Three traders open the same long at the same price with the same initial stop. One trails with ATR, one exits when price closes below a moving average, and one moves the stop under each new higher low. Fourteen candles later, they have three different results, and none of them is “wrong.” This article runs that experiment on one illustrative trade so you can see exactly where each method gains and loses, then shows how to pick one for the way you trade.
The setup
All numbers below are illustrative, not historical data. ETH/USDT, 4-hour chart.
- Entry: long at 2,400 (close of bar 1).
- Initial stop: 2,330, below the swing low that defines the setup. Initial risk = 70, so 1R = 70.
- ATR(14): held at 70 for simplicity. In a live chart it would drift a little each bar.
- Every method starts from the 2,330 stop and only ever moves it up. That one-way ratchet is what makes a stop “trailing,” as the SEC’s investor bulletin on stop and trailing stop orders describes: the trigger follows favourable moves and stays put when price moves against you.
The three rules
- ATR trail: after each bar, stop = highest high since entry − 2 × ATR (2 × 70 = 140). Triggered intrabar when the low touches it.
- Moving-average exit: exit at the close of the first bar that closes below the 20-period EMA. No intrabar trigger; the initial 2,330 stop stays as a disaster stop.
- Structure trail: when price makes a new high above the previous swing high, move the stop to 10 below the most recent pullback low. Triggered intrabar.
The trade, bar by bar
| Bar | High | Low | Close | EMA(20) | ATR stop | Structure stop |
|---|---|---|---|---|---|---|
| 1 | 2,410 | 2,370 | 2,400 | 2,350 | 2,330 | 2,330 |
| 2 | 2,445 | 2,390 | 2,435 | 2,358 | 2,330 | 2,330 |
| 3 | 2,470 | 2,420 | 2,460 | 2,368 | 2,330 | 2,330 |
| 4 | 2,465 | 2,425 | 2,430 | 2,374 | 2,330 | 2,330 |
| 5 | 2,490 | 2,428 | 2,485 | 2,384 | 2,350 | 2,415 |
| 6 | 2,540 | 2,475 | 2,530 | 2,398 | 2,400 | 2,415 |
| 7 | 2,560 | 2,500 | 2,510 | 2,409 | 2,420 | 2,415 |
| 8 | 2,530 | 2,455 | 2,470 | 2,415 | 2,420 | 2,415 |
| 9 | 2,580 | 2,465 | 2,575 | 2,430 | 2,440 | 2,445 |
| 10 | 2,620 | 2,560 | 2,600 | 2,446 | 2,480 | 2,445 |
| 11 | 2,610 | 2,540 | 2,550 | 2,456 | 2,480 | 2,445 |
| 12 | 2,560 | 2,470 | 2,490 | 2,459 | hit | 2,445 |
| 13 | 2,500 | 2,430 | 2,462 | 2,460 | out | hit |
| 14 | 2,470 | 2,400 | 2,420 | 2,456 | out | out |
The stop columns show the level in force after that bar closes. How the stops moved:
- ATR: after bar 10 the highest high is 2,620, so the stop is 2,620 − 140 = 2,480. Bar 12’s low of 2,470 reaches it.
- Structure: bar 4 made a pullback low at 2,425. When bar 5 broke above bar 3’s high (2,470), the stop moved to 2,415. Bar 8’s pullback low was 2,455; when bar 9 broke above bar 7’s high (2,560), the stop moved to 2,445. Bars 11 and 12 pulled back without a new high, so the stop stayed put until bar 13’s low of 2,430 hit it.
- EMA: bar 13 closed at 2,462, just above the 2,460 EMA, so the trade stayed open. Bar 14 closed at 2,420, below 2,456. Exit at the close.
The scoreboard
| Method | Exit | Gain per ETH | Result in R | Share of the 220-point peak move kept |
|---|---|---|---|---|
| ATR trail (2×) | 2,480 on bar 12 | +80 | +1.14R | 36% |
| Structure trail | 2,445 on bar 13 | +45 | +0.64R | 20% |
| EMA(20) close | 2,420 on bar 14 | +20 | +0.29R | 9% |
The best possible exit was near the 2,620 high, a gain of 220 or about 3.14R. No trailing method gets close to that, and none is designed to. A trailing stop trades part of the move for staying in until the trend visibly fails. If R-multiples are new to you, our risk-to-reward guide explains them.
What changes if the trend resumes
Now rerun bars 12 onward with a different ending. Suppose bar 12 still dips to 2,470, but bar 13 holds above 2,450 and bars 14 to 20 rally to 2,800.
- ATR trail: already stopped out at 2,480 on bar 12. It misses the rally entirely unless the trader has a re-entry rule.
- Structure trail: stop at 2,445 is never touched. Bar 12’s low of 2,470 becomes the next pullback low, and the stop moves to 2,460 once price breaks above 2,620.
- EMA exit: if closes keep holding above the rising EMA, it stays in as well.
Same entry, same stop, opposite ranking. This is the honest answer to “which trailing stop is best”: the one that won in the first scenario lost in the second. The tighter the trail, the more often you keep a good chunk of a move that reverses, and the more often you get shaken out of one that continues.
Side-by-side: what each method really responds to
| ATR trail | Moving-average exit | Structure trail | |
|---|---|---|---|
| Reacts to | Distance from the peak, scaled by volatility | Closing price vs. average price | Breaks of the trend’s own swing lows |
| Moves the stop | Every new high | Every bar (the average moves) | Only after a confirmed higher low |
| Strength | Mechanical, adapts to the coin’s volatility | Smooth, ignores single-bar wicks | Matches how the trend is defined |
| Weakness | Can exit on a normal pullback in a fast trend | Lags; exits after a close, often well below the peak | Swing points are subjective; stops sit where many others do |
| Main parameter | ATR length and multiplier | MA type and length | Swing definition and buffer |
For the ATR mechanics, including true range and Wilder smoothing, see our ATR indicator guide. For the swing-point rules behind the structure trail, see market structure: higher highs, lower lows, BOS and CHoCH.
Failure modes worth planning for
- Gaps and fast candles. All three methods can fill below their level. A stop order becomes a market order once triggered, and the bulletin linked above warns the execution price can deviate from the trigger.
- Close-based exits take bigger hits. The EMA exit waits for a close, so a single crash candle can travel far past the average before you act. Keep a hard stop resting on the exchange.
- Obvious structure stops cluster. Ten points under a clean higher low is where many traders put theirs. Price often wicks into those levels before the trend resumes, so a small extra buffer can matter.
- Switching methods mid-trade. Tightening to the ATR stop when nervous and loosening to the EMA when hopeful turns three rules into none.
How to choose: a short decision procedure
- Name your holding period. Hours to a few days: ATR or structure on your entry timeframe. Weeks: moving-average exit or a wider ATR multiple on a higher timeframe.
- Decide what you hate more: giving back open profit (choose tighter ATR) or getting shaken out before the big move (choose structure or MA).
- Write the rule so someone else could follow it. “Stop = highest high − 2 × ATR(14), updated at each 4h close, never lowered” is testable. “Trail under structure” is not, until you define a swing.
- Test it on 30 or more past trades, not one. For each, log entry, initial R, exit under each method, and the peak price. Compare the average R and the share of the peak captured. A trading journal makes this a spreadsheet job rather than memory.
Financial disclaimer: This article is for educational purposes only and is not financial, investment, or trading advice. Trading involves substantial risk of loss. Example figures are illustrative.