Where to Place Stop Loss: Structure, ATR or a Fixed Percent

Three stops on the same 50.6 entry

A daily chart slides from 51.6 to 48.2 over seven sessions, holds, and recovers to close at 50.6. You buy the close at 50.6 (1). The swing low at 48.2 (2) is the last place buyers stepped in. The question is where the stop goes, and three methods give three different answers on this one chart.

The first method reads structure. A long is wrong once price trades back through the swing low, so the stop sits a little under 48.2, at 48.0. Distance from entry: 50.6 minus 48.0, or 2.6 a share. Fidelity’s support and resistance page describes support as a price area where buying has stopped a decline before; the stop belongs on the far side of it.

The second method reads volatility. ATR(14) on this chart is 1.1, meaning the average daily range over the last 14 sessions is 1.1 points. Two ATR is 2.2, so the stop goes at 50.6 minus 2.2, or 48.4. It lands 0.2 above the swing low. That is a coincidence of these numbers, and on another chart the same rule can land above or below structure.

The third method is a fixed percent. Two percent of 50.6 is about 1.0, so the stop goes at 49.6. It ignores the chart entirely. Price can revisit 49.6 on any ordinary pullback in a market whose daily range is 1.1, and the chart shows two closes below 49.6 inside the recovery. On this chart the fixed stop sits inside the noise.

Three stops, three distances: 2.6, 2.2 and 1.0. In percent of the entry, that is 5.1%, 4.3% and 2.0%. None of them is the stop. Each is a stop, and the next section shows what the distance costs.

The stop distance sets the share count, not the other way round

Start from the account, not the trade. A $10,000 account with a 1% risk budget can lose $100 on this trade. Shares equal the budget divided by the stop distance, rounded down. That is the whole calculation, and it runs the same way for contracts, lots or coins.

Read the last two columns together. The fixed stop holds 100 shares, more than two and a half times the structure stop’s 38, and puts $5,060 of a $10,000 account into one name. Every one of the three loses the same $100 if the stop fills at its level. The difference is how often that happens.

The nearer stop fails more often. A 1.0 stop in a market with a 1.1 daily range can be hit by one session’s noise with the trade idea still intact. The 2.6 stop survives that session and gives up fewer shares in exchange. That trade-off, more shares against more frequent exits, is the choice you are actually making when you pick a stop.

The percentage of the account at risk is a separate dial. CME Group’s note on the 2 percent rule presents 2% of capital as an example threshold that keeps one loss small, and it is an example, not a law. This page uses 1%. The position size calculator runs the division for any budget, entry and stop, in shares or units.

Reward scales with the same distance. If the target is 55.8, the structure stop gives 5.2 of reward against 2.6 of risk, or 2R; the fixed stop gives 5.2 against 1.0, or 5.2R on paper. The risk reward calculator shows both ratios side by side, and the paper ratio is only worth something if the stop is not hit first.

Below the wick or below the close: pick the level the market must break

The swing low candle on the diagram has a low of 48.2 and a close of 48.5. A stop under the wick sits at 48.0; a stop under the body could sit at 48.3. The wick version is 0.3 further away, so it allows fewer shares (38 against 43) and survives a retest that trades through 48.3 and closes back above it.

Wicks are where stops get run. A single daily bar can probe 0.2 under the prior low, fill resting orders there and close higher, which is why the long-tailed candles in the pin bar candlestick guide so often print exactly at an obvious level. A stop at 48.0 under the 48.2 wick still gets hit by that probe if it runs 0.3 deep. No offset is safe from every probe. Wider costs shares.

The close-based alternative is a rule, not a level: exit if the daily candle closes below 48.2, whatever the low prints. It ignores intraday probes, and it also means the fill comes at the next open, which can be well under 48.2. Size it as if the stop were at 48.0 or lower, because the exit price is unknown until it happens.

One placement rule works for both. Put the stop where the market has to do something it should not do if the trade is right. Under 48.2, a higher low has failed. At 49.6, nothing has failed yet. Support is a zone rather than a line, so read the whole cluster of lows before picking the number; the support and resistance page covers how levels are marked and rated on a screenshot.

A stop is an order, not a promise of the fill price

A stop-loss order becomes a market order once the stop price trades. Investor.gov’s page on order types spells this out: the fill can come at a price well away from the stop in a fast or gapping market. A stop-limit order avoids the bad fill by not filling at all, which in a gap is worse.

Take the structure stop at 48.0 with 38 shares. If the stock gaps from 48.6 to 47.0 on an earnings morning, the stop triggers at the open and fills near 47.0. The loss is 3.6 a share, about $137, not the $100 the plan said. The fixed stop at 49.6 with 100 shares fills at the same 47.0, and that loss is 3.6 × 100, or $360.

That is the second cost of the near stop. A gap jumps over every stop by the same distance, so the larger position loses more. Holding a daily-chart position through a scheduled event with 100 shares and a 1.0 stop is a different risk from the $100 in the table.

This page does not say which of the three is right. It depends on the market’s range, on whether you hold through events, and on how many stop-outs you can take in a row. Crypto perpetuals trade through weekends; single stocks gap; forex majors rarely gap outside the Sunday open. Not financial advice, and not a formula.

Checking stop placement on a chart screenshot

Before you set a stop from a screenshot, three things should be in the frame. Each one changes the number.

1. Enough bars to see the swing low and the lows around it, roughly 40 or more, so the 48.2 is a level and not the edge of the crop.
2. The ATR pane if you use a volatility stop; without it the 1.1 is a guess.
3. A legible price axis, because a stop read as 48.0 on a blurred axis can be 48.6 on the platform.

Crop the order panel and any account balance before uploading; the whole screenshot is what gets sent. On your own chart, check where the nearest swing low sits and how far it is from your entry in ATR terms.

TradeGPT reads the screenshot as an image. When the report includes an Execution section, the trade plan carries a stop loss, an invalidation condition and a stop distance in percent, and the levels in Key Levels carry a strength rating and a distance from price, which is the input the position size calculation needs. A plan can also be flat, with no stop at all.

The sample report shows how the stop, the invalidation and the level ratings are laid out. To compare the report’s stop with the one you would draw under the swing low, analyze your chart and check the first price it names against the axis.

Questions traders ask about stop loss placement

Where do you place your stop loss?

On the far side of the level that proves the idea wrong. For a long taken at 50.6 after a 48.2 swing low, that is a few ticks under the low, at 48.0. Then size the position from that distance: $100 of risk divided by 2.6 is 38 shares. The stop comes first and the share count follows.

How far should a stop loss be?

Far enough that ordinary daily noise does not reach it. With ATR(14) at 1.1, a 1.0 stop can be hit by one average session, while 2.2 (two ATR) or 2.6 (under structure) needs a real move. The wider stop allows fewer shares, 38 or 45 instead of 100, for the same $100 risk.

Should the stop go below the wick or the close?

Below the wick, if the stop is an order, because wicks are where probes reach. On a candle with a 48.2 low and a 48.5 close, 48.0 survives a probe that 48.3 does not. A close-based exit rule ignores probes, but the fill comes at the next open, so size it for a worse price than the level.

Does a stop loss guarantee the loss stays at $100?

No. A stop order becomes a market order when the stop price trades, and a gap fills it wherever the market opens. A gap from 48.6 to 47.0 turns the $100 plan into about $137 on 38 shares and $360 on 100 shares. The share count decides how much a gap costs.