Use measured, instrument-specific and order-type-specific slippage as your baseline in backtests. When you lack fill data, apply a conservative percent or tick cost and add uplifts for opens, news and thin liquidity. The main levers are a fixed percent or tick value, separate rules for market, open and limit orders, and buckets by liquidity and time of day.
TL;DR:
- Measure realized slippage from tick or level one fills for each instrument and order size, then bucket results by liquidity, time, and volatility.
- Large orders can walk the book, so a model that ignores order size may understate market impact even when its baseline is measured.
- Professional entities can achieve transaction costs below 1 basis point, while retail traders may face more than 50, so match assumptions to your execution setup.
- A bar range cap can drop orders, while allowing fills beyond the high or low preserves trade counts but can use prices that never traded.
- Run sensitivity sweeps to find the slippage level that turns profits into losses, and publish each assumption beside performance results.
Table of Contents
- What slippage is and why it changes backtest results
- Common methods to model slippage in backtests
- How to pick realistic slippage values: a data-driven checklist
- Practical implementation notes and examples
- Practitioner perspective on execution trade-offs and slippage drivers
- My checklist for backtests I would trust
- Test fills with the platform built for scalpers
- FAQ
- Sources
What slippage is and why it changes backtest results
Slippage is the gap between the price you expected and the price you actually got. It runs both ways: negative slippage costs you, positive slippage helps you. It shows up most in volatile or low-liquidity markets.
Two drivers matter for scalpers. The first is market impact: your order size against the depth available at the touch. The second is latency and race effects: the quote moves while your order is in flight. Both eat into small per-trade edges.
Ignore slippage and the backtest books every fill at the signal price. Returns inflate, win rates look better than they are, and cases where a live order would have missed or filled badly vanish from the record. A scalping edge of a few ticks can disappear completely once honest execution costs go in.
Common methods to model slippage in backtests
Pick the method that matches the data you have, then layer on complexity as needed.
- Fixed percent or fixed ticks: the simple baseline. Quick to code, but blunt: it treats a calm midday fill like a news-bar fill.
- Per-order-type rules: slip market and stop orders, treat limits differently, and handle the open bar as an exception.
- Capped vs. out-of-bar slippage: capping fills inside the bar's high and low avoids impossible prices but can drop executions. Allowing out-of-bar fills keeps trade counts intact but can print prices that never traded.
- Stochastic draws: sample slippage from your own historical fill distribution or per-bar microstructure data for realistic variability.
- Event-based and size-based uplifts: widen costs around news, opens and volatility spikes, and scale them with order size to reflect market impact.
Pro Tip: Start with a fixed baseline, then replace it bucket by bucket as your measured fills come in.
How to pick realistic slippage values: a data-driven checklist
Defensible numbers come from measurement, not guesses. Work through this:
- Measure realized slippage from tick or level-1 fills for your instrument and your typical order size.
- Bucket the results by liquidity (average daily volume or depth), time of day and volatility regime, then set parameters per bucket.
- Adjust for who is executing. Professional desks and retail traders live in different cost worlds, so match the parameter to your real execution setup.
- Run sensitivity sweeps and note the slippage level where performance flips from profit to loss.
- Print your assumptions right next to the performance tables.
On step 3, the gap is wide: professional entities can achieve transaction costs under 1 basis point while retail traders may face more than 50 basis points. One optimistic number for every trader type is not a model.
For historical microstructure samples across venues, Assymetrix's slippage guide covers measuring slippage and backtesting with high-resolution data.
Practical implementation notes and examples
Backtrader's slippage documentation is a clean reference for how engines handle this on OHLC bars. Its main parameters:
- slip_perc: slippage as a percentage of price.
- slip_fixed: slippage as a fixed number of price points.
- slip_open: whether to apply slippage to orders that execute at the open price.
- slip_match: cap the slipped price to the bar's high and low so you never fill outside the bar's range.
- slip_out: allow slippage to push the fill outside the high/low range.
The trade-off is plain. With matching on, impossible fills disappear, but some orders find no valid price and get dropped, which changes your trade count. With slip_out on, fills may occur at prices outside the bar's range.
Example defaults, as pseudo-config: for scalping, use small fixed ticks with slip_open and slip_match both on. For swing trading, use a higher percent value, bucketed by instrument liquidity.
Watch for two pitfalls: applying the same slippage to every instrument, and ignoring the extra cost large orders create by walking the book.
Practitioner perspective on execution trade-offs and slippage drivers
Execution algorithms live inside a trilemma: market impact, market risk and opportunity cost. BIS research on FX execution algorithms describes the implementation shortfall benchmark and shows that algorithm choice and benchmarking method change measured execution costs. Your backtest inherits the same choice.
Speed matters too. SEC staff research on automated liquidity provision indicates that speed-related effects can shape spreads and adverse selection costs in some instruments. For fast strategies, model race and latency costs separately from baseline market impact when you can.
My checklist for backtests I would trust
Most backtests fail on honesty, not math. A single flattering slippage number applied to every trade is the usual culprit. My view: a result that survives a harsh execution assumption is worth more than a pretty equity curve that needs perfect fills. Scalpers feel this first, because their edge per trade is thin and every tick of friction compounds across hundreds of entries.
- Document where each slippage figure came from and how you derived it.
- Put a sensitivity table in every report.
- Prefer measured, instrument-specific values over one global number.
- Log and review non-matches and dropped fills. They usually expose unrealistic assumptions.
Pro Tip: If dropping a few fills flips your results, your edge depends on prices you could not have had.
— Tran
Test fills with the platform built for scalpers
We built our Command Center to support your backtesting workflow. Our backtesting dashboard and indicator suite help you test trading strategies. Our slippage reduction writeup pairs well with the measurement steps above.
[image:cta_image]
- Backtest as part of your workflow.
- Test adaptive take-profit and stop-loss placement.
- Pair it with Smart Scalping Signals on lower timeframes.
See our plans and start testing today.
FAQ
What does slippage mean in crypto trading?
It is the difference between the price you expected and the price your order actually filled at. It tends to be larger when liquidity is thin or the market moves fast. Investopedia's slippage definition covers the basics.
How do beginners backtest with slippage?
Start with a simple fixed percent or tick cost on every trade, then run it at several levels to see where results break. Our retail trader's backtesting guide walks through the workflow.
Is slippage the same as the spread?
No. The spread is the gap between the best bid and ask. Slippage is the gap between your expected price and your actual fill, which can include the spread plus movement or depth effects.
How much slippage should I assume in a backtest?
Use measured fills for your instrument, order size and time of day wherever possible. Without data, pick a conservative value and sweep it higher to find your strategy's breakpoint.
Sources
- Broker - Slippage - Backtrader
- Understanding Slippage in Finance: Key Insights and Examples
- FX execution algorithms and market functioning