Why are levels zones rather than lines?

Beginners draw hairline-thin levels at exact prices and then spend their time arguing about whether $50.03 "counts" as a touch of the $50 line. Experienced chart readers draw bands. The reason isn't sloppiness — it's that a single-price level implies a precision the underlying data does not have.

Four reasons precision is fake

1. Which price is the level? A swing high has a wick high and a closing high. Take an illustrative turn: intraday high $51.40, highest close $49.80. Which is "the level"? Both are defensible. The honest answer is that the level is the region between them.

2. Multiple touches never land on the same number. Real turns cluster; they don't repeat. Highs of $118, $121 and $119 don't identify a line at $119.33 — they identify a zone roughly $117–122, and pretending otherwise adds decimals, not information.

3. The data itself shifts. Prices get adjusted for splits and dividends, and adjusted history is not the price people actually traded at. A level at "$50" on an adjusted chart may correspond to $63 on the unadjusted series. Both series are correct for their purpose, but a level is only meaningful on the series you drew it on.

4. Different participants see different prices. Different feeds, venues and session definitions produce slightly different highs and lows. Chasing the last cent is chasing an artefact of whose data you happen to be looking at.

A worked example

You've identified an area on an illustrative stock from three sources:

• a swing high at $51.40 (by wick),
• a cluster of closes near $49.60–50.10,
• the round number $50.00.

Rather than pick a favourite, mark the zone $49.50–51.50 — roughly 4% wide — and write "3 touches, weekly, most recent March" beside it.

Now every subsequent question gets easier. Price at $50.80 isn't "above resistance" or "below" it — it's inside the zone, which is a real and honest state to be in. Price at $54 is clearly outside it. The zone converts a false binary into a description that matches what you can actually know.

The schematic below is the same argument at a different price. Three turns, no two of them on the same number, and a band drawn from the highest and the lowest of them.

How wide?

There's no universal answer, and anyone offering one is bluffing. Sensible anchors:

• Scale with volatility. A zone should be wide enough to cover the normal noise of the instrument on your timeframe. A stock that moves 3% a day needs a wider band than one that moves 0.5%.
• Scale with timeframe. Weekly zones are wider than daily zones, because weekly swings are bigger.
• Let the touches define it. Often the cleanest method: the zone spans the actual reactions, from the lowest to the highest.

What you must not do is quietly widen a zone until it contains whatever price is doing now. Fix the width when you draw it, then leave it — otherwise the tool has stopped measuring anything.

In the data

Zone width has a measured yardstick: the average true range, how far the instrument typically travels in a session. Here is Apple's latest 14-day figure and its latest close.

The first number is in dollars, not percent, so on its own it says nothing about whether Apple is lively or quiet. Divide it by the price first; that is also the only way to compare it with another stock's.

Try it now

1. The three reactions on the schematic above are 118.00, 121.00 and 119.00. Draw the zone from the highest and the lowest of them and express its width as a percentage of price.
2. Now put a measured number beside your judgement. Take Apple's latest 14-day average true range and its latest close from the two tables in the section above, and divide the first by the second. Is your zone wider or narrower than one typical day, and should it be?
3. Redraw your strongest level from Unit 3 as a zone on the year below, then check whether price is currently above, below, or inside it — and notice how much more comfortable the honest third option feels.