Quick Start

The goal of this page is not to get you configured. The goal is to get you reading the pane honestly before you start turning knobs. Almost every costly misread on STR comes from a reader who started changing inputs before they understood what the defaults were already telling them. Spend twenty minutes here before you touch anything. You will not get that time back if you skip it.

Before you change anything

Load the indicator onto a 1-minute chart of a liquid, actively traded symbol. Leave every input at the shipped defaults. Let the chart accumulate for a while — STR has several warm-up windows that have to fill before the pane reports completely, and a partially warm pane looks like a broken one if you do not know what to expect.

Out of the box:

• Slots 01, 02, 03 are enabled at timeframes 5, 15, 60, weight 33.3 each, with Hide Plot on. You will not see the three slot lines yet; only the blend and its structure features. This is deliberate — the default pane is tuned to let you read the blended conversation first, without five colors competing for your eye.

• Slots 04 and 05 are disabled at weight 0 and no timeframe selected. They ship off because STR is not trying to push you to five slots; it is letting you grow into them.

• The blend plot is on. Divergence is on. BBWP columns are on. Donchian is on. Keltner is off — one of the four features ships off so the pane does not look crowded at first contact.

• The reference lines you see are 0, 30, 50, 70, and 100. The 30 and 70 lines are the user-configurable bands; 0, 50, and 100 are fixed by the indicator. Fifty is the midline. Zero and one hundred are structural rails.

The shipped configuration is not the answer. It is a credible starting posture for a reader who has not yet decided what their five slots are for. Its value is that it lets you watch the blend move through its own range for long enough to notice what the pane is actually reporting, before you begin shaping it.

The four-stage reading order

Before you react to a color or a column, walk the pane in this order. The same order shows up on Visuals and Logic and Workflows. It is the single most useful habit this pack can install, and it is the single most ignored habit among readers who show up expecting an indicator to tell them what to do.

1. Slot color first. When you turn Hide Plot off for a slot, its line is either full-tone or faded. Full-tone means that slot's smoothed RSI is above its own signal. Faded means it is at or below. This is an RSI-vs-signal read for the slot itself — not a comparison with 50, and not a price-direction call. If a slot's color flips, something changed between the slot's own smoothed RSI and its own signal line; nothing more, nothing less.

2. Slot value second. Where on the 0..100 axis the slot is sitting. Fifty is the midline. Seventy and thirty are reference bands to orient you; they are not triggers, and no alert on this pane reads them as thresholds. If a slot is full-tone at 45, it is in an RSI-above-signal regime while sitting below the axis midline — both facts are true at the same time and the pane is not in disagreement with itself.

3. Blend color third. The blend's lime-or-red reports the RSI-vs-signal relationship at the blend level — not a vote count across slots. It is possible for several slot colors to look bullish while the blend line has just crossed under its own signal; the blend color goes red on that bar because the weighted-mean math says so, and the slot colors are answering a separate question at a separate scale.

4. Structure features fourth, in a deliberate sequence. Read the silver Donchian steplines first (is the blend pressing its own range edges?), then the orange Keltner envelope if it is on (is the blend stretched against its own smoothed basis?), then the BBWP columns (is the blend's own Bollinger width compressed or expanded by its own history?), then the divergence triangles (is there a confirmed pivot pair disagreeing between chart price and the blended RSI?).

The reason for the ordering is practical, not aesthetic. The slot and blend colors come from the faster, more direct reads. The structure features come from slower, derived reads that will often mislead you if you anchor on them before you have read the underlying line they are describing. If you start at stage four and work backward, you will regularly find yourself explaining a structure feature with a mental model you have not yet calibrated against the slot stack.

First-bar sanity check on the defaults

With the indicator fresh on the chart, confirm the shape. This is not theater; it is how you catch a broken load before you spend fifteen minutes reading a misconfigured pane.

• Every plotted line sits between 0 and 100. The rails hold. If any line excurses, the load is broken and you should reload rather than continue.

• The BBWP columns print from a baseline of zero upward. A tall column means the current Bollinger width is ranking high against its own history; a short column means the opposite. Nothing about price volatility is being asserted here — the whole read is the blend against itself, and the qualifier "from the blend, not from price" rides every mention of BBWP in this pack.

• Donchian steplines bracket the blend. When the blend prints a fresh recent extreme, the stepline moves to it. Until it does, the lines hold.

• Divergence triangles print only when a confirmed pivot pair has formed in the opposing-direction pattern. On a fresh chart you may see none for a long while. That is expected and is a feature of the strict-pivot construction; it is not the engine failing.

• The first several hundred bars of the chart can carry warm-up nulls: the RSI smoothing pass, the signal pass, the BBWP lookback window, and the divergence state all wait for enough bars. If the BBWP columns are absent near the left edge of your chart, or if a slot tone sits in an early faded/down equality state while its signal pass is warming up, those are not bugs. They are the timing cost of not pretending every derived value is ready before it is.

Two looks-broken-but-isn't traps specific to STR

Both are common on a first session. Name them and you stop losing time to them.

BBWP columns are missing for a long stretch near the start of the chart. The BBWP lookback defaults to 252 prior blend-derived widths. Until that full window is populated, the percentile-rank helper has no ranking to report, and BBWP returns nothing. On a symbol with sparse history, or when you scrub to the very start of a chart, this shows up as a silent BBWP strip. It is not silence about volatility; it is silence about not yet having enough blend history to rank against. If you drop the lookback to see columns appear earlier, you are trading the stability of the percentile rank for quicker visibility — a fine diagnostic move, not a production setting, and you should restore the lookback once the point is proven.

A divergence triangle appeared to move backward after it printed. With Plot On Pivot off (the default), the triangle prints at the right-shoulder confirmation bar. If you toggle Plot On Pivot on, the same triangle back-shifts by Pivot Len bars to the original pivot. This is a visual choice, not a timing claim. Either way, the alert fired on the confirmation bar, and you did not have the marker in real time at the back-shifted bar. Hold both truths next to each other from the very first time you see this setting: ON is honest about where the geometry formed, and ON is misleading-looking about when a reader could have acted on it. Pair the two any time you use the setting. More on this in MTF and Repainting.

One guided progression: add a fourth slot

Run this once, with intent, before you start composing your own configuration. The point is not to teach you to use slot 04; the point is to let you feel what a single configuration change does to the blend — once, under controlled conditions, while nothing is at stake.

1. In the indicator settings, enable slot 04. Leave its source, length, smoothing, and signal type at defaults. Enter 240 as its timeframe. Give it a weight of 10.0. Turn Hide Plot off on slot 04 so you can see the line. Leave slots 01, 02, 03 with Hide Plot still on.

2. Look at the blend. It has shifted slightly from where it was with only the first three slots, because the weighted mean now includes a slower, 240m voice at a 10.0 weight against three 33.3 contributors. The shift is not dramatic; it is proportional to the relative weight you gave it. That proportionality is the whole design of the blend.

3. Walk the four-stage reading order with slot 04 visible. You should see a slower orange line sitting somewhere in the bounded range, with full-tone color when its own 240m smoothed RSI is above its own 240m signal, faded otherwise. Watch it step at the top of each 240m candle close, not every minute.

4. Raise slot 04's weight from 10.0 to 30.0. Watch the blend's shape flatten — the slower 240m voice now pulls harder against the faster three-slot baseline, and the blend becomes less reactive to intraday movement. If a structure feature is active (BBWP, for instance), it reshapes with the blend. It did not recompute; its input changed.

5. Drop slot 04 back to 10.0. Turn on Keltner (Show Keltner). Watch how the envelope wraps the blend, not any of the slots. A touch of the envelope is a stretch of the blend against its own basis. Nothing is asserted about the 240m slot directly, and nothing about price is being claimed at all.

That is the posture this trim wants from you: each change has a cost in one direction and a cost in the other, the blend's behavior is not always what any single slot's behavior alone would predict, and the structure features are not voting — they are reporting the same line from different framings.

What a finished first session actually looks like

Not much. That is the point.

You leave the indicator at defaults or very close to them. You have watched the blend move through the four-stage reading order at least once across a stretch of chart that included both a quiet passage and a busier one. You understand that the reference lines at 30 and 70 are orienting bands, not triggers, and that the alert list does not carry threshold-crossing conditions. You have toggled Plot On Pivot on and off on a single confirmed divergence and seen where the marker lands in each mode. You have noticed that the per-slot Hide Plot toggles do not turn the slots off — they still compute, still contribute to the blend, and still fire their own alerts.

You have not built a five-slot configuration you will trade on. You have not decided which MA family you prefer. You have not wired any alerts. You have not concluded anything about what the tool is "best at." All of those come later, after you have read the defaults honestly enough to know what you are changing away from and why.

When you are ready to configure on purpose, the next stop is Settings. It walks every input with a stated cost in both directions and flags the tripwires that typically bite readers who configured in a hurry.