Stop-loss vs stop-limit: what each one costs on a fast drop

The usual way this choice gets explained is that a stop-loss guarantees an exit but not a price, and a stop-limit guarantees a price but not an exit. That is correct, and it settles nothing, because it does not tell you how much price you give up or how often the exit fails to happen. Both answers depend on one thing: what the market looks like in the second your stop fires. That second is not an average second, and we have it recorded. What follows uses our own measurements of 4,400,640 recorded seconds across 38 sessions, and every number here links back to the study it comes from.

1. What each order actually does

A stop-loss, more precisely a stop-market order, is an instruction held by your broker. When the stock trades at or through your stop price, it converts into a market sell. A market sell takes the best bid standing at that instant. Your stop price is the trigger, never the fill. A stop-limit converts into a limit sell instead. It fills only if a bid exists at or above your limit price, and only for as much size as is standing there. If nothing is there, nothing happens and you still own the stock. So the real question is what the bid looks like at the moment of conversion.

2. The second your stop fires is not a normal second

We measured the quoted spread at every second where the last price fell below the previous second, grouped by how far it fell. The spread you must cross is 4.2 times wider on the biggest falls, and the money standing at the front of the book is smaller at the same time. Both move against a seller together, which is the part most explanations miss. Measured session by session rather than pooled, the thinning held on 35 of 37 sessions. This is the whole reason the two order types diverge. They are not two ways of leaving the same market. They are two responses to a market that has changed underneath you.

3. What a stop-loss actually costs

Because it converts to a market sell, a stop-loss pays whatever the spread is at that instant. We measured the distance from the last print down to the best bid on every falling second. The typical case is mild. In 28.2% of falling seconds the bid was at or above the last print, so a market sell would have received the print price or better. The problem is not the median, it is the shape: the bad cases are concentrated in exactly the seconds where the fall was largest, which are the seconds your stop was placed for.

4. What a stop-limit actually refuses

A limit price is a refusal to cross a spread wider than you chose. Read that against the table in section 2 and the difficulty becomes obvious. The refusal binds most often precisely when the spread is widest, which is when the stock is moving fastest, which is when you most wanted out. There is a second failure that gets less attention. Even when a bid exists at your limit, the size resting there is finite. On falls of 1.00% or more the median second had $720 at the best bid, and the lower quartile had $190. An order larger than that does not fill at your limit. It fills partly and leaves you holding the rest, in a stock that is still falling.

5. Side by side

Neither column is the safe one. They fail in different directions, and the direction you can survive is a decision about your account, not about order types.

6. Choosing between them

Start with size. If the position is worth more than the money standing at the front of the book on a fast fall, both orders behave badly, one by walking down and one by stalling. Between $2 and $20, where most day trading scan lists live, the median falling second offered between $1,530 and $1,740 at the best bid. A position several times that size has no clean exit at the quoted price, and the fix is smaller size rather than a cleverer order. Then look at what the stock quotes when nothing is happening. Cheap stocks quote wider: under $2 the mean spread on falling seconds was 1.22% against 0.72% for $5 to $20 names, and the bid sat 0.34% below the last print at the median against 0.13%. A stop-loss on a sub-$2 stock is a more expensive instrument than the same order on a $10 stock, before anything goes wrong. Finally, be honest about whether you are watching. A stop-limit that does not fill is only safe if somebody notices and acts. If the answer is that you are at work, the unfilled order is not protection, it is a position you no longer have a plan for.

7. What neither order fixes

A quote can be pulled in the milliseconds between when you saw it and when your order arrives, and our one-second recording cannot see that. Every figure here is therefore the optimistic case, not a promise. Beyond the front of the book there is more interest. The median falling second showed $145,700 of total bid interest against $1,810 at the touch. That depth is not a rescue, it is a price schedule: each further level is a worse fill, and a large market sell simply walks down it. The one thing that does change the arithmetic is the size of the position relative to the book. That is decided before the trade, not during it.

8. Where these numbers come from

The spread and fill distances are from https://traderwe.com/research/stop-loss-slippage-the-spread-widens-with-the-drop-that-triggers-it and the resting size figures are from https://traderwe.com/research/liquidity-when-a-stock-drops-the-bid-thins-on-95-of-sessions Both measure the same sample: 745 symbol-days across 38 sessions drawn from 264 recorded sessions, 4,400,640 quoted seconds of which 387,948 were falling seconds, regular session only, on the US stocks our own scan surfaced each day. Those are the day's active movers rather than a cross-section of the market, which is the population these orders are usually placed in.