Stop Loss Placement — Technical, ATR and Percentage Stops

A stop loss makes a simple promise: it caps your loss at a known amount decided in advance. But around that simple promise a whole mythology has grown — from "professionals always use stops" to "exchanges hunt your stops." The truth, as usual, is less comfortable for both camps: a stop loss is a tool that saves accounts in some strategies and measurably hurts results in others.

In this article we walk through three main methods of placing a protective order (technical, ATR and percentage) with BTC examples, and then honestly present both sides of the debate over whether a stop even makes sense — including an inconvenient backtest result from mean-reversion strategies.

Educational framework: this article is for educational purposes only and is not investment advice. Trading cryptocurrencies, especially with leverage, carries the risk of losing your entire capital.

What a Stop Loss Is and Three Ways to Place It

A stop loss is an order that closes a position once price reaches a defined loss level. It serves two functions: a mathematical one (you can't calculate position size without a known distance to the stop) and a behavioral one (it takes the decision to "admit you were wrong" out of your hands mid-move, when emotions run highest).

1. Technical Stop — Behind the Invalidation Level

You place the stop where your trade idea stops making sense: below support, below the last swing low, behind a demand zone (mirror-image for a short). The logic: if price gets there, it's not "noise" — it's proof the scenario failed.

A practical add-on: a buffer of about 0.3-0.5×ATR below the level itself, because the market loves to poke through an obvious line by a few dollars before reversing.

2. ATR Stop — Based on Market Volatility

ATR (Average True Range) measures how much an instrument typically moves. A stop sized off ATR automatically widens on a choppy market and tightens on a calm one:

SL (long) = entry price − ATR × multiplier
SL (short) = entry price + ATR × multiplier

The standard is a 14-period ATR with a multiplier of 2-3 (a smaller multiplier chokes the position in noise, a larger one gives back a lot on a reversal). A variant is the Chandelier Exit — a stop measured from the highest point of the move instead of from entry, popular as a trailing stop in trend-following strategies.

3. Percentage Stop — Fixed Distance

The simplest one: always, say, 2% from entry. The upside is zero decisions. The downside is serious — it ignores both chart structure and volatility. The same 2% on BTC is a reasonable distance in a calm week and the middle of ordinary noise in a week right after macro data. Treat it as a starting solution you grow out of.

What the Numbers Say

Let's start with the case for. The strongest argument for a stop is the asymmetry of drawdowns: a 50% loss requires a +100% recovery, a 70% loss requires +233%. A single unprotected position, held "because it'll come back," can set an account back years. On the crypto market, where altcoins can fall 80-90% and never return, and leveraged positions face liquidation, a hard per-trade loss limit is, for most traders, a condition for survival.

Now for the side the guides stay quiet about. Backtests published by QuantifiedStrategies indicate that in mean-reversion strategies, a classic stop loss usually made results worse — it lowered total profit and win rate instead of protecting capital. The mechanism is logical: a mean-reversion strategy, by definition, buys weakness — you enter when price has overextended down and wait for a bounce. A stop loss in that kind of strategy sells at exactly the point of maximum extension, often moments before the reversal move that was supposed to deliver the profit. In other words, the SL systematically cuts the trades that, statistically, were closest to paying off.

Important caveats, to be fair: the QS data comes from equity and index markets, and we're citing it from summaries — verify specific figures directly at the source before relying on them. And, as always: past results don't guarantee future ones. The takeaway from these tests isn't "ditch your stops," but rather: this is a trade-off between average outcome and tail risk. Without an SL, mean-reversion results have historically been better on average, but a single black-swan sequence (like a crash) can then take everything. In trend-following and breakout strategies it's the opposite — there, a stop is a natural part of the logic (a breakout that reverses is an invalidated signal), and the tests back it up.

How to Apply It Step by Step

Step 1. Match the stop type to the strategy type. Trend/breakout → a technical or ATR stop is non-negotiable. Mean reversion → instead of a tight SL, consider a time-based stop (close after N candles with no bounce), a wide "catastrophic" stop (e.g., 3-4×ATR), and above all a smaller position.

Step 2. Place the stop BEFORE opening the position and size the position from it.

Step 3. Numerical example on BTC (ATR stop):

• Account: $20,000, risk 1% = $200
• Long entry: $100,000
• ATR(14) on D1: $2,800
• Stop: 100,000 − 2×2,800 = $94,400 (5.6% from entry)
• Position size: 200 / 5,600 = 0.0357 BTC (~$3,570 notional)

Same setup with a technical stop: support at 97,200, buffer 0.5×ATR = 1,400, stop at 95,800. Risk per 1 BTC = $4,200, position = 200 / 4,200 = 0.0476 BTC. Both versions risk an identical $200 — only the notional differs.

Step 4. Set the rules for moving the stop in advance. Allowed: moving it in the direction of profit (trailing, moving to break-even after a 1R move). Forbidden: moving the stop farther away "to give the position a chance" — that's the moment you stop having a system.

Two notes on managing a stop mid-trade. Moving to break-even is often overhyped: it sounds like a "free position," but pulling the stop up to the entry price too early regularly turns good entries into scratches, because price retests the entry point more often than intuition suggests. A safer framework: break-even only after a move of at least 1R-1.5R in your favor, and before that the stop stays where the analysis put it. The second note is about trailing: a moving stop (e.g., a Chandelier at 3×ATR from the peak of the move) is really a change of exit strategy — it extends profits at the cost of giving back part of the move on a reversal — and deserves its own test, covered here.

Step 5. Plan what happens after the stop is hit. A stop executed according to plan is the end of a trade, not the end of analysis. Three acceptable paths: letting the instrument go, re-entering on a fresh, fully valid signal (not "because it bounced $200"), or a journal entry and a calm review. The fourth, most popular path is not acceptable: an immediate revenge position in the same direction, bigger, "because this time it's really going to work." Statistically, the trades opened fifteen minutes after a stopped-out position have the worst results in traders' journals — not because the market is out to get you, but because they're being made by adrenaline, not a plan.

When It Doesn't Work and the Most Common Traps

• A stop in an obvious spot. Right below support, on a round number (95,000, 100,000) — right where the crowd's stops sit, liquidity regularly reaches for them. A buffer and a volatility-based stop reduce the problem; they never eliminate it.
• A stop tighter than market noise. An SL closer than ~1×ATR from entry is statistically a coin flip — ordinary oscillation, not a change of scenario, will trigger it. If your account can't afford a wider stop with a sensible position, the instrument is too big for the account, not the stop too wide.
• Gaps and slippage. A stop doesn't guarantee your execution price — during a flash crash on an illiquid altcoin, the fill can land far below the level. A stop-limit protects against a bad price but might not fill at all. On crypto, the only complete safeguard is position size.
• The mental stop. "I'll close it manually when it gets there" — in practice, half of traders renegotiate with themselves the moment the level is touched. An order sitting on the book doesn't negotiate.
• One ATR multiplier for everything. 2×ATR on BTC and 2×ATR on a low-liquidity altcoin are entirely different levels of noise-triggering risk. Parameters should be tested per instrument and timeframe.
• Thin-liquidity hours. The crypto market runs 24/7, but not equally so: on weekends and at night (European time) order books are thinner, and moves can be sharper on lower volume. A stop set "right at the edge" on a Friday evening has a genuinely higher chance of getting wicked out than the same stop in the middle of the US session.

A stop loss isn't dogma — it's a tool with a measurable cost and a measurable benefit, and its value depends on the strategy. One thing stays constant: you need some mechanism to cap your loss. If not a price stop, then a time stop and a small position. The market doesn't forgive accounts that have neither.

FAQ

Rafał — Krypto Bez Ściemy

Trader and founder of Strefa Tradingu. He’s been breaking crypto down on YouTube for years — no hype, no signals, with a focus on market structure and risk management.

kryptobezsciemy.plInstagramYouTubeKlub Zero