Do trading strategies work? 38 popular rules, tested
The rules did not pass the tests Tickfloor ran. Of 517 strategies backtested net of modelled costs, 0 passed the corrected test. That does not prove no strategy works, and it says nothing about versions that were not tested. 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.
Each page below takes one widely repeated rule, codes it from its source's description where one exists, backtests it with modelled trading costs against a benchmark (an equal-weight basket of the same assets for most pages, random-entry controls or other comparisons on the rest), and reports the result with the exact number of trades and the corrected p-value. Pick the rule you have been told about.
Indicator rules
- RSI Trading Strategy
- RSI(2) Mean Reversion
- MACD Crossover
- MACD + RSI Confluence
- StochRSI Cross
- Stochastic, CCI and Williams %R
- ADX and Parabolic SAR
- Bollinger Band Bounce
- Bollinger Band Squeeze
- VWAP Bounce
- Ichimoku Cloud
- Fibonacci Retracement
Trend, moving averages and breakouts
- Moving Average Crossover
- Golden Cross / Death Cross
- 9/21 EMA Cross
- The 200-day rule on SPY
- Donchian / Turtle Breakout
- Crypto Trend Following
- Dual Momentum (GEM)
- 12-1 Month Momentum
- 52-Week High
Chart patterns and price action
- Candlestick Patterns
- Support and Resistance
- The TJR / ICT sweep and fair value gap model
- Opening range breakout, 81% win rate
- Volume profile POC pullback
Calendar and seasonal effects
Sentiment, volatility and crypto positioning
- Fear & Greed Index
- Fear & Greed Timing
- VIX Term Structure Timing
- Funding Rate and Open Interest
- Buy the Dip
- Low-Beta Crypto
- Pairs Trading
By market
- crypto pairs: 146 strategies tested, none passed
- US stocks: 233 strategies tested, none passed
- ETFs: 138 strategies tested, none passed
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.
Do trading strategies work?
The rules did not pass the tests Tickfloor ran. Of 517 strategies backtested net of modelled costs, 0 passed the corrected test. That does not prove no strategy works, and it says nothing about versions that were not tested. 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.
How were these rules chosen?
They are the rules that turn up most in trading videos, forums and tutorials, plus the classic published ones. Most were coded from the source's description of the rule, as we interpreted it. The page for each rule says what was actually implemented, and where that differs from the popular teaching.
What does 'net of costs' mean?
Every trade in the backtest pays an estimated cost for each side of the trade. A rule that looks good before costs and bad after them did not survive the modelled costs; real fills, fees and slippage can differ.
Is this advice?
No. This reports what happened to a set of rules in the past. General information only, not personal advice.