Trend analysis is the family of techniques used to answer the most basic question in technical analysis — which way is this market biased, and how durably? — and then to trade in alignment with that bias rather than against it. The folk distillation is "the trend is your friend": a market that has been rising tends, on average, to keep rising for a while before it reverses. The core tension is that this is simultaneously the best-evidenced idea in TA and one of the hardest to operationalize: the underlying tendency for price to persist is real and measurable, but the discretionary tools used to identify a trend (drawn lines, moving averages) lag price and whipsaw badly in sideways markets, and naming a trend in real time is far harder than spotting it in hindsight. This node defines the concept and the major lenses; the mechanics live in the children below.

The lenses (and where to route)

There are four recognized ways to identify and qualify a trend, each a child node:

• Market structure — higher highs / lower lows — the most primitive definition. An uptrend is a sequence of higher swing highs and higher swing lows; a downtrend, lower highs and lower lows. This is the language of Dow Theory: a trend is assumed to remain in force until structure breaks (a failed higher low, a lower high). Route here for swing-point logic and trend-change confirmation.

• Trendlines & channels — drawing the trend geometrically. A rising trendline connects successive higher lows; price holding above it confirms the uptrend, and a clean violation is often the earliest warning of a change. Route here for line construction, the "two points to draw, third to confirm" rule, and channel boundaries.

• Moving averages — a sub-section with its own children, because MAs are the largest and most-used trend toolkit. The canonical rule: trend is up when price is above a chosen MA (commonly the 50- and 200-day), down when below; crossovers and slope add signals. Route here for SMA vs EMA, period selection, crossovers (golden/death cross), and MA support/resistance.

• Trend strength & exhaustion — qualifying the trend rather than just naming it. Is the trend strong (worth following) or tired (prone to reversal)? Route here for ADX, momentum divergence, and exhaustion signals.

A practical workflow uses several together: structure defines direction, a trendline or MA gives an objective line in the sand, and a strength gauge filters out weak/choppy trends where the whole approach fails.

Standing & evidence — the honest meta-take

Trend has more empirical backing than most of technical analysis, but the distinction between the phenomenon and the discretionary tools matters enormously and is routinely blurred.

The phenomenon is documented. Moskowitz, Ooi & Pedersen (2012), "Time Series Momentum" (Journal of Financial Economics), found that an asset's own past ~12-month return positively predicts its next-month return across 58 liquid equity-index, currency, commodity and bond futures, with the effect persisting for roughly a year and then partially reversing — exactly the under-reaction-then-over-reaction shape a trend implies. This is a measured, cross-asset effect, not folklore, and it is the academic backbone of the managed-futures / CTA industry. Note the distinction from cross-sectional momentum (Jegadeesh–Titman): time-series momentum is about an asset versus its own past, which is what "trend following" actually trades.

But the effect is cyclical and the tools lag. Trend strategies are famous for "crisis alpha" — they tend to do well in sustained dislocations (e.g. 2008) because trends extend (Hurst, Ooi & Pedersen / AQR; Man Group). The flip side: in trendless, low-volatility, mean-reverting regimes they bleed. Trend-following materially underperformed equities through much of the 2010s, and at major turning points the signal stays on the wrong side for weeks because it is backward-looking. The discretionary tools are worse: a raw moving-average crossover is a lagging indicator that whipsaws in ranging markets, generating clusters of false signals and small losses, and in choppy regimes its profit factor hovers near break-even (sources below). So: the idea of trend persistence has real evidence behind it; any specific drawn-line or MA-crossover rule does not inherit that evidence and must prove itself on its own.

Strengths & limitations

• Strength: trend alignment improves the odds on directional trades and is grounded in a genuine, cross-asset empirical effect; it also imposes useful discipline ("don't fight the tape").
• Limitation: every trend tool is regime-dependent — designed for trending markets and actively harmful in ranges, which is where most markets spend much of their time. Tools lag, so they enter and exit late.
• The #1 misuse: treating trend tools as predictive timing signals on their own, especially forcing a trendline or MA onto a sideways chart and "seeing" a trend that isn't there. Trend analysis is a contextual filter, not a standalone entry system; pair it with a strength gauge and accept that it will be wrong at turning points.

System relevance

Within Delvantic, trend analysis is a context layer, not a trigger. The Augustus trade-setup agent consumes trend direction and strength to decide whether a setup runs with or against the prevailing bias and to size conviction accordingly; the regime engine independently classifies trending vs ranging conditions, which is precisely the filter that decides whether the lagging trend tools are usable at all. Hard caveat for any consumer of this branch: a trend reading is a probabilistic, lagging filter, not a forecast — its real-world value on a given trade is Augustus's call at decision time, informed by live data and Cairn's measured record, not something this doc asserts.

Sources

• Moskowitz, T. J., Ooi, Y. H. & Pedersen, L. H. (2012). "Time Series Momentum." Journal of Financial Economics 104(2), 228–250. — NYU Stern PDF / AQR.
• Hurst, Ooi & Pedersen — "Trend Following: Equity and Bond Crisis Alpha" (AQR / Man Group commentary).
• StockCharts ChartSchool — Dow Theory, Trend Lines, Moving Averages (trend-identification mechanics and the 50/200-day convention).
• Practitioner / educational sources on moving-average crossover whipsaw and ranging-market underperformance (TradingPedia, ChartsWatcher, FractalCycles).
• Background on the cyclical 2010s underperformance and crisis-alpha profile (AlphaArchitect, pfolio academy reviews of the TSMOM literature).