Drawing Tools and How to Use Them
Drawings are a record of an analysis, and they are also the easiest way to convince yourself of something that is not there.
MadStockAlerts Research · Updated August 28, 2026
What to take away
- Horizontal levels are the most objective drawing and the most useful.
- Every drawing involves choices that a later reader cannot see.
- Fewer drawings usually read better than more.
- Dating a drawing is what makes it reviewable later.
- Given enough tools, any chart can be made to support any conclusion.
MAD Academy Training Video · 0:45
Fewer Lines, Drawn Earlier
Drawing tools are for recording a decision before the move, and a chart covered in lines is a chart nobody can read.
This lesson is part of a Stock Alerts + Tools plan.
The tools, ranked by objectivity
| Tool | How subjective | Best used for |
|---|---|---|
| Horizontal line | Low | A price where something clearly happened |
| Rectangle / zone | Low | A region rather than a single price, which is more honest |
| Trendline | High | A slope, when the touches are genuinely unambiguous |
| Channel | High | Framing a trend's boundaries |
| Fibonacci retracement | Very high | Reference fractions of one obvious swing |
| Measured move projection | Very high | A rough scale, never a target |
The ranking is a rough guide to how much weight a drawing should carry as evidence. A horizontal line at an obvious prior high is close to a fact; a Fibonacci level derived from a chosen swing is an opinion with a decimal point.
The discipline
- 1Draw before, not afterA level drawn after the move is a description of the past that will look prophetic and is not.
- 2Use zones for levelsA rectangle admits the imprecision that actually exists; a line pretends it does not.
- 3Keep one conventionWicks or bodies, log or linear. Consistency is what makes your own drawings comparable over time.
- 4Delete what stopped matteringA chart carrying every line ever drawn always has one near the current price, and that is not information.
The self-deception problem
Given enough tools and enough freedom, any chart can be made to support any conclusion. This is not a hypothetical; it is the ordinary result of drawing after forming a view. The only reliable countermeasure is drawing first and recording the date.
It is worth noticing how the failure feels from the inside, because it does not feel like fitting. It feels like finding, and the sense of discovery is exactly as strong whether the level was drawn before the move or after it.
Scroll the chart sideways to see all of it.
Saving and reviewing
Persistence is what turns a drawing from a doodle into a record. A level drawn weeks ago and still present when price returns to it is testable; one drawn today about the past is not.
Drawings persist per symbol, so a level drawn weeks ago is still there when the price returns to it, which is what makes reviewing your own calls possible.
Drawings and saved layouts on Pro Charting — for membersWhat a drawing is anchored to
A drawing has to be attached to something, and the choice determines what happens when the chart changes. A horizontal level is anchored to a price, so it survives every timeframe change, every zoom and every scale change untouched. That stability is most of the argument for preferring horizontal levels over everything else.
A trendline is anchored to two points, each of which is a price at a time. Change the timeframe and the underlying bars change, so the anchors land somewhere slightly different; change from arithmetic to logarithmic and a straight line in one is a curve in the other. A trendline drawn on a daily arithmetic chart and viewed on a weekly logarithmic one is a different line, and it will break at a different place.
This is the mechanism behind a common and entirely innocent error: a trendline that held perfectly is opened later on a different scale and appears to have broken weeks ago, or the reverse. Nothing moved. The drawing was never a property of the security, only of one rendering of it.
Some platforms let a drawing be anchored to a bar index rather than to a date, which makes it stable under timeframe changes and unstable under any change in the data, such as switching extended hours on. There is no anchoring scheme that is stable under everything, which is the underlying reason a fixed convention matters more than the choice within it.
Deleting is part of the discipline
Drawings accumulate. A chart worked on for a year carries levels from conditions that no longer exist, trendlines from a trend that ended, and channels drawn during a range that has since resolved. The accumulation is not neutral: a chart with thirty lines on it will have one near almost any price, so almost any price looks significant.
The convention that helps is treating a drawing as having a lifespan tied to the reason it was drawn. A level marked because it was the high of a base is meaningful until price is far enough past it that the base is no longer the relevant structure. Past that point it is a historical note, and leaving it on the chart gives it a weight it no longer earns.
- A level that has been broken decisively and left behind has done its work, whatever it once meant.
- A trendline from a trend that has ended is describing a condition that is over.
- A drawing nobody can remember the reason for is not evidence of anything, because the reason was the evidence.
- Anything drawn to justify a position already held was never a reading of the chart.
The test that separates the useful drawings from the rest is whether the drawing was made before the move it describes. That is also the only test that cannot be applied retrospectively, which is why the date a drawing was made is worth more than the drawing.