Skip to content

You Can't Always Trend When You Want: summary and key takeaways

The paper splits trend-following performance into three drivers: the size of market moves, how well the strategy turns moves into profit, and diversification. It finds the weak 2010 to 2018 decade came from small market moves, not from a weaker strategy or less diversification.

Babu, A.; Hoffman, B.; Levine, A. et al. · 2019 · SSRN Electronic JournalSummary updated 20 November 2026
Published2019Working paper
Asset focusMulti-assetTrend following
TypeStrategyTrend & Momentum
Reading time4 minthis summary

You Can't Always Trend When You Want at a glance

The question
Why did trend following earn lower returns in the current decade than in its long history, and did the strategy itself change?
Dataset & period
Monthly futures-based returns for 67 markets: 29 commodities, 11 equity indices, 15 bond markets and 12 currency pairs. Daily futures prices go back as early as 1877, with analysis from 1880 to 2018; the 'current decade' is 2010 to 2018.
Method
The authors regress each market's yearly trend-following Sharpe ratio on the absolute Sharpe ratio of the underlying market move. The slope and intercept measure trend efficacy. A diversification multiplier links average per-market performance to the portfolio Sharpe ratio. They then swap each factor, one at a time, for its full-sample average to see how much each decade's gap versus the long-run average is due to market moves, efficacy or diversification.

Key findings of You Can't Always Trend When You Want

  • Across the history, bigger absolute market moves in a market and year go with higher risk-adjusted trend-following returns.
  • The fitted line has a negative intercept and a slope below 1: trend following loses in directionless markets and keeps only part of a directional move.
  • In 2010 to 2018, market moves were small compared with the rest of the sample.
  • The line fitted to 2010 to 2018 closely resembles the full-sample line, so trend efficacy was not impaired.
  • Diversification, measured by the multiplier, looked much like its long-run level in the current decade, so rising correlation among trends is not the cause of the weak returns.
  • The 1970s were the strongest decade for trend following, with all three factors favorable.

Robustness, as reported in the paper

  • The model is fitted separately for each decade from 1880 through 2018, and the impact of each factor is compared decade by decade.
  • No clear or persistent pattern appears in any one factor's impact across decades.
  • In the current decade, big moves such as 2014 still paid off for trend following, as history would predict.

Caveats and limits

  • Transaction costs are estimated with large uncertainty, and costs of rolling futures contracts are not included.
  • Early data rely on manually transcribed prices and, for some periods, the average of high and low prices instead of closes.
  • The decomposition is exact only under assumptions listed in an appendix (inferred from the text).
  • The paper does not forecast future market moves; the outlook depends on move sizes reverting to long-term levels (inferred).

Items marked (inferred) are our reading of the design, not statements by the authors.

Why it matters

The paper offers a way to separate a bad stretch caused by quiet markets from one caused by a broken strategy. A practitioner reviewing a trend allocation can check move size, efficacy and diversification separately. The authors say muted moves, not a loss of efficacy, explain the recent gap.

Questions about this paper

Why was trend following weak in the 2010s according to the paper?

The paper attributes it almost fully to few big risk-adjusted moves in markets, up or down, during 2010 to 2018.

Did trend following stop working?

The paper reports no meaningful shift in efficacy and no meaningful drop in diversification.

What are the three factors in the framework?

Average magnitude of market moves, trend efficacy (intercept and slope of the move-to-return relationship), and a diversification multiplier.

Where can I read the original paper?

The original is at its publisher: https://doi.org/10.2139/ssrn.3487134. We link to the paper and do not host the PDF.

Follow new quant research

Download MacroFinance market data as CSV or Excel, or pull it by API, and test the ideas in these papers yourself. New summaries are added as we review them.

Create a free account
Summaries are written by MacroFinance from the published papers, in our own words, for information only. They are not investment advice and may omit detail. Papers are not hosted here: each page links to the original at its publisher.