Trading concepts

5 Swing Trading Strategies for Beginners: Rules and Tested Results

Trend following, breakouts, pullbacks, gap and go and Fibonacci retracements, each with an exact rule; the first three are run on five years of S&P 500 ETF bars and compared with random entries, every trade counted.

· Founder of PatternsWizard
Checked 10 min read
SMA(20) and SMA(50) crossover on S&P 500 ETF, dividend-adjusted prices, daily, August 5, 2026 to October 28, 2026. Illustration of how the indicator reads, not a backtest.
On this page
  1. How we tested the strategies
  2. 1. Trend following: the 20/50 moving average cross
  3. 2. Breakout trading: the 20-day high
  4. 3. Pullback trading: buying a dip in an uptrend
  5. 4. Gap and go
  6. 5. Fibonacci retracement
  7. Which strategy should a beginner start with?
  8. Frequently asked questions

Five swing trading strategies suit a beginner because each can be stated as a simple rule: trend following, breakouts, pullbacks, gap and go, and Fibonacci retracements. We wrote an exact rule for the first three and ran it on five years of daily S&P 500 ETF bars, from October 18, 2021 to October 28, 2026 (1,263 sessions). None of the three beat random entries held for the same number of days by a margin larger than chance, and none came close to the +84.65% of simply holding the ETF over the window. The other two need data or judgment that a fixed rule on daily bars cannot capture, so they are described without results.

20-day and 50-day moving averages on the S&P 500 ETF

SMA(20) and SMA(50) crossover on S&P 500 ETF, dividend-adjusted prices, daily, October 30, 2025 to October 28, 2026. Illustration of how the indicator reads, not a backtest.
  • Price, daily candles (hollow up, filled down)
  • SMA(20) of the close (solid line)
  • SMA(50) of the close (solid line, second color)
  • Fast average crosses above the slow one
  • Fast average crosses below the slow one
SMA(20) and SMA(50) crossover on S&P 500 ETF, dividend-adjusted prices, daily, October 30, 2025 to October 28, 2026. Illustration of how the indicator reads, not a backtest.

How we tested the strategies

Each rule was run on the dividend-adjusted daily bars of an S&P 500 ETF. The window starts on October 18, 2021, the first session with a 200-day average in the five-year series we used, and ends on October 28, 2026. Trades are long only, one at a time, entered and exited at the close of the signal bar, without commissions, spread, slippage or taxes, so real results would be lower. To judge each strategy, we compared its mean trade with 10,000 sets of random entries in the same window. A set has one random entry for each closed trade of the strategy, drawn uniformly among the sessions that leave room for that trade's length, and held for the same number of sessions; the percentile is the share of sets whose mean trade is below the strategy's, 50 being the middle. The draws come from the mulberry32 generator, restarted at seed 1 for each rule, set by set and trade by trade in the order of the trades; an entry is the first session of the window plus the integer part of the draw times the number of eligible sessions. The median is the 5,001st of the 10,000 set means sorted in ascending order, and the 90% range runs from the 501st to the 9,500th. Other seeds move the percentiles by about one point. That is the same idea as the random baseline of our pattern bench, and the method is explained in our backtesting guide.

StrategyClosed trades (winners)Mean trade vs random entries
1. Trend following, 20/50 cross11 (5)+1.29% vs median +3.80% (percentile 14)
2. Breakout, 20-bar high24 (12)+1.47% vs median +1.51% (percentile 49)
3. Pullback above the 200-day average49 (38)+0.40% vs median +0.18% (percentile 81)

Mean trade of each rule against the median of random entries

Mean trade of each rule against the median of random entries. Trend following, 20/50 cross (11 trades): Rule +1.29%, Random entries, median +3.80%; Breakout, 20-bar high (24 trades): Rule +1.47%, Random entries, median +1.51%; Pullback above the 200-day (49 trades): Rule +0.40%, Random entries, median +0.18%. Our computation; the inputs and the method are in the text.
  • Rule
  • Random entries, median
Our computation; the inputs and the method are in the text.S&P 500 ETF, daily, October 18, 2021 to October 28, 2026, before costs. Random entries: 10,000 sets with the same number of trades and the same holding lengths. Percentiles 14, 49 and 81: none outside the middle 90% of random sets.

The details of each test, losers included:

StrategyMedian, best and worst tradeClosed trades compounded, time in the market
Trend followingMedian −3.44%; best +17.83%; worst −8.49%+11.22%; 70% of sessions, including the open trade
BreakoutMedian +0.18%; best +10.02%; worst −4.35%+39.67%; 55% of sessions
PullbackMedian +0.57%; best +3.02%; worst −6.13%+21.05%; 13% of sessions

1. Trend following: the 20/50 moving average cross

Trend following buys after the market has started rising and stays in as long as the trend lasts. The rule we tested: buy at the close when the 20-day simple moving average crosses above the 50-day, and sell at the close when it crosses back below. The chart at the top of the page shows both averages on the ETF. More on the signal in moving average crossover and on the style in trend trading.

Every trade of the rule, entries and exits at the close:

Entry to exitResultSessions
Oct 28, 2021 to Jan 25, 2022−4.87%60
Apr 1 to May 2, 2022−8.49%20
Aug 1 to Sep 16, 2022−5.75%33
Nov 11, 2022 to Jan 3, 2023−4.00%34
Jan 27 to Mar 14, 2023−3.44%31
Apr 13 to Aug 28, 2023+7.48%94
Nov 20, 2023 to Apr 26, 2024+12.69%108
May 20 to Aug 9, 2024+0.88%56
Sep 3, 2024 to Jan 14, 2025+6.13%91
Feb 4 to Mar 7, 2025−4.30%22
May 14, 2025 to Mar 2, 2026+17.83%199
Apr 23, 2026, open on Oct 28+9.57%130

The 11 closed trades of the 20/50 cross, S&P 500 ETF

The 11 closed trades of the 20/50 cross, S&P 500 ETF. Oct 2021 to Jan 2022: −4.87%; Apr to May 2022: −8.49%; Aug to Sep 2022: −5.75%; Nov 2022 to Jan 2023: −4.00%; Jan to Mar 2023: −3.44%; Apr to Aug 2023: +7.48%; Nov 2023 to Apr 2024: +12.69%; May to Aug 2024: +0.88%; Sep 2024 to Jan 2025: +6.13%; Feb to Mar 2025: −4.30%; May 2025 to Mar 2026: +17.83%. Our computation; the inputs and the method are in the text.
Our computation; the inputs and the method are in the text.Entries and exits at the close, dates in the table above, before costs. The trade opened on April 23, 2026 is still open and not shown.

Six of the 11 closed trades lost money, all of them between 3.4% and 8.5%, and the results came from a few long trades: the one from May 2025 to March 2026 alone made 17.83%. On this ETF, the rule made many small losses and a few large gains, which is why the median trade is negative while the mean is positive. Here the rule ranked below most random sets held as long (percentile 14), though still inside the range that 90% of them covered (+0.06% to +7.10%): the 20/50 cross reacts late by construction, since both averages lag the price.

2. Breakout trading: the 20-day high

A breakout trader buys when price closes above the top of its recent range. The rule we tested: buy at the close when it is above the highest high of the 20 previous sessions, which is the upper band of a Donchian channel; sell at the close when it falls below the lowest low of the 10 previous sessions.

20-day high and low (Donchian channel) on the S&P 500 ETF

Donchian Channel(20) on S&P 500 ETF, dividend-adjusted prices, daily, October 30, 2025 to October 28, 2026. Illustration of how the indicator reads, not a backtest.
  • Price, daily candles (hollow up, filled down)
  • Upper, highest high of the previous 20 bars (solid line, second color)
  • Lower, lowest low of the previous 20 bars (solid line)
  • Middle, (upper + lower) / 2 (dashed line)
  • Close above the upper line (new 20-bar closing breakout, first day)
  • Close below the lower line (first day)
  • Shaded area between the bands
Donchian Channel(20) on S&P 500 ETF, dividend-adjusted prices, daily, October 30, 2025 to October 28, 2026. Illustration of how the indicator reads, not a backtest.

Half of the 24 trades won, the median trade was +0.18% and the worst lost 4.35%, with losers closed at the first close below the 10-day low. The mean trade, +1.47%, was almost exactly the median of random entries (percentile 49): on this ETF, buying the breakout did no better than buying any day and holding as long. Patterns such as flags, rectangles and ascending triangles are breakouts too, with a shape added; their results on many markets are on our bench below, and the general idea is on breakout.

3. Pullback trading: buying a dip in an uptrend

A pullback trader waits for a rising market to dip, then buys before the rise resumes. The rule we tested: when the close is above the 200-day simple moving average and lower than the previous close for the third day in a row, buy at the close; sell at the first close above the previous session's high, or after 10 sessions.

The pullback rule: three lower closes above the long average

The pullback rule: three lower closes above the long average: candles above a rising long moving average. Three closes in a row, numbered 1, 2 and 3, are each lower than the one before; a green dot marks the buy at the third lower close. Two bars later, a close rises above the prior bar's high, marked by a dashed line: that close is the exit.
Schematic, not market data. The rule tested on this page, drawn by hand; not a real trade.The pullback rule as tested on this page, drawn by hand: buy at the third lower close in a row above the long average, sell at the first close above the prior session's high.

This rule won most often, 38 of 49 trades, but each gain was small: the best trade made 3.02% and the worst lost 6.13%, twice that. With a median hold of 3 sessions, the rule was in the market only 13% of the time. Its mean trade ranked at percentile 81 against random entries: better than most random sets, but inside the range that 90% of them covered (−0.26% to +0.60%), so not clearly better than chance. On this ETF, the rule won often and lost rarely but larger.

4. Gap and go

Gap and go trades a stock that opens well above the previous close, for example by 2% or more, and buys if it keeps rising after the open. Descriptions vary on the size of the gap and on the length of the opening range, so this version is our synthesis of the common one: buy above the high of the first 15 minutes of trading, with a stop below the low of those 15 minutes and a target set from the stock's usual range. The kinds of price gaps are explained in our gap guide.

Gap and go: buy above the opening range

Gap and go: buy above the opening range: yesterday's candle and a dashed line at its close. Today the price opens well above that close, a gap up, and swings inside a shaded opening range. A green line marks the range high, where the buy is placed; a red line marks the range low, where the stop goes. The price then breaks above the range high and keeps rising.
Schematic, not market data. The rule as it is commonly described, drawn by hand; not tested on this site.Gap and go as commonly described, not tested here: the opening range is the first minutes of trading; buy above its high, stop below its low.

5. Fibonacci retracement

The Fibonacci strategy draws levels between a recent swing low and swing high, at 38.2%, 50% and 61.8% of the move, and buys when price pulls back to one of them and turns up, with a stop below the next level. We did not test it either: the result depends on which swing high and low you pick, a choice a fixed rule only approximates. The levels and how to draw them are on Fibonacci retracement.

How a Fibonacci retracement is drawn

How a Fibonacci retracement is drawn: an up move from a swing low to a swing high, with horizontal levels at 23.6%, 38.2%, 50%, 61.8% and 78.6% of the move measured down from the high, and a pullback that stops near one of them.
Schematic, not market data.The levels are measured on the move from the chosen swing low to the swing high; another swing gives other levels.

Which strategy should a beginner start with?

Start with the one whose rule you can write in two lines and test on past data before risking money. In the three tests above, the trend rule lost more often than it won (6 of 11) but won big, the breakout rule won half the time, and the pullback rule won often and small; on another market or period, the profile can differ. Whichever you choose, set the stop before the entry and size the position from it: our guides on where to place your stop and position sizing show how. Whether swing trading suits your schedule better than day trading is discussed in day trader vs swing trader.

For further reading, see our list of the best swing trading books and more styles in trading concepts. If you prefer structured lessons, our comparison of swing trading courses gives the rating, length and focus of each.

Every pattern's results on one sheet

The only pattern cheat sheet with tested stats: 152 patterns, what the textbook says and what really happened on 6 markets. Free PDF, sent to your inbox.

The PDF by email, plus our newsletter: new tests and results that change. Unsubscribe anytime. Privacy.

Frequently asked questions

What is the most profitable swing trading strategy?

No strategy is profitable in every market and period. In our test on the S&P 500 ETF from 2021 to 2026, none of the three rules beat random entries beyond chance, and none beat holding the ETF.

How long should you hold a swing trade?

It depends on the strategy's exit. In our tests, the median trade lasted 3 sessions for the pullback rule, 24.5 for the breakout rule and 56 for the 20/50 trend rule.

What is the 3-5-7 rule in swing trading?

It is an informal rule of thumb with no standard definition, and we found no primary source for it, so this page gives no version of it. A written risk limit per trade, such as the 2% rule below, is easier to test on your own trades.

What is the 5-3-1 rule in trading?

It is an informal rule of thumb with no standard definition, and we found no primary source for it. Whatever the version, it narrows what you trade; it is not a tested strategy.

What is the 2% rule in swing trading?

Risk no more than 2% of the account on one trade, meaning the loss if the stop is hit. With a $10,000 account, that is $200; the number of shares follows from the distance to the stop.

Is swing trading good for beginners?

It needs less screen time than day trading and can be run with orders placed after the close, which suits a beginner with a job. It still loses money with a bad rule or poor risk control, so start on paper or with small positions.

Sources

  1. StockCharts ChartSchool, Moving Averages: Simple and Exponential
  2. StockCharts ChartSchool, Price Channels (Donchian Channels)
  3. StockCharts ChartSchool, Fibonacci Retracements
, Founder of PatternsWizard

Passionate about finance, quantitative trading and data analysis. Amaury built the PatternsWizard test bench, which detects each pattern on full price history and measures what happened next against random entries.

PatternsWizard tests chart and candlestick patterns on full market history, against random entries. How we test patterns.

179 patterns testedRandom-walk self-test passed

Related guides

All trading concepts articles