A last-traded price is not a promise that your order can trade there. For a small order in a quiet book, the gap may look trivial. For a larger order, or a fast market, the available quantity can thin out across several prices. To estimate execution cost, start with the quotes, work through the visible size, and keep commissions separate.
Measure the quoted spread
At one moment, the best bid is the highest displayed buying price and the best ask is the lowest displayed selling price. The quoted spread is ask minus bid. The midpoint is their average, a useful reference that is not itself an executable order.
Suppose a fictional BTC spot book shows a best bid of $99.90 and a best ask of $100.10. Its midpoint is $100.00, so the $0.20 spread is 0.20% of the midpoint, or 20 basis points. A buyer crossing the spread pays the ask; a seller crossing it receives the bid. This estimate describes only the top quote at that instant. It says nothing yet about the quantity available there.
Walk the book for your order size
Now suppose the visible book contains these fictional levels:
| Side | Price | Quantity available |
|---|---|---|
| Bid | $99.90 | 0.50 BTC |
| Bid | $99.80 | 0.50 BTC |
| — | Midpoint | $100.00 |
| Ask | $100.10 | 0.50 BTC |
| Ask | $100.20 | 0.50 BTC |
A market buy for 1 BTC would take 0.50 BTC at $100.10 and the next 0.50 BTC at $100.20, if the book stayed unchanged and all displayed size remained available. The notional is 0.50 × $100.10 + 0.50 × $100.20 = $100.15, so the average fill is $100.15 per BTC. A 1 BTC market sell against the shown bids would receive 0.50 × $99.90 + 0.50 × $99.80 = $99.85 in total, also before fees.
The difference between buying and selling 1 BTC across this unchanged snapshot is $0.30. Of that, $0.20 is the quoted spread; walking the second level adds another $0.10 across the round trip. Relative to the $100 midpoint, that is 30 basis points of price friction before commissions or any market move. It is an illustration of the arithmetic, not a quote, fill promise or expected cost for a real account.
Define slippage before you report it
For this example, define slippage as the difference between a chosen reference price and the actual execution price. The reference might be the arrival midpoint, best quote or last trade; state which one you use. A market order that consumes multiple levels can have an average fill different from the top quote, as the displayed fictional book illustrates. This definition is an accounting convention for the example, not a promise about how a venue calculates or displays slippage.
For your own analysis, state the reference: arrival midpoint, best quote, last trade or another price. A buy filled above its chosen reference and a sell filled below it have adverse slippage under that convention. The spread and the movement through deeper levels can both contribute to that gap. If you calculate fills at ask and bid already, do not add the same spread again as a separate cost.
Fees are another line. If, purely for illustration, you assume a 5-basis-point taker fee on each side of the 1 BTC example, entry and exit commissions add about $0.10. Price friction plus those assumed fees is about $0.40 per BTC before funding, latency, partial fills or a price change between orders. Replace the chosen fee rate with the current schedule and your account’s actual tier; see the venue fee example. Funding matters for a perpetual only when the holding interval crosses a settlement under that contract’s rules.
Liquidity is more than one number
A tight quoted spread is useful, but it does not tell you how much size is available. Depth describes displayed quantity at multiple price levels. A market is easier to trade in a given size when it has competitive quotes and enough near-price depth, but a single snapshot cannot prove that the depth will remain until your order arrives. Activity or 24-hour volume is not a substitute for book depth at the decision time.
A market order may execute across several prices or fill only part of its size. Coinbase’s order guide describes both behaviors for its platform. A limit order can cap the worst acceptable price, but it can also remain unfilled or partially filled; a candle touching the limit does not show that your order had queue priority. Compare the order-type guide before treating different instructions as equivalent.
Match the evidence to the claim
A useful execution-cost record names the venue, instrument and contract type; timestamp and reference price; bid and ask; quantities consumed at each level; order side and size; average fill; fee schedule; and any missing or stale data. If you are evaluating repeated orders, preserve snapshots or appropriately synchronized book updates around each decision. Check results for several sizes and faster-market periods instead of presenting one average as a universal cost.
OHLC candles do not preserve the order book or queue. An hourly backtest cannot reconstruct available depth at the signal time or prove a limit fill. A fixed slippage input can show how a modeled rule responds to higher assumed costs, but it is not a historical impact curve. SuperQuant’s included examples use synthetic hourly data and simplified cost assumptions; they do not replay historical books or verify live execution. For shorter-horizon ideas, the scalping cost guide explains when the data may be too coarse for the claim.
The practical sequence is simple: choose one reference, calculate the quoted spread, sweep the required size through the visible levels, add fees once, and report what the snapshot leaves unknown. This is an execution estimate, not a market quote or investment advice.