Risk & execution
Orders, spreads, and slippage: from signal to fill
Compare market, limit, stop, and stop-limit orders, and learn why execution costs can change a trading signal’s apparent performance.
By atradeaday · Published
A signal price is not an execution report. Between identifying a setup and receiving a fill, the market can move, available size can change, and the venue can apply fees or order restrictions. A strategy that ignores this gap may look better in a chart than in an account.
Match the order to the intended behavior
A market order prioritizes execution against available liquidity, with an uncertain final price. A limit order sets a price constraint but may remain unfilled. A stop order is triggered under the venue's rules and commonly becomes a market order. A stop-limit order becomes a limit order after triggering and can fail to exit during a fast move.
The SEC's investor bulletin on stop orders explains why stop prices are not guaranteed execution prices. Specific mechanics differ across products and venues, so verify the relevant order documentation.
Separate the components of cost
The bid-ask spread is the difference between quoted selling and buying prices. Slippage compares the actual execution with a chosen reference. Fees are explicit charges. Funding or financing can apply over time. Record these separately so that a deterioration in liquidity does not look like a change in signal logic.
For a hypothetical quoted market of 99.9 bid and 100.1 ask, buying and immediately selling one unit at those quotes loses 0.2 before fees. A historical strategy that buys and sells at the midpoint has omitted that friction.
Do not overstate limit-order fills
A candle touching a limit price does not prove your order would have filled. The quote may represent a different side of the market, and other orders may be ahead in the queue. A large order can also fill partially.
For candle-only research, document a conservative fill model and test sensitivity to worse assumptions. If the strategy depends on queue placement or millisecond timing, candles are not sufficient data to validate it.
Check operational rules
Before automating a trade plan, verify minimum size, price increments, time-in-force, reduce-only behavior where relevant, and rejection handling. A stale order after a cancelled signal creates exposure the strategy did not intend.
Keep the signal timestamp, submission timestamp, acknowledgement, and individual fills when evaluating live results. Aggregated profit alone cannot explain whether a problem came from the model or execution.
Continue with risk sizing and backtesting to make the same assumptions consistent from research to execution.