RESEARCH AND PRACTICE
Crypto execution shortfall: from decision price to actual fills
Audit the gap between a crypto decision and its fills, separating market movement, execution prices, fees and the opportunity cost of unfilled quantity.
A strategy can identify a favourable price move yet produce a disappointing account result. To understand why, retain the price available when the decision was made and compare it with the actual sequence of fills. This execution audit is separate from asking whether the forecast was directionally correct. It should also distinguish realised account transactions from a hypothetical benchmark, especially when only part of the intended quantity was traded.
Choose and preserve the decision benchmark
Record the intended side, quantity, decision timestamp and a clearly defined reference price before execution. A midpoint benchmark can be useful analytically but is not necessarily a price at which the whole quantity could trade. Preserve the relevant quotes so the distinction remains visible. Add an arrival timestamp and a separate arrival-price reference if you want to investigate delay before the order reached the venue. Investor.gov explains the general importance of execution price and routing for stock orders; the crypto workflow here applies the accounting principle without importing stock-market legal duties into unrelated venues.
Keep the arithmetic tied to actual quantities
For a buy, paying above the chosen benchmark creates a positive cost difference; for a sale, receiving below it does the same. Weight each fill by its executed quantity and include fees in the asset in which they were charged, converting consistently for the report. Compare filled quantity with the intended quantity rather than assuming that an accepted order completed in full. An order acknowledgement, a partial fill and a final fill are separate events. The ledger should be able to reconcile its calculated cost with the account's transaction records.
A hypothetical partial-fill calculation
Suppose the intention is to buy 100 units at a decision reference of 20.00. Only 40 units fill at 20.05 and 30 at 20.08. The 70 filled units cost 1404.40, compared with 1400.00 at the benchmark, so the filled-price shortfall is 4.40 quote units. A hypothetical 0.10% fee adds 1.4044. If a predeclared evaluation price for the unfilled 30 units is 20.30, their opportunity-cost component is 9.00. The combined benchmark shortfall is 14.8044, but only the fills and charged fee are actual transactions in this invented example.
Attribute delay without double-counting it
If the reference price changes between decision and arrival, part of the difference may be associated with waiting, transmission or processing time. The further difference between arrival reference and fills describes another stage. These components must add back to the same total under the chosen definitions; do not add the entire decision-to-fill difference again as a separate slippage charge. Nor does a temporal association prove that a particular software delay caused the price move. Compare repeated observations with aligned clocks and similar conditions before attributing a persistent cost to one part of the process.
Make cancellations and non-execution comparable
A patient limit order may reduce the price paid when it fills while increasing the chance of not trading. Evaluating only completed orders can therefore make it look better by excluding opportunities that moved away. Define in advance when an unfilled remainder is cancelled and the price used for its benchmark comparison. Report filled fraction, waiting time, price shortfall and opportunity-cost assumptions separately. A negative opportunity-cost value is possible when avoiding a purchase proves favourable, but it remains a comparison with the stated counterfactual, not cash credited to the account.
Review execution quality alongside the signal
Group the audit by order size, spread, volatility and data freshness, then compare medians and adverse tails rather than one aggregate average. Keep rejected and partially completed orders in the coverage report. Investigate whether a seemingly strong signal is usually visible only after most of its price advantage has disappeared. A chart-based paper result should remain labelled as a simulation when no fills exist. The useful output is a traceable explanation of where costs arose and which quantities were exposed, not a relabelling of every favourable market move as executed profit.
Sources and example scope
Sources support the definitions and mechanisms. Numerical scenarios are hypothetical teaching examples, not live prices, forecasts or reported HOSTuvo returns. Images are editorial illustrations.