What is slippage? Separate reference prices from fills

Slippage describes the difference between a reference or expected price and the fill price. The reference matters: a quote at submission, a decision price and another benchmark answer different questions.

A higher fill is adverse for a purchase, while a lower fill is adverse for a sale. The sign of fill minus reference alone therefore does not establish whether a difference was favorable. Record trade direction.

In this fictional teaching example, a purchase of 100 shares has a reference price of 100 yen and fills at 101. The adverse difference is (101−100)×100 = 100 yen, or 1% of the reference value. Fees are excluded.

Selling 100 shares at 99 against the same 100 reference also creates an adverse difference of 100 yen. These are not real trades or software results; they illustrate consistent treatment of buy and sell directions.

Commissions are a separate charge. Bid-ask spread, price movement since the decision and the order’s market impact can overlap within a measured price difference, depending on the reference.

Adding the same difference as both spread and slippage can double-count it. Before adding costs, identify the reference, fill and each additional component.

When a signal uses a close and simulated execution uses the next open, their difference includes intervening market movement. The entire difference cannot be identified as slippage experienced by a real order.

Daily bars cannot reconstruct submission times, quotes, available depth or partial fills. Actual execution requires matching order and fill records. Keep simulated fills explicitly labeled as simulated.

KabuForge’s public course explicitly excludes slippage and market impact. Using a later opening price does not by itself model either effect.

Its delay diagnostic keeps quantities fixed and executes at a further observed-session open. It is neither a new signal nor a complete slippage model. Distinguish observations from calendar days and examine changes in fills and skips. Related explanations: transaction costs, look-ahead bias, daily-bar limitations.

Workflow and display source: public GUI research course (reviewed October 6, 2026). Fictional examples illustrate calculations and are not course results.

Public course results are research-only; historical information availability is unverified. Explaining a general definition does not establish full software support. No securities or trades are recommended. Read the disclaimer.

Explore the backtesting learning path

• What is survivorship bias? The problem with today’s stock list
• Historical Paper Replay and Forward Testing: Different Evidence
• What Is Backtesting? What Historical Tests Can and Cannot Tell You