How do you find the swing points that define a trend?

Markets Foundations defined a trend as a pattern of highs and lows. Fair enough — but a chart contains thousands of small highs and lows. Before you can talk about trend structure, you need a repeatable way to decide which ones count. Those are swing points.

A definition you can actually apply

A swing high is a candle whose high is higher than the highs of a set number of candles on both sides of it. A swing low is the mirror: a low lower than the lows of the candles either side.

The "set number" is your choice, and it is the whole game:

• 2 candles each side — sensitive. Marks lots of small pivots. Good for short-horizon reading; noisy on a long chart.
• 5 candles each side — a common middle setting. Filters most wiggles, keeps the meaningful turns.
• 10+ candles each side — only the major turns survive. Useful for describing a multi-year picture in five or six points.

There is no correct number. There is only a number you chose deliberately and applied consistently — which is the difference between analysis and drawing shapes you like.

A worked example

Illustrative daily prices over three weeks, in dollars — one figure per session, standing in for that session's high and its low so the counting rule is visible on a single row. (Real swing points read each session's high and low separately; see "In the data" below.)

100, 103, 106, 104, 101, 105, 110, 113, 111, 108, 112, 118, 121, 119, 116

Using 2 candles each side, the swing points are roughly: a high near 113, a low near 108, and a high near 121 — plus an earlier high near 106 and low near 101.

Now use 5 candles each side on the same fifteen bars and nothing survives — not one candle qualifies. Three of them never get the chance: 106 sits third from the start with only two candles to its left, 101 sits fifth from the start with only four, and 121 sits third from the end with only two to its right. The two interior candidates fail on their merits: 113 is beaten on the right by 118 and 121, and 108 is undercut on the left by 101 and 105.

That is not a failure of the data. It is the right-edge problem arriving early, and it is the honest result: a coarse rule needs a long window before it can say anything at all. Same prices, and the coarser setting leaves you with no skeleton whatsoever until more data arrives.

Neither setting is wrong. But if you casually switch settings mid-analysis — sensitive when you want to see more pivots, coarse when you want a cleaner story — you can produce whatever structure you were hoping for. That's not reading a chart; that's decorating one.

Why swing points earn their keep

Once you have them, several things become concrete rather than vague:

• Trend structure becomes a countable sequence of highs and lows (next lesson).
• Levels get objective locations: prior swing highs and lows are exactly where support and resistance are usually drawn (Unit 3).
• Trendlines get real anchor points instead of "wherever the line looks nice" (two lessons from now).

The right-edge problem

Here is the honest catch, and it never goes away. Under a 5-candle rule, a swing high cannot be confirmed until five candles have passed. The most recent potential pivot on any live chart is always provisional — it may be confirmed tomorrow, or price may exceed it and erase it. Every clean structure you see on a historical chart was fog when it was the right edge. Hold on to that discomfort; it is the correct feeling, and it is what separates chart reading from chart storytelling.

In the data

Swing points live in the highs and lows, not in the closes. Put a Level on the highest high of the month below and look at the close of that session: a session's extreme is often a price the close never touched.

Then switch the bars to Weekly. Each weekly bar's high is the highest of that week's daily highs, however many sessions the week held, so a swing point you can see on daily bars may not exist at all on the weekly ones.

Try it now

1. Check the five marked pivots on the schematic against the two-either-side rule yourself. Then apply a five-either-side rule to the same bars and find the high that stops qualifying. Same bars, coarser rule, smaller skeleton.
2. Now mark every swing high and low in the year of daily candles below with a consistent three-either-side rule, then redo it at eight and count how many survive.

3. Look at the last two weeks of that chart. Can you confirm the most recent swing yet — or is it still provisional? Say the answer out loud; that is the right-edge reality.