The Test Bench › scanner

Does Random-50 (null control) actually work?

NO VERDICT

Tested, but the result is not clean enough to call.

The claim

Nothing — 50 stocks picked at random. Proves the harness isn't lying. — Us (the control)

What we did

Tested against Equal-weight survivorship-free peer universe + SPY, net of cost, 2003–2026, OOS ≥ 2021.

The result

Result: -0.2 net-αSPY %/yr (beats index net of cost)

vPEERS −0.2 (lands at zero, as it must)

Verdict

Our honesty check. It SHOULD show no skill — and it does. When it briefly showed +25% we knew the engine had a bug, found it, and fixed it. The control caught the lie.

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.

← IBD Relative-Strength Leaders All findings Value (Earnings Yield E/P) →

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.