The Test Bench › gate

Does CAPE / P-E10 (as a timing signal) actually work?

DEAD

No edge left once tested on a fair footing.

The claim

CAPE / P-E10 tells you when to be in or out of the market — Robert Shiller / valuation crowd

What we did

Tested against Buy & hold the S&P, modern 1997-2026 daily + Shiller 1920-2023 monthly, OOS, null-floor controlled.

The result

Result: +0.08 in/out edge (vs floor); noise over 100 years

+0.08 in/out edge (vs floor); noise over 100 years

Verdict

High CAPE genuinely predicts low long-run returns -- and genuinely cannot time entries and exits. Zero edge as a buy/sell gate over a century. The most expensive mistake in markets is confusing 'expensive' with 'about to fall.' They are not the same thing.

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.

← Advance/Decline Ratio All findings Cross-Sectional Dispersion →

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.