaint mechanics in detail: MTF and Repainting.