Market order vs limit order: what each one actually costs, measured

Most explanations of this stop at the definitions: a market order fills now, a limit order fills at your price. Both true, and neither tells you which one to press. The useful version is what each one costs. One of them has a price you can measure, and we measured it across 264 recorded US sessions and 35,207,932 quotes. The other has a cost that is real but is not money, which is exactly why people underestimate it.

1. What you are actually instructing

A market order specifies quantity, not price. You are telling the broker to fill you now at whatever the book offers. You control whether it fills. You do not control what you pay. A limit order specifies price, not quantity filled. You are telling the broker to fill you only at your price or better. You control what you pay. You do not control whether it fills. That is the entire trade-off. Market order vs limit order is really a choice about which uncertainty you would rather carry: an unknown price, or an unknown fill. Every practical difference below comes out of that one decision.

2. What a market order costs

Crossing the spread is the base cost, and it is not one number. Median spread from our recording, in basis points, by time of day and price band: A market order in the first half hour pays the worst pricing of the day, in every band. That is also when the most people are trading. On top of that, 32.2% of all quotes we recorded were already one cent wide, which is as tight as the rules allow. On a $1.50 stock that single penny is a 0.67% round trip. A market order there is not slightly expensive, it is structurally expensive, and no order routing fixes it. The full measurement is in what a US stock actually costs to trade.

3. The size limit nobody mentions

A market order does not pay the quoted price. It pays the quoted price **for as much as is sitting there**, then the next price, then the next. We walked market orders against the recorded ladder and measured how much the top of book actually holds: If your order is bigger than that number, part of it fills worse than the price you saw, and you will experience it as slippage. It is not slippage in any mysterious sense — it is you buying the second and third best offers. It gets worse with size. A $5,000 market buy in a $1-5 name costs 88bp one way against 45bp for a $1,000 order. And a $25,000 order could not be filled inside ten visible levels 28.7% of the time. Full depth measurement here: what Level 2 market data actually shows.

4. What a limit order costs

A limit order looks free. It is not, and the cost does not appear on a statement. You pay in fills you do not get. Your limit sits at a price. If the stock runs the way you wanted without coming back, you never got in. The trade that would have worked is the one that did not fill, and you will never see it in your P&L. The fills you do get are selected against you. A resting buy limit fills when a seller is willing to hit it, which happens more readily when the price is heading down. Sitting passively means the market chooses when to trade with you, and it chooses the moments that suit the other side. Nothing in this is dishonest, it is simply what passive execution means. We have not put a number on that cost, because it requires order-level fill data we do not have. Be suspicious of anyone who quotes you one without saying how they measured it. What we can say is that it is real, it grows the further you place your limit from the touch, and it is why "always use limits" is bad advice for time-sensitive entries.

5. The order type does not change your commission

A common assumption is that limit orders are cheaper in fees, because that is how crypto exchanges work — maker and taker rates differ there. On US stocks, for the commission-free brokers and the fixed per-share plan we priced across 36,947 real round trips, the commission is the same either way. Order type changes what you pay the market, not what you pay the broker. The exception is plans that pass exchange fees and rebates through to you, where adding or removing liquidity can change the bill. If you are on a tiered or pro plan, check your own schedule rather than assuming.

6. Stop orders are market orders in disguise

This one costs people real money. A standard stop order is not a separate species. When the stop price is touched, it **becomes a market order**. So it pays the market-order cost from sections 2 and 3, and it pays it at the worst possible moment: during a fast move, when spreads are widest and the book is thinnest. The reason stops so often fill far from the stop price is not broker mischief, it is the definition of the instrument. A stop-limit avoids the bad price by adding a limit, and in exchange takes on the risk from section 4: in a genuine collapse the price can move straight through your limit and leave you holding the position you were trying to exit. Neither is wrong. But you should choose knowing that one guarantees the exit and not the price, and the other guarantees the price and not the exit.

7. Which to press

Not rules, just what the numbers above imply. Cheap stocks lean strongly toward limits. When the minimum possible spread is already 0.67% of the trade, paying it twice a day is not survivable for most targets. Time-critical entries lean toward market. If the reason for the trade expires in seconds, a limit that does not fill has already cost you the whole idea. Size relative to the touch decides more than either. If your order is several times the numbers in section 3, no single order type saves you. Splitting the order matters more than choosing between market and limit. Avoid the open when the method allows it. The first thirty minutes is the most expensive window in every price band we measured, for both order types. Set the limit price against the spread, not against a round number. A limit one cent inside a one-cent spread is not a limit, it is a market order with extra steps.

8. What we did not measure

Limit order fill rates. We measured the cost of crossing, not how often passive orders get filled. That needs your own order records, and it is worth building. Price improvement. Some brokers route in ways that fill you better than the quote. We did not model it, which means our market-order costs are, if anything, slightly pessimistic for those brokers. Hidden and iceberg liquidity. Our recording sees ten displayed levels. Size resting out of sight would make market orders look better than we measured. The honest summary is that a market order has a knowable price and a limit order has an unknowable one, and most traders reverse that: they treat the market order as the simple default and the limit as the clever choice, when it is the market order whose cost you can look up in a table.