Chapter 4 — Stop-Losses: Decide Your Exit Before the Market Does

Chapter 3 made you promise to write down your exit before you enter. This chapter turns that promise into an actual order that fires while you sleep — because hoping a losing trade "comes back" is not a strategy, it's a bedtime story.

4.1 What a stop-loss actually is

A stop-loss is an order that sits dormant until price touches a trigger you chose — then it fires and sells your position automatically. It's you, in advance, answering the only question that matters about any trade: "At what price is this idea wrong?"

Every trade has an invalidation point whether you set one or not. The only choice is whether you pick it calmly beforehand, or the market picks it for you at 4am while you're mid-panic scrolling.

4.2 The two flavours: stop-market and stop-limit

Exchanges dress them in slightly different names (Binance calls them Stop-Limit and Stop-Market under the "Stop" tab; Bitget and OKX group them under "Trigger" orders), but there are only two mechanics:

If you do use a stop-limit, set the limit price a little below the trigger (e.g. trigger $94,000, limit $93,700). That gap is the cushion that keeps a fast market from skipping over your order entirely.

4.3 Where to put the stop (the part everyone gets wrong)

• At the invalidation level, not at a feel-good number. Ask: "what price would prove the reason I bought is dead?" Usually that's just below the support level or dip-low you bought against. If price breaks it, the idea failed — you want out.
• Not at the obvious round number. Half the internet's stops sit exactly at $90,000.00. Fast wicks love sweeping those levels before price reverses. Place yours a step beyond the crowd (e.g. $89,600, below the level rather than on it).
• Not so tight that noise kills you. Crypto routinely wobbles 3–5% for no reason at all. A stop 1% away isn't risk management, it's a donation schedule.
• Not so wide that it's meaningless. A stop 40% away just means "I'll think about it later". If the invalidation level is that far from your entry, the entry is wrong — wait for a better price (Chapter 2's ladder).

Level 2 of the 2% rule: size from your stop

In Chapter 3 you traded a flat 2% of portfolio. The grown-up version: risk at most 2% — meaning the loss if your stop fires stays under 2% of portfolio. The formula:

Example (numbers illustrative): portfolio $10,000, so max risk $200. You buy at $100,000 with a stop at $92,000 — an 8% stop distance. Position size = $200 ÷ 0.08 = $2,500, and if the stop fires you lose $200. Tight 4% stop? Size can be $5,000 — the risk is identical. Wide stop, small size; tight stop, bigger size. Either way, the most you can lose is the same $200.

While you're learning, either version of the rule keeps you alive. Flat 2% positions are simpler; risk-based sizing is what you graduate into.

4.4 Placing one, step by step

Say you bought 0.002 BTC at $100,000 (illustrative), and your invalidation is a break below $92,000:

1. Open the Spot market for your pair and find the order form's Stop / Stop-Limit / Trigger tab.
2. Side: Sell. You're protecting a position you hold.
3. Set the trigger price: $92,000 — the "if we get here, it's over" line.
4. Stop-market: done, set the amount. Stop-limit: set the limit a cushion below — e.g. $91,600 — then the amount.
5. Amount: your full position for that trade (partial stops are a later refinement).
6. Confirm, then find it under Open Orders. It reserves your coins but executes nothing until the trigger touches.
7. Leave it alone. The only permitted edit is moving a stop up to protect profit — never down to "give it room". Moving stops down is just losing money in slow motion.

4.5 The stop's twin: take-profit, and OCO orders

The same trigger mechanics work in your favour on the way up — a take-profit order sells automatically at your target. Better yet, most exchanges (Binance, Bitget, OKX) offer OCO — "One Cancels the Other": one bracket holding both your take-profit limit and your stop-loss. Whichever fires first cancels the other. Your trade from 3.4 — entry, target, stop, all written down — becomes one order:

Set the OCO the moment your buy fills, and the trade runs itself — no screens, no 4am panic, no "she'll come back" bedtime stories.

4.6 Hard truths and fine print

• Stops only work where the coins are. An exchange stop-loss can't protect coins on your Ledger — and that's correct: the stack (Chapter 3) isn't traded, so it doesn't need stops. Only the float gets bracketed.
• Flash crashes slip. In a violent wick a stop-market can fill noticeably below the trigger. Unpleasant, survivable — that's why size stays small.
• Stop hunting is real, but it mostly eats the predictable. Obvious round numbers and exact support levels get swept. Being slightly less obvious is most of the defence; the rest is accepting the occasional unlucky wick as a cost of doing business.
• Trailing stops follow price up at a fixed distance and lock in gains automatically. Handy in strong trends, fiddly to tune — file under "after you've run ten normal stops".
• Getting stopped out then watching price recover happens to everyone. It stings. It's still cheaper than the times it doesn't recover. You're buying insurance, and insurance costs premiums.

4.7 Chapter checklist

• Before every entry I can state the trigger price where the idea is wrong
• My stop sits beyond the obvious level, not on the round number
• I know the difference between stop-market (certain exit) and stop-limit (certain price, maybe no exit)
• I can size a position from my stop distance so the worst case stays ≤ 2% of portfolio
• Winning trades get their stop moved up — no stop ever gets moved down
• My open trades carry an OCO bracket: target and stop, set the moment the buy fills
• The Ledger stack stays out of all of this

4.8 The stop-loss you can't rely on: outages and flash crashes

A stop-loss lives on the exchange's servers, not on the blockchain and not on your Ledger. That single fact creates two failure modes worth knowing about before you assume "I've got a stop, I'm covered."

• Exchange outages happen, usually exactly when you need them not to. Every major exchange has, at some point, gone down or throttled trading during extreme volatility — the exact moment your stop order needed to fire. Using a large, reputable exchange rather than a brand new or thinly-resourced one reduces the odds; nothing eliminates them.
• Flash crashes blow through stops, not just past them. A stop-market order guarantees an exit, not a price. In a genuine flash crash, price can gap straight through your stop level and fill you far worse than expected — sometimes brutally so — before recovering minutes later. Rare, but it's exactly the scenario a stop exists for, so it's worth knowing it isn't a perfect guarantee.