Updated 28 September 2026
We test popular trading strategies with costs, against holding the index, on years they were not tuned on and at realistic fills (how we test). So far none has cleanly beaten simply holding the index. Here is every verdict, newest first.
08 · Not a Winner. Maybe a Sidekick. Opening-range breakout, rescued. A gap filter brings it to the edge of the luck bar. Alone it trails QQQ; beside QQQ it helps. Works as an add-on, on borderline evidence.
07 · The 47-Second Tax A realistic fill costs a tenth to a sixth of profit, and at two minutes the breakout's edge stops looking real. Delay is a cost; test for it.
06 · We Broke Our Own Strategy Saty Phase Oscillator. Strong on the years it was tuned on; on the years it never saw, the evidence is too thin to call. Unproven. Verdict revised 27 September 2026; the note at the top of the post says why.
05 · The Winner That Doesn't Win Connors RSI(2). A real edge that wins about three trades in four, inside an account that trails buy-and-hold by a wide margin. Loses to buy-and-hold; works as an overlay.
04 · The Century That Ended in 1990 Industry trend following, 100 years. Zarattini and Antonacci's result replicates, and it was strong before 1980. Since then it loses to the market. Replicates; loses to the index today.
03 · The Missing Ten Weeks Weekly SPX put credit spread. The 95% win rate reproduces. Include the 2020 crash and drawdowns grow five to ten times. Replicates; the risk hides in the tail.
02 · The $25,000 Door Day trading after the PDT rule. FINRA replaced the $25,000 minimum; the day-trading strategies it unlocks collapse under realistic slippage. Fails.
01 · Five Strategies, One Survivor A batch of five popular strategies. The 5-minute ORB, UT Bot, TTM Squeeze and Nadaraya-Watson envelope fail after costs. Only Connors RSI(2) survives, per trade; Issue 05 tests it as an account. Four fail, one survives per trade.
Your turn. Send us the strategy you are about to risk money on: reply to any issue or write to [email protected].
Backtested results are hypothetical. This is research, not investment advice. Full disclaimer.