Market structure
How often markets trend
No trustworthy universal percentage exists for how often markets trend or range.
Repeated in many places and sourced in none.
The often-repeated 30% trending and 70% ranging split has no traceable origin. Time-series momentum at one-to-twelve-month horizons across 58 futures and forwards is documented; it is not the same as trends on a 15 minute chart.
What it is
Any figure depends on the definition used (a swing size, an efficiency ratio threshold, a timeframe). Measured properly, currency and futures markets show mild persistence over months and near-random behavior over short horizons.
What it suggests
Two books that give different percentages are not really contradicting each other. They are using unstated definitions.
How to check it
If a percentage is quoted, ask for the definition and the timeframe. The Exchange can measure a trader's own chart with a stated definition.
Common mistakes
- Planning around a trending share nobody can source.
Claims worth checking
- unsourced Markets trend about 30% of the time. Repeated on forums and prop-firm blogs with no primary source.
- unsourced Markets range more than 70% of the time. Same. The two claims are consistent with each other and with any definition that suits the author.
How we will test it
On public price history, never on anyone's account. Until it is done, the note above is all we can honestly say.
- Cases: windows of price history on public data
- Measure: the share of windows an efficiency ratio above a stated threshold would call trending, per market and timeframe
- Window: rolling
- Compared with: shuffled price changes
Sources
- Moskowitz, Ooi and Pedersen (2012), Time series momentum, Journal of Financial Economics 104(2) (citation and headline finding only)
- Liu and He (1991), A variance-ratio test of random walks in foreign exchange rates, Journal of Finance 46(2) (citation and headline finding only)
Related lessons
Written in our own words. Education only, not financial advice.