The Test Bench › gate

Does Higher Highs & Higher Lows (Dow Theory) actually work?

Real but small -- not worth what it costs to run.

The claim

A classic breakout/trend system that beats the market — Dow Theory / classic chart-reading

What we did

Tested against Buy & hold SPY, same window, SPY 1993-2026, prior-day, OOS 2014+.

The result

Result: -0.12 dSharpe (vs buy-hold); negative IS and OOS

Long/cash on SPY: +5-7%/yr vs buy-hold +10.8%; drawdown -55%->-30%; dSharpe -0.07 to -0.16, negative out-of-sample too

Verdict

The textbook definition of a trend — hold while each swing high and swing low tops the last, step aside when it flips to lower-high + lower-low. We coded it with causal pivots (a swing is only confirmed a few bars after the peak, no peeking) at three sensitivities (5/10/20-bar swings). It cuts the worst drawdown (-55% -> -30%) but trails buy-and-hold badly on return (+5-7%/yr vs +10.8%) and has no risk-adjusted edge anywhere — negative dSharpe in-sample AND out-of-sample, robust across every pivot setting. Waiting for structure to break costs you the rebounds; on a market that mostly goes up, that’s a losing trade.

What the test did and did not do

• tested out of sample
• point-in-time data
• checked for look-ahead
• net of real costs

Almost nobody publishing a backtest says which of these they did. We list the ones we failed as well as the ones we passed.

Tested 2026-06-23. The same treatment, applied to every other system, is on the test bench.

Written with AI assistance from our own test database. The numbers are computed, not composed.
General market commentary for education, not personalised investment advice. Backtested results are hypothetical. Trading carries a real risk of loss.