You can see the same shape many times and still read it differently. The result changes because context changes: where the chart sits, how the larger flow is behaving, and whether the pullback is holding structure.

This post does not present a new entry method. It focuses on what to review first and how to decide candidate priority from a workflow perspective.

1) Scan wide first: why dense mode should come first

1-1) What to do right after launch

After opening the app, check the loaded workspace first. A stable routine starts with this order:

• Ctrl/Cmd + 7: switch to dense grid and scan a wider candidate field.
• Ctrl/Cmd + 8: switch to expanded grid to narrow focus.
• Space: expand the current chart for deeper verification.

1-2) Column and row logic

A common confusion is to read rows and columns in the wrong orientation. In practice, the layout works as column = symbol, row = timeframe. So one column lets you read one symbol’s flow across multiple horizons from top to bottom.

When the same pattern is judged differently, it is often because the column-level comparison is inconsistent. Start by comparing candidates column by column.

2) For the same pattern, the useful lens is context, not a truth label

Use the same two references only as lenses:

• EMA/NRZ are not entry switches.
• EMA (persistence view): a first pass to estimate whether momentum can be sustained.
• NRZ: a context check on whether pullback behavior still keeps structure.

The practical point is this: they help you rank observation candidates, not replace your decision.

2-1) A context checklist when outcome differs

Even if a chart shape is similar, reading changes when one of these differs:

1. Does the candidate match the broader context in that same column?
2. Is structure still being held after pullback?
3. Is this candidate “watch now” or “defer” in your own sequence?

This prevents emotional selection and keeps the workflow objective.

3) The practical meaning of consensus cues (border color)

Many users treat consensus as a direct action trigger. In this context, it is better understood as an attention allocation cue.

• When several signals point in the same direction, the border tends to show a clear directional frame (long/short emphasis) and becomes a stronger focus candidate.
• If signals are mixed, neutral, or noisy, the cue is often low-emphasis, so you usually move it to “wait” or “hold-off.”

When consensus appears, the right action is still not immediate entry. It is candidate filtering plus next-step verification.

4) Routine for same-pattern cases: scan → focus → record

4-1) Scan phase (about 2 minutes)

1. Start with Ctrl/Cmd + 7 to widen the candidate view.
2. Confirm the same symbol is aligned by column.
3. Before deep zoom, compare timeframe continuity in that column first.

4-2) Focus phase (1~2 minutes)

1. Read top-to-bottom within the candidate column.
2. Keep only likely candidates.
3. Use Ctrl/Cmd + 8 to narrow the board and preserve attention.

4-3) Record phase (30 seconds)

Do not overfit a decision in one glance. Record it.

• V: toggle check-note on the current chart.
• N: open the CheckNote section and review recent candidate logic.

A short one-line note for why you skipped or kept a candidate is enough to preserve continuity into the next session.

5) Fix your workflow with templates

If you recreate the setup manually every time, workflow drift increases. Keep it stable through templates.

• Set Grid Size, Timeframes per row, and Exchange, then run Generate Template (by size)
• Save with F12
• Load later with Ctrl/Cmd + L

Having separate templates for scan mode and focus mode reduces startup noise before markets move.

Closing

The same pattern is not the same outcome. For stable execution, use the 3-step loop: scan wide, focus deep, and record decisions.

Remember: 1k_scanner is not a document scanner. It is a Rust+egui-based multi-market, multi-timeframe trading scanning app.