Mean Reversion After Three Down Days: Tested

After three consecutive declines the probability of a bounce is higher than usual. The simplest counter-trend formulation, with no indicators at all.

Mean reversion: the rule without indicators

Three consecutive lower closes go long, three consecutive higher closes go short. The position is held for one bar and closed. The rule stays out of the market more than three quarters of the time.

Three down days: what the run showed

Run with the portal's own engine over 518 daily bars for each of 12 pairs, costs of 1 pip round turn, canonical parameters. The method and its limitations are set out in how we run backtests.

• Median return — 0.27%, profitable on 6 of 12 pairs
• Sharpe — 0.07, drawdown — 3.7%
• Trades — 63, win rate — 35%, time at risk — 23%
• Extremes: AUDJPY +13.4%, GBPUSD -4.6%

A near-zero result with a very low drawdown

The median is essentially zero with six profitable pairs out of twelve — a coin flip. But note the drawdown: it is among the lowest of all the rules, and that follows from the construction rather than from signal quality. A one-bar holding period and rare presence in the market limit the loss regardless of whether the signal has an edge.

Why zero is more honest than a negative

A zero result here is more informative than a negative one: it says that after three declines the market behaves roughly as it usually does. Counter-trend logic in its pure form provides no edge, but it does not ruin anyone either, because it spends little time at risk. The practical lesson for anyone building such systems: establish that the signal has an edge first, and only then layer position management on top — the reverse order manufactures the appearance of a working system out of nothing. For comparison: simply holding the position on the same data returned 4.79%, and this rule did not beat it — why a benchmark is not optional.

A caveat about the holding horizon

Specific to this rule: it holds for exactly one bar, so it does not test the counter-trend idea in full — only its shortest-term version. A bounce may unfold over several days, in which case our holding horizon is simply the wrong one. The general limitations — a short sample, correlated pairs, a signal tested without risk management — are listed in the methodology. The data is open and the run reproduces from a script in the repository.

This material is educational and is not individual investment advice. Backtested results do not guarantee similar results in the future. Trading forex carries the risk of losing capital.