Do trading strategies work on US stocks?
None passed the corrected test in these backtests. Of 233 strategies tested on US stocks, 36 finished ahead of the equal-weight benchmark on the point estimate, 1 of those with an unadjusted p-value under 0.05, and none met the corrected Benjamini-Hochberg bar (q below 0.10). The median strategy returned −10.9 pts/yr against the benchmark. These are historical diagnostics, not validation under Testing Standard v2: the backtest harness predates that standard and has not been re-run to meet it.
The universe is US stocks only, with data up to 2025-03-15. Most strategies are published or popular rules coded from a source description (a few are Tickfloor research-desk hypotheses), charged modelled trading costs and scored against an equal-weight benchmark of the same assets.
The numbers
- 233 strategies tested on US stocks.
- 36 finished ahead of the benchmark on the point estimate, 1 with an unadjusted p-value under 0.05.
- None met the corrected bar (q below 0.10).
- The median strategy returned −10.9 pts/yr against the benchmark.
The closest results
These five came out furthest ahead of the benchmark. Being furthest ahead among 233 tries is what luck alone produces, which is why the corrected q matters more than the headline figure.
- 12-1 cross-sectional momentum, top 4 of 40 large caps, monthly (wide US stocks) (Momentum): +18.8 pts/yr vs the benchmark, unadjusted p=0.047, corrected q=1.000.
- Crypto-style top-N cross-sectional momentum, long-only (wide US stocks) (Momentum): +9.7 pts/yr vs the benchmark, unadjusted p=0.231, corrected q=1.000.
- 12-1 cross-sectional momentum, top 4 of 40 large caps, monthly (Momentum): +8.8 pts/yr vs the benchmark, unadjusted p=0.137, corrected q=1.000.
- Aggregate-volatility risk: long the lowest VIX-beta stocks (Volatility): +6.2 pts/yr vs the benchmark, unadjusted p=0.187, corrected q=1.000.
- 13-week up/down volume ratio leaders (Momentum): +5.8 pts/yr vs the benchmark, unadjusted p=0.069, corrected q=1.000.
Tickfloor's research desk has backtested 517 strategies, most against an equal-weight benchmark of the same assets rebalanced daily that pays no costs (the strategies pay theirs). After correcting for the 598 tests run (Benjamini-Hochberg), 0 passed. That does not prove no strategy works. These are historical diagnostics, not validation under Testing Standard v2: the backtest harness predates that standard and has not been re-run to meet it.
General information only. It doesn't consider your objectives, finances or needs. Tickfloor holds no financial services licence and never places trades.
Strategy families tested on US stocks
- Momentum (42 on US stocks)
- Equity factors (31 on US stocks)
- Volatility (26 on US stocks)
- Trend following (24 on US stocks)
- Pairs and stat arb (20 on US stocks)
- Mean reversion (18 on US stocks)
- Social media chart patterns (13 on US stocks)
- Event driven (10 on US stocks)
Do trading strategies work on US stocks?
None passed the corrected test in these backtests. Of 233 strategies tested on US stocks, 36 finished ahead of the equal-weight benchmark on the point estimate, 1 of those with an unadjusted p-value under 0.05, and none met the corrected Benjamini-Hochberg bar (q below 0.10). The median strategy returned −10.9 pts/yr against the benchmark. These are historical diagnostics, not validation under Testing Standard v2: the backtest harness predates that standard and has not been re-run to meet it.
What is the benchmark?
An equal-weight long position in the same assets, rebalanced daily, before any benchmark costs. The strategies pay trading costs and the benchmark does not, which makes the test harder to pass, not easier.
Why correct the p-values?
When 598 tests are run, some will look significant by luck. The Benjamini-Hochberg correction accounts for how many were tried before calling any result real.
Is this advice on what to trade?
No. It reports what happened to published and popular rules in the past. General information only, not personal advice.