The Test Bench › gate

Does NYSE Advance/Decline Line actually work?

OVERSOLD

Works, but nothing like as well as it is sold.

The claim

NYSE Advance/Decline Line times market turns — Classic market-breadth canon

What we did

Tested against Buy & hold SPY, 2012-2026, thresholds swept, null-floor controlled.

The result

Result: 0.71 dSharpe (vs buy-hold), OOS

OOS edge +0.71 vs buy-hold; data-mining noise floor ~+0.40 (clears it)

Verdict

The single most-watched breadth gauge. It looks good in the recent test years (+0.71) -- but in the earlier years the very same rule HURT (its score was negative). A signal that helps in one stretch and hurts in another can't be trusted to help in the next; the recent win is most likely luck. Famous, not dependable.

What the test did and did not do

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

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.

← NFCI_risk<0 All findings Nasdaq Advance/Decline →

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.