Effective spread (and realized spread)
The effective spread measures what a trade actually paid relative to the mid-price at the moment it executed: twice the distance between the trade price and that mid. The realized spread measures the same distance against the mid a little later, which is what the liquidity provider kept once the price had moved; the difference between the two is the price impact of the trade.
Senzoukria · Glossary · Updated September 2026
Three spreads, three questions
| Measure | Formula (for a buy) | Question it answers |
|---|---|---|
| Quoted spread | Best ask − best bid | What would crossing cost right now? |
| Effective spread | 2 × (trade price − mid at the trade) | What did this trade actually pay? |
| Realized spread | 2 × (trade price − mid some time later) | What did the liquidity provider keep? |
| Price impact | Effective − realized | How much did the mid move after the trade? |
A worked example
The book shows 5,000.00 bid and 5,000.25 offered, so the mid is 5,000.125. A market buy fills at 5,000.25. Its effective spread is 2 × (5,000.25 − 5,000.125) = 0.25, one tick, the same as the quoted spread. One minute later the mid is 5,000.375. The realized spread is 2 × (5,000.25 − 5,000.375) = −0.25: the seller who provided the liquidity is now behind by one tick. The price impact is 0.25 − (−0.25) = 0.50. The buyer was informed, or lucky; the seller was adversely selected.
Why effective and quoted differ
- An order larger than the size at the best price walks the book and pays more than the quoted spread.
- An order can execute inside the spread on venues that allow it, paying less.
- Quotes change between the decision and the fill, so the relevant mid is the one at execution.
- On a one-tick futures market, the effective spread of small market orders is usually exactly one tick; the interesting variation lies in larger orders and in the realized spread.
When quotes are not available
Computing effective spreads needs the quote at each trade. When only prices are available, estimators infer an average effective spread from price behavior: Roll's estimator from the negative autocovariance of price changes, Corwin and Schultz's from the relation between one-bar and two-bar high-low ranges. Both describe an average over a window and both rest on assumptions that intraday data can violate.
In Senzoukria
The Quantitative group of the indicator library includes Roll Implied Spread and Corwin-Schultz Spread, both estimating an effective spread from bar prices without reading the book. They are distinct from Spread Proxy (ticks) in the Tape & flow group, which measures how many ticks price skipped between traded levels inside a bar, a geometry of the prints rather than a spread model. The live quoted spread is visible on the heatmap through the best bid and ask lines and on the DOM ladder.
Related
- Bid-ask spread
- Roll implied spread
- Corwin-Schultz spread estimator
- Adverse selection
- Spread Proxy indicator
- Transaction costs
In the same section
- Effort vs result
- Drawdown lock
- End-of-day drawdown
- Drawdown
- Equity curve
- Double distribution
- Equity curve trading
- DOM software
This page in other languages
Frequently asked questions
- Why multiply by two?
- The distance between a trade and the mid is half of a round trip. Multiplying by two makes the effective spread comparable with the quoted spread, which is the full distance from bid to ask. Some sources report the half-spread instead, so check the convention before comparing figures.
- Can the realized spread be negative?
- Yes. A negative realized spread means the mid moved past the trade price in the direction of the aggressor within the measurement horizon, so the liquidity provider lost money on that trade. Averaged over many trades, a negative realized spread is a sign of toxic, informed flow.