Most traders think about stops backwards. They decide how many shares to buy, then slap a stop "10% below" or "$200 of risk" onto it — an arbitrary line that has nothing to do with the actual trade. The result is stops that get hit by normal noise, or stops so wide the loss blows the risk budget. The fix is to reverse the order: place the stop where your idea is wrong, then size the position to fit.
A stop marks where you're wrong — not where it hurts
Every trade has a price at which the reason you took it no longer holds. Bought a breakout? If it falls back below the level it broke out from, the breakout failed — that's your stop. Bought a bounce off support? If support gives way, the thesis is dead. The stop belongs just beyond that structural level (a swing low, support/resistance, a moving average you're trading off of), because that's the price that objectively invalidates the idea. Where the loss "hurts" or a round number sits is irrelevant to the chart.
Common ways to place the stop
- Structure-based (best default): just below the recent swing low (for longs) or above the swing high (for shorts). It's where other traders' stops cluster and where the setup is genuinely broken.
- ATR-based: a multiple (say 1.5–2×) of the Average True Range from entry. This adapts to the stock's normal volatility so you're not stopped out by routine wiggles — useful on choppy names.
- Percentage: a fixed % below entry. Simple, but blind to structure — fine as a hard maximum, poor as a primary method.
Whatever the method, the stop must be far enough that normal noise won't trigger it, but close enough that when it hits, you were genuinely wrong. Too-tight stops are the most common way good trade ideas die from a thousand small cuts.
Now size around the stop — this is the whole trick
Once the stop is placed by structure, you know your per-share risk (entry − stop). Your position size then flows from your fixed dollar risk:
Shares = (Account × Risk %) ÷ (Entry − Stop), rounded down
A wider (but correct) stop simply means fewer shares for the same dollar risk — not a bigger loss. This is the key insight: you never widen your risk to accommodate a stop; you shrink the position instead. The stop-loss calculator shows the exact dollar and percentage loss at your stop for any share count, and the position size calculator turns your structural stop into the right share count automatically. Together they let the chart decide the stop and the math decide the size.
A few rules that keep stops honest
- Set it before you enter, and honor it. A stop you move down "just this once" isn't a stop.
- Don't cluster on the obvious round number where everyone's stop sits — give it a little room beyond the level.
- Check the reward first. If a structurally-correct stop makes the risk/reward ugly, the trade isn't worth taking — skip it rather than using a bad-but-comfortable stop.
Place the stop where you're wrong, size so being wrong costs exactly what you planned, and a stopped-out trade becomes a small, expected cost of doing business instead of a gut punch.