The stop: where it goes and why

Every trade has a reason, an idea you can say in one sentence. Marc's: "Bluefin fell for three weeks, found a low at 37.32 and bounced; the bounce holds as long as the price stays above that low." The idea itself tells you where it fails: below 37.32 the bounce has failed. That is where the stop belongs.

Nora asked a different question: "how much would I like to lose?" She wanted to lose at most 40 dollars on 100 shares, so she put her stop 40 cents under the entry. The market knows nothing about her 40 dollars; her stop sat at a price that meant nothing for the idea.

So the sequence is: the idea decides the stop, the stop decides the loss per share, and only then the risk rule decides how many shares fit. A comfortable loss is controlled by the size of the position, never by moving the stop.

Once you know the price where the idea fails, don't put the stop exactly on it. Prices often poke a few cents through an old low and turn back up, and a stop sitting right on the level gets taken by that poke even though the idea is still alive.

So the stop goes a little beyond the level. Marc's idea fails below the 37.32 low, and his stop sits at 37.12, 20 cents lower. The stop now fills only if the low really breaks.

How big should the buffer be? There's no magic number and no rule from the broker. Look at how far this price usually wanders past levels before turning, and stay outside that. A calm instrument needs a small buffer; a jumpy one needs more.

Note the cost: the buffer moves the stop away from the entry, so the loss per share gets a little bigger. That's fine, because the size of the position, not the stop, is what keeps the total loss in line.

With the stop placed, measure the loss honestly: (entry − stop) × number of shares, plus costs.

Marc bought 100 Bluefin Energy at 38.65 with his stop at 37.12. The distance is 38.65 − 37.12 = 1.53 a share, so the planned loss is 1.53 × 100 = 153 dollars before costs. On his 10,000 account that's about 1.5%.

That number comes from the market, not from a wish. Now compare it with your risk rule. If 153 is more than your rule allows, there are two honest fixes: buy fewer shares (lesson 3 of this module turns this into a formula) or skip the trade. The dishonest fix is to pull the stop closer until the loss looks right, which puts it back inside the price's normal wiggle.

If the right stop is so far away that even a few shares break the rule, the trade simply doesn't fit your account today.

To know whether a stop sits inside ordinary noise, measure the noise. A simple gauge is the average daily range: for each day take the high minus the low, then average those numbers over recent days.

For five days with ranges 0.62, 0.44, 0.51, 0.39 and 0.54, the sum is 2.50 and the average is 2.50 ÷ 5 = 0.50. Bluefin Energy averaged the same 0.50 a day over the 20 days before Marc's and Nora's entry.

Compare your stop distance with it. Nora's stop was 0.40 under her entry: less than one ordinary day's range, so an ordinary day can reach it without anything being wrong. Marc's stop was 1.53 away, about three ordinary days, and placed under a real low.

The range is not a rule for where the stop goes (the idea decides that). It's a check: if your stop is closer than a normal day's movement, expect noise to hit it. Module 6 uses the range again for sizing stops.

A stop can be hit for two different reasons. Either the idea really failed, and the stop did its job, or the price just made an ordinary move and the stop was in the way.

Nora bought Bluefin Energy at 38.65 on 30 January with her stop at 38.25. On 2 February the low was 38.07, so her stop filled: minus 40 dollars, as planned. But the idea relied on the 37.32 low, which was never touched. The next day the price closed at 38.76, above her entry, and on 13 of the next 14 days it closed above 38.65.

Her idea wasn't wrong; her stop was inside the normal wiggle. A string of such exits feels like bad luck, but it's a placement problem: many small losses on trades whose ideas never failed.

The test after any stopped trade: did the price reach the level that disproves the idea? If not, the stop was in the wrong place.

A stop is an instruction to sell once the price reaches a level. It needs someone to trade at or near that level. Usually there is. But sometimes the price jumps: the market closes at one price and opens far away, with no trades in between. That jump is a gap.

Bluefin Energy reported results before the open on 23 February. The day before closed at 38.95; the next morning opened at 36.15, a gap of (36.15 − 38.95) ÷ 38.95 ≈ −7.2%. Nobody traded at Marc's stop of 37.12, so it sold at the first available price, the open.

His planned loss was 153 dollars. His real loss: (38.65 − 36.15) × 100 = 2.50 × 100 = 250 dollars, 97 more than planned.

So a stop limits the loss in normal trading, not in a jump. The planned loss is a floor for what can go wrong, not a ceiling. Gaps cluster around company reports, big news and weekends.

You can't prevent a gap, but you can expect one and keep it small.

Know the calendar. Shares and funds gap mostly around company reports, big news and weekends. Check the dates before you enter. Made-up coins like Spark Coin have no reports but trade all night, so a jump can come at any hour.

Hold less into the event. Fewer shares, or none, through a report means a gap hurts less. Marc's 7.2% gap cost him 250 dollars on 100 shares; on 50 shares it would have cost 125.

Plan for a bad gap, not just the stop. Ask: if it opens 7% against me, is the loss still bearable for the account? If not, the position is too big for this moment.

What doesn't help: moving the stop further away "to give it room" (the planned loss just grows, and the gap can still jump it) or trusting that the stop fills at its exact price.

Placing a stop takes five short steps, and their sequence matters.

1. Write the idea in one sentence. Marc keeps a journal line for each trade: "Wrong if…". If you can't finish it, you have no stop and no trade.
2. Find the price where the idea is clearly wrong, such as below a recent low.
3. Put the stop a little beyond it, outside the normal daily wiggle.
4. Work out the loss if the stop fills, and allow for a gap around reports.
5. If the loss is too big, buy fewer units or skip the trade. Never move the stop to make the numbers fit.

Starting from "how much would I like to lose" turns the list upside down: the stop lands wherever the money feels right, usually inside the noise. Nora's line would have read "Wrong if… I lose 40 dollars", which says nothing about the price.

How it fits together: the idea sets the stop, a buffer and the daily range keep it out of the noise, and gaps can still make the loss bigger. Next lesson: how many shares fit between your risk rule and that stop.