What are the best swing trading setups?
Out of 49 classic setups we walk-forward tested across 28 tickers, only 4 held an edge on 10 or more of them. The best lists you read elsewhere rank setups by fame or by one spectacular backtest. We rank by breadth: on how many of the 28 tickers did the edge actually hold, out of sample, against a 100-run random-entry gauntlet. By that measure the winners are unglamorous, and several of the most famous setups do not make the list at all.
Based on the same process behind every Library report. How we test.
How we rank: breadth beats a hot backtest
A setup that returned 800 percent on one volatile ticker is a story. A setup that held its edge on 12 different tickers is a pattern. Breadth, the count of tickers where a setup passed walk forward testing and beat random entries, is the hardest number in our data to fake, so it is the one we rank by. The two style families also read differently: mean-reversion setups win often and small, trend-following setups win rarely and big. Both appear below, ranked within their family.
The best mean-reversion setups
| Setup | Ticker | Return | Sharpe | Win rate | Edge held on |
|---|---|---|---|---|---|
| Williams %R bounce | SPY | +35.5% | 0.69 | 74.2% | 12 / 28 |
| RSI(2) below 5 dip buy | QQQ | +35.4% | 0.56 | 82.1% | 10 / 28 |
| RSI(2) dip, MA exit | SPY | +44.4% | 0.76 | 74.7% | 9 / 28 |
| Stochastic %K/%D cross | SPY | +46.2% | 0.66 | 95.7% | 7 / 28 |
| CCI oversold bounce | DIA | +63.6% | 0.86 | 93.8% | 7 / 28 |
Buy weakness inside an uptrend, sell the snap back. High win rates and modest per-trade returns are the profile. RSI(2) dip buy, the Bollinger dip, and the four-down-days bounce also held an edge on 7 of 28 each. Return shown is a representative index or large-cap.
The best trend-following setups
| Setup | Ticker | Return | Sharpe | Win rate | Edge held on |
|---|---|---|---|---|---|
| SuperTrend follower | QQQ | +199.8% | 1.11 | 52.5% | 11 / 28 |
| Parabolic SAR flip | SPY | +99.0% | 1.13 | 60.0% | 10 / 28 |
| SMA 20/50 cross | SPY | +103.4% | 0.83 | 61.9% | 9 / 28 |
| EMA 8/21 trend rider | SPY | +52.2% | 0.75 | 44.7% | 9 / 28 |
| EMA 20/50 trend rider | QQQ | +110.6% | 0.91 | 51.9% | 7 / 28 |
Buy strength, ride it, exit on the give-back. Lower win rates, bigger winners, and deeper drawdowns than the dip buys. The Awesome Oscillator zero cross also held an edge on 7 of 28.
The famous names missing from this list
Notice who is absent. The Golden Cross, the RSI(14) oversold signal, the 52-week-high breakout, bullish RSI divergence, the MFI bounce, and Wilder's DMI cross are six of the most preached setups in trading, and every one busted in our test. They share a single failure mode: they fire so rarely that a decade produces only a handful of trades, and a handful of trades cannot prove an edge no matter how good the score looks. Fame and breadth turn out to be almost unrelated.
See the full Golden cross report →
See the full RSI(14) oversold turn report →
See the full 52-week high breakout report →
See the full Bullish RSI divergence report →
The pattern: even the best setups fail on most tickers
The broadest edge we found, the Williams %R bounce, held on 12 of 28 tickers. That is the ceiling. Every setup on this page failed on more tickers than it worked on, which is the most useful thing our data says about setup lists in general: no setup is an everywhere edge, and anyone selling one as universal has not tested it honestly. The practical consequence is that pairing a setup with the right ticker matters as much as picking the setup, which is exactly what each setup's full report shows, ticker by ticker.
How to use this
- Rank candidate setups by breadth, not by the best single backtest you can find for them.
- Match the style to your temperament: mean reversion pays in frequent small wins, trend following in rare large ones. The equity curve you can actually sit through is the one that works.
- Check the per-ticker matrix in a setup's full report before running it. Even the broadest setup here failed on 16 of 28 tickers.
- Be suspicious of famous setups. The six best-known names in our test all busted on sample size.
- Run your own backtest on your own ticker before trusting any row in these tables.
Common questions
What is the best swing trading setup?
By tested breadth, the Williams %R bounce: it held a walk-forward edge on 12 of 28 tickers, the most of the 49 classic setups we tested. Among trend-following setups, the SuperTrend follower led with 11 of 28.
Do most swing trading setups actually work?
Most held an edge somewhere: 43 of the 49 we tested passed on at least one ticker. But only 4 held an edge on 10 or more of the 28 tickers, so a typical setup works in far fewer places than its reputation suggests.
Is mean reversion or trend following better for swing trading?
Neither dominated our test. Mean-reversion setups showed higher win rates with smaller wins, trend-following setups lower win rates with bigger winners. The broadest edges came from both families.
Which swing setup has the highest win rate?
The Stochastic %K/%D cross posted a 95.7 percent win rate on SPY in our test. A high win rate alone is not an edge, though: several busted setups also had strong win rates on tiny trade counts.
Why do famous setups like the Golden Cross rank so poorly?
Sample size. The six famous setups that busted all fire rarely, a handful of trades in a decade, so their good-looking scores rest on too few signals to separate edge from luck.
We take each setup's rules as commonly stated. Your parameters may differ, so run your own on your own account and watch the walk-forward result before you trust it.