Measuring Market Trendiness Against a Random Walk Baseline

Summary

This article seeks a reproducible way to distinguish trend-like behavior from flat or random movement. It first defines trading success through positive expected payoff, which depends on win probability and the sizes of average gains and losses. It cautions that changing the win rate or reward-to-loss ratio alone does not create an edge in a random process, and that costs can make an otherwise zero-expectation approach lose money. Martingale sizing, the author argues, does not change that underlying expectation.

For its market-structure analysis, the article represents price changes as fixed-size point steps and blocks, then uses a random walk as a reference. Combinatorial calculations estimate the distribution of vertical movement over a chosen number of steps; observed price-series distributions can then be compared with this baseline at different scales. The text reports qualitative examples across stocks, oil, and EURUSD, including instruments whose apparent trendiness changes with scale. This is a descriptive statistical framework, not proof of a profitable strategy; the author leaves the causes and trading use for further work.

Key ideas

• Positive expected payoff requires a favorable combination of win probability and average gain and loss sizes.
• Changing trade outcomes’ size ratios alone does not establish an edge in a random process.
• Fixed-size price blocks make point-by-point movement easier to compare with a random-walk model.
• A reference distribution of random-walk movement can be used to assess trendiness at different scales.
• The reported instrument comparisons are descriptive and do not demonstrate profitable trading rules.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.