Market Order: Certainty at Whatever Price
A market order executes immediately at the best available price, with no upper or lower bound on what that price turns out to be. You gain certainty of execution and give up control of the price, which is the exact opposite trade-off to a limit order.
How it works
A market order says: execute immediately, at whatever the best available price is. No limit, no condition, no waiting.
That is the exact inverse of a limit order. You are certain to trade and uncertain what you will pay. Which of those matters more depends entirely on the situation, and neither answer is generally correct.
A market buy lifts the offer; a market sell hits the bid. So the position starts down by the spread and pays it again on exit. That is the round trip, and it is charged whether the trade works or not.
What the cost actually is, as a share of the move
On this site’s shared 576-bar history the round trip is 2% of a median bar’s range — 0.0098 in the series’ own price units against a median bar of 0.493.
The same fixed cost is 45% of the smallest bar in the series. Bar ranges here run from 0.17 at the tenth percentile to 1.10 at the ninetieth, a 6.5-fold spread, and the cost does not move with them.
Which turns “what does trading cost?” into a much more useful question: what does it cost relative to what I am trying to capture? A scalper working the tenth-percentile bars is paying a materially larger share of the opportunity than a swing trader working the ninetieth. Same broker, same order type, different business.
Slippage is the gap between the quote you saw and the price you got. In a fast market the quote you clicked is already stale, and the fill lands beyond it. That is not a broker fault; it is what “immediately” means when the book is moving.
A large order does not fill at one price. It consumes the best offer, then the next, then the next, and the average fill is worse than the top of the book. On a thin instrument that difference can dwarf the spread.
In practice
Most venues charge a taker fee for removing liquidity and pay a small rebate for adding it. So a market order pays the spread, the taker fee and any slippage; a limit order pays none of those and may not fill at all.
Every stop order becomes a market order at its trigger price. Which means every stop inherits everything on this page — the spread, the slippage, the walking of a thin book — and it does so at exactly the moment conditions are least favourable, because a stop triggers when price is moving.
That is the single most under-appreciated consequence of the order type, and it is why a stop’s real cost is systematically larger than its notional distance suggests.
Volume is what sets the price. Heavy participation means a deep book and a small spread; thin participation means both are worse. Cost is a function of when you trade as much as of what you trade.
Spreads are widest in the first minutes of a session, which is when a great many people place market orders. The opening gap page covers why that combination is expensive.
On a long-horizon position the cost is invisible rather than absent. A spread that is 2% of a daily bar is a rounding error against a move measured in weeks — which is a genuine argument for longer holding periods that has nothing to do with being right more often.
What a market order is not
It is not a guarantee of price. It is a guarantee of execution, at a price the market decides.
It is not the same as the quote you saw. The quote is an offer to trade that can be gone by the time your order arrives.
It is not more expensive on every instrument. On a deep, liquid future the spread is a tick and slippage is rare; on a thin small-cap it can be several percent.
And it is not avoidable inside a stop. Anything that converts to a market order on trigger carries these costs.
When it fails
In a tight range the spread is most of the available move. Bars are small, the target is small, and a fixed cost that is 2% of a median bar can be a fifth of what a range-bound trade is trying to capture. That is where the cost decides whether to trade at all rather than how.
The second failure is size in a thin book. An order larger than the top few levels fills progressively worse, and the average price bears no relationship to the quote.
A third is the news moment. Placing a market order into a scheduled release is asking to be filled during the widest spread of the day.
A fourth is the stop that triggers into nothing. A stop is a market order arriving exactly when the book is thinnest, which is why stop fills so often look worse than the level.
And a fifth is treating cost as a constant. It is fixed in price units and variable as a share of the opportunity, and the second is the one that matters.
The original data
On this site’s shared 576-bar history the round-trip cost is 0.0098 in price units — 2% of the median
bar range of 0.493, 45% of the smallest bar of 0.022, and more than 10% of the bar’s range on 15 of the
576 bars. Bar ranges run 0.17 at the tenth percentile to 1.10 at the ninetieth, a ratio of 6.5. All of
it is in research/series-measurements.json, produced by site/measure_series.py.
The ratio between those two figures — 2% and 45% — is the whole argument for measuring cost against the bar rather than against the account. A cost quoted as a percentage of capital tells you nothing about whether a setup is worth taking; a cost quoted as a percentage of the bar you are trying to capture tells you immediately. Compute your own round trip in price units, then divide it by your instrument’s tenth-percentile bar range. If that number is large, quiet conditions are unprofitable for you before any question of skill arises, and knowing that in advance is worth more than any entry technique.
Related
Order types is the parent page. Limit order is the alternative and the opposite trade-off. And bid-ask spread is the cost this order type pays in full.
I stopped worrying about order types the day I worked out what my costs were as a share of the bar I was trying to capture. On the moves I was actually trading, the spread was not a rounding error - it was a fifth of the target, and that changed which setups were worth taking.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.