Limitations and Trust Boundaries

This is the heaviest misread-prevention page in the pack. A reader who configures Base correctly can still misread it. The misreads live in what the instrument looks like from outside — a bounded oscillator with familiar numbers — versus what it actually reports under the hood. This page names the misreads in order of how much they cost if you carry them into a trading decision.

No disclaimers. No hedges. The traps are real, and naming them plainly is the point.

1. The alignment trap

This is the costliest misread on Base, and the one the pack has been pointing at since README.md. It deserves the first section.

Base exposes an All Stoch Slots Bullish alert and an All Stoch Slots Bearish alert. They sound like three timeframes independently converging on the same direction. If the slots have been deliberately differentiated across source, K length, and MA families, that read is close to correct — each slot is doing different work, and when all three land on the same K-vs-D state, something coherent is happening across three different measurements.

If the slots share configuration and differ only by timeframe, the alignment alert is not reporting consensus. It is reporting one measurement — the classical slow-stoch K-vs-D relationship with length 14, SMA 3, SMA 3, close source — sampled at three cadences. On trending instruments, the three cadences tend to agree with each other, because the same trend shows up in all three lookback windows. On choppy instruments, they tend to disagree for the same reason. The alert becomes a trend-persistence detector in disguise, and its confidence is wrong.

This misread is larger on Base than on CTX. CTX has ten slots. One misconfigured slot on CTX is a tenth of the alignment tally. Base has three slots. One misconfigured slot on Base is a third. If two of your three slots are near-duplicates of each other, the alignment alert is effectively a two-slot alert, and the two slots are not independent.

How to avoid it. Do not load Base with three slots that differ only by timeframe if you plan to rely on the alignment alert. If you only want "same stoch, three timeframes," the alignment alert is the wrong tool for you on this pane — the per-slot alerts on the three timeframes carry the same information without the consensus framing. If you do want alignment to mean something, differentiate the slots across source (some slots on hlc3), K length (short, medium, long), and MA families (SMA on one, EMA on another, RMA on a third) so the three observations are doing different work. The Workflows page describes the differentiated three-slot stack as its first named scenario.

How to verify. Turn on both All Stoch Slots Bullish and All Stoch Slots Bearish on three identical slots across three timeframes on a trending instrument. Watch the alerts fire in long runs — 20, 40, 60 bars at a stretch. Then differentiate the slots across source, K length, and MA family, and watch the same alerts fire more selectively. The two configurations are reporting different things.

What to do if you are already relying on the bad version. Ignore the alert on your current configuration and rebuild the slot differentiation before you trust it again. There is no software fix for a misconfigured three-slot alignment. The fix is in your slot choices.

One more thing. If the cleanest honest reading of your current three-slot configuration is that it does not earn the word "alignment" — three near-duplicate slots, or two duplicates and one outlier — the right move is not to wire the alignment alert at all. The per-slot bull/bear alerts carry the same K-vs-D information on each slot, and they do so without the consensus framing. Leaving alignment unwired on a configuration that cannot support it is a perfectly reasonable answer. The alert is not load-bearing on Base; it is a tool you use when you have built the differentiation to earn it.

2. The stoch-habit transfer trap

If you have been reading single-timeframe stoch for a while, you have a set of reflexes built up around what 80 means, what 20 means, what a K-D cross means, and what saturation means. Most of those reflexes transfer onto this pane correctly. The ones that do not are the expensive ones.

The classical single-timeframe stoch reflex is value-first: the line crosses 80, you notice; it crosses 20, you notice. That reflex reads the value and treats the color (or the K-D relationship, in a traditional pane) as secondary or decorative. On Base, the slot color is the primary decision on each slot. A slot at 55 in full tone and a slot at 55 faded are different events. A value-first read collapses them into the same event and throws away the part of the information that matters most.

The 80/20 reflex — "above 80 sell, below 20 buy" — is the other large transfer. It works on some instruments and some timeframes, sometimes. It does not work reliably on any instrument under trend, and it does not work on this pane as a rule: the 80 and 20 lines are reference zones with no internal logic attached, and the K-vs-D color discipline carries more information than the value-against-threshold read.

How to avoid it. Read color before value, every time, for the first month. After that you can keep reading in whatever order, but the habit will be there. The Visuals and Logic page walks the reading order with concrete examples; if the habit has not landed, re-read that page.

How to verify. Find a bar on your chart where a slot is at 55 in full tone. Find another bar where it is at 55 faded. Describe the difference in words — "K above D, leaning up" versus "K below D, leaning down despite value above midline." If you can do that out loud from the pane, the habit is installing.

How to tell if the trap has you. You are making decisions from slot values alone and checking color as an afterthought. Symptom: decisions feel unanchored from trend and cluster around false overextensions.

3. The saturation-as-signal reflex

Stochastic saturates at 0 or 100 under sustained trends. A 60m slot pinned at 95 for a multi-hour run is saying price has been trading near the top of its 60-bar, 60m-timeframe lookback window for that entire run. Under trend, that is the expected state — not a setup for mean reversion, not a sell, not evidence of exhaustion. It is the slot doing its job, accurately.

Readers trained on single-timeframe stoch fade setups will carry the "pinned at 95, fade it" reflex onto this pane and be wrong about it under trend. The per-slot bull alert will fire on every confirmed chart bar the state holds, for the entire duration of the saturation. That is not a malfunction. State alerts fire on every confirmed bar their state is true.

What is actually informative about saturation:

• Slot-spread saturation patterns. Slot 01 pinned at 95, slot 03 at 30 is saying the short-timeframe slot has been near the top of its short lookback window while the long-timeframe slot has been near the bottom of its long window. That is a texture read about where you are inside different lookback scales — useful.

• Saturation release. A slot leaving saturation is a genuine transition. It is not wired as an alert on Base; you notice it with your eyes or by wiring a crossover alert externally on the slot line.

• Saturation on one slot, not on others. A slot pinned at 95 while another slot at the same configuration on a longer timeframe is at 45 is saying the short-window trade is different from the long-window trade.

What is not informative about saturation:

• Saturation as a fade signal. The pane pinned at 95 does not mean sell. It means price has been near the top of a lookback window.

How to avoid it. When you see a slot pinned at a boundary, your first question should be what the other slots are doing, not whether to fade the pinned slot.

How to verify. On a strongly trending session, with defaults, watch slot 03 pin near 95 for an extended run. Confirm the per-slot bull alert on slot 03 fires on every confirmed chart bar during the saturation. Notice that the saturation persists through the trend and releases when the trend releases.

4. Color-vs-value neglect

Adjacent to the stoch-habit transfer trap but specific enough to warrant its own entry.

The color on every slot line encodes K-vs-D. Every slot on every bar. If you are reading a slot's value against the midline or against the 80/20 reference lines without looking at its color, you are discarding the K-vs-D information the pane generated for you. Roughly half of what each slot is telling you on any given bar is in the color; the other half is in the value. A reader who ignores either piece is reading the pane at half bandwidth.

This is not a decorative claim. A slot at 62 in full tone is a different event from a slot at 62 faded. A slot at 78 in full tone is different again from a slot at 78 faded; the second is particularly worth noticing, because it says K is already crossed below D while the value is still in saturation territory — the classical fade-at-80 reflex would treat those two states identically, and they are not.

How to avoid it. Force the color read first. Make it physical — train your eye to hit the color before the number, not the other way around. If your chart layout makes the colors hard to see, change the layout before you change anything else; per-slot Line Width: and your TradingView theme are both in your control. A reader who cannot see the color confidently cannot read the pane confidently, and no amount of knob tuning fixes that.

How to verify. For one session, every time you register a slot's value, pause and articulate its color out loud or in writing. If you hesitate or guess, that is a data point — the habit has not landed. After a few sessions, the check becomes unnecessary, because the color will have become the first thing you see.

How the trap compounds. Color-vs-value neglect compounds with the alignment trap. A reader who is not reading color accurately is also not checking whether three slots "aligned bullish" are genuinely aligned on their K-vs-D relationship or merely sitting above the midline. The two failures reinforce each other, and the cost of holding both at once is larger than the cost of either alone.

5. Weight-zero masking

You zero a slot's blend weight to try a two-slot blend. Later — maybe the next session, maybe the next day — you forget the weight is still at zero. The slot is still plotting. The slot's per-slot alerts are still firing. The slot is still counting toward alignment. But the blend is being driven by the other two slots only.

The symptom: the blend moves in a way the three visible slots cannot explain if you assume they are all contributing. They are not — two of them are. The misread is subtle because the pane looks normal.

How to avoid it. Decide whether a slot is contributing or not, and pick the right control for your intent. If you want the slot fully out of every downstream behavior — blend, per-slot alerts, alignment — disable it. If you want the slot plotted and alerting but excluded from the blend, zero its weight intentionally and remember you did it. The middle ground of "zero weight, forgot it" is the trap.

How to verify. Open your input panel and walk the weights on all three slots before a session. Takes ten seconds.

6. Hide-Plot masking

Related to weight-zero masking but different in cause. You hide a slot's plot for visual cleanliness. The slot is still computing. If its weight is non-zero, the slot is still contributing to the blend. Its per-slot alerts are still firing. It is still counting toward alignment.

The symptom: the blend moves in a way the two visible slots cannot explain. A third slot is contributing, but you do not see it.

How to avoid it. Same discipline as weight-zero masking. If you want a slot out, disable it. If you want it invisible but contributing, know that you chose that and remember you did.

How to verify. Every slot has a row on the input panel. If you are ever surprised by the blend, walk the three rows and check the Enable / Hide / Weight triad on each.

7. Master-smoothing lag

Enabling master smoothing adds a smoothing pass on top of the blended K and D. The blend visuals get calmer. The blend-based alerts fire later than they would on the unsmoothed blend.

The misread: treating the smoothed blend alert as early confirmation of what the unsmoothed blend would print. It is not. The smoothed alert fires on a lagged version of the blend. A reader who enables master smoothing because the blend looked choppy and then treats the smoother alerts as "cleaner signals" has inverted the actual relationship — the smoother alerts are later, not earlier.

How to avoid it. Treat master smoothing as a visual calm layer over the blend, not as a correction of it. If the blend looks choppy and that bothers you, the next move is often to revisit the slot configurations — because choppy blend behavior typically has its cause on one of the slots — not to hide it with another pass.

How to verify. Compare blend behavior side by side with master smoothing off and master smoothing on (EMA 3 or EMA 5) on the same chart. Notice that the flip timing shifts later on the smoothed version.

8. Warm-up misreads

On a fresh load, every slot needs enough bars for ta.stoch plus both smoothing passes to fill. During warm-up:

• The slot's K line draws using raw %K as a fallback for the first smoothing pass.

• The slot's D falls back to the K value while the second smoothing pass is still na.

• The slot's color falls back to a midline-based rule while D is na.

A reader who adopts early values before the slot-timeframe history has filled is reading fallback behavior. The fallback is a design choice to keep the pane visible during warm-up rather than leave it blank, but it is not yet the instrument running in its intended state.

How to avoid it. On a fresh load, let the chart breathe before drawing firm conclusions from the pane. With the default 60m slot on a 1m chart, think in loaded 60m history, not in the first few visible 1m bars.

How to verify. Reload the indicator and watch the early higher-timeframe slot behavior. The color can look a little strange until D fills, then snap into its normal K-vs-D behavior.

9. Chart-timeframe changes

You have been running the defaults on a 1m chart. You switch the chart to 1H to take a longer-timeframe look. The default 5m and 15m slots are now below the chart timeframe, so the script raises a runtime error instead of silently resolving them to the chart.

The cause is not mysterious. The code explicitly refuses any slot timeframe lower than the chart timeframe. It does this with a runtime error naming the slot, because a lower-timeframe slot request is not a valid configuration for this Base script.

The tool is correct. The configuration has lost its teaching purpose.

How to avoid it. Keep every slot timeframe equal to or higher than the chart's timeframe, and keep the slots meaningfully differentiated. When you change chart timeframes, revisit the slot configuration.

How to verify. Switch from a 1m chart to a 1H chart with the defaults still loaded. You should see the lower-timeframe slot error, not a quiet collapse. MTF and Repainting treats this case explicitly.

10. The clamp-as-reason misread

The defensive clamps at 0..100 on each smoothing pass output are a final guardrail. They are not the reason the pane is bounded. The stochastic transform is bounded 0..100 by construction, and the current Lite MA families are averaging methods that should not normally push a bounded input outside that range.

A reader who believes the clamps are load-bearing for the bound will tend to distrust values near the rails — treating a pinned slot at 95 as "the clamp holding it up" rather than as "the transform reporting where price sits inside its lookback." That misread distorts interpretation at exactly the points where the pane is most informative: extremes.

How to avoid it. Remember that stochastic's native range is 0..100 by definition. The clamp is a defensive bound in the code, not evidence that values near the rails are fake or being forced there.

How to verify. For the Geeks covers the clamp-as-safety-net distinction in its pipeline walk.

The honest limits of this pane

This is not a complete-market tool. A few things it does not do, named plainly.

• It does not produce trade instructions. The pane reports state. Entries and exits are your call. No number of slots agreeing is a trade signal.

• It does not account for spread, liquidity, session regime, or news. Anything outside the stochastic measurement itself is outside the pane.

• It does not carry symbol-specific tuning. The defaults are sensible for a broad range of instruments. They are not optimized for any specific one. Symbol-specific tuning is your work to do, using the per-slot inputs.

• It does not expose the wider Pro MA palette on Base. The current Lite library supports SMA, EMA, RMA, WMA, VWMA, and SWMA. HMA, ALMA, Jurik/JMA, KAMA, FRAMA, and their inside-family controls are not available in this Base source.

• It does not expose per-slot On Bar Close? on Base. One switch, all three slots. CTX exposes per-slot.

• It does not expose cross-asset ticker slots on Base. Every slot runs on the chart's symbol. CTX exposes outside-symbol slots.

• It does not expose more than three slots on Base. CTX has ten.

Each of those is a deliberate tier choice, not a missing feature. If the choice does not fit, CTX is the trim to look at. Base is not pretending otherwise.

Cross-references

• Reading-order discipline and saturation as context: Visuals and Logic.

• Scenario-based application of the differentiated-stack contrast: Workflows.

• The pipeline that produces clamp, smoothing, and blend behavior: For the Geeks.

• Symptom-first triage when something reads wrong: Troubleshooting.