What is a trend — really?

"The trend is your friend" might be the most quoted phrase in markets. Before the Technical Analysis domain examines what that's worth, let's define the thing itself — because "trend" has a precise, observable meaning that has nothing to do with prophecy.

Three weathers of price

Prices, observed honestly, do one of three things:

• Uptrend — successive peaks AND valleys both landing higher than the previous ones. The crowd's willingness to pay keeps ratcheting up; dips find buyers earlier each time.
• Downtrend — the mirror: lower peaks, lower valleys. Rallies exhaust sooner; sellers accept less at each stage.
• Range — peaks and valleys at roughly repeating levels; the market has (temporarily) agreed on a fair zone and oscillates inside it.

That's the entire definition: a pattern in the sequence of highs and lows. You can verify it with a ruler — no indicators, no faith required.

Why trends exist at all

Nothing forces them to — and yet they recur, because the world moves slower than the tape. Information spreads unevenly; institutions build million-share positions over weeks, not seconds (Course 2's big feet); narratives recruit followers gradually. Behavior stretches reactions over time, and stretched reactions ARE trends.

The one-line definition most desks actually use

Peaks and valleys need judgement about which wiggles count. So many desks reduce the question to a single overlay: the 200-day moving average, the mean closing price of the last 200 sessions, roughly ten months. Price above a rising average is called an uptrend, below a falling one a downtrend, and everything else a range. The cost of that tidiness is lag: an average of ten months turns several months after the price did, so it confirms a trend long after it began and announces the end long after it ended. It is a ruler with a delay built in, and knowing the delay is most of using it.

The honest asterisks

Trends are descriptions of the past: a textbook uptrend can end the minute after you label it, and every downtrend in history was an uptrend that did. Ranges break. Labels drawn on hindsight look magically obvious — drawing them at the RIGHT EDGE of a live chart, where tomorrow is blank, is the humbling exercise. That gap between hindsight clarity and right-edge fog is where the Technical Analysis domain begins its real work: rules, statistics, risk control — not vibes.

Try it now

One year of a real price, with tomorrow deliberately missing:

1. Mark three consecutive peaks and three consecutive valleys — a Level at each keeps them on screen — and read the six prices off them.
2. A trend is a claim about arithmetic, so check yours against it. Each high above the one before, each low above the one before: that is four comparisons, and count how many of them actually hold. Four out of four is an uptrend by definition. Two or three out of four is what a range looks like in the weeks before anybody admits it is a range, and it is the honest label far more often than the charts in tutorials suggest.
3. Label the current state from your count: up, down, or range. Then note how confident you honestly feel at the right edge, where the next point is blank. That gap between hindsight clarity and right-edge fog is the whole lesson.
4. Now stretch the same instrument to five years and label it again:

The one-year label often sits inside a longer trend pointing the other way. Which window you chose decided your answer, and nothing about the company changed. 5. One more ruler, one button away: switch either chart to Weekly bars and run the same four comparisons a third time. A trend that survives on daily bars and dies on weekly ones was a fortnight of weather rather than a season. Unit checkpoint next — then the dimension charts hide by default: volume, the market's conviction meter.