Slippage and Execution Speed
Slippage and Execution Speed: How Latency Costs You Money
Slippage is the difference between the price you expected to receive and the price you actually received when your order executes. It's one of the most underestimated costs in active trading because it's invisible—there's no commission line item or fee notification, just a worse fill than you anticipated. A 1-cent slippage on a 1,000-share trade is $10. On 20 trades per day, 250 trading days per year, that's $50,000 per year in hidden losses. Execution speed is the primary determinant of slippage for day traders. The faster your orders reach the exchange and the faster they execute, the less time the market has to move against you.
Quick definition: Slippage is the difference between your expected fill price and your actual fill price; execution speed (latency) is the time it takes for your order to reach the exchange and be filled.
Key takeaways
- Slippage has two sources: the bid-ask spread and market movement — The spread is unavoidable; market movement while your order is in transit is partially avoidable with faster execution.
- Execution speed matters only for day traders and scalpers — If you hold positions for hours or days, a few milliseconds of latency doesn't matter. If you hold for 30 seconds, it's everything.
- Fast execution requires three things: low-latency platform, low-latency connection, and DMA — You can't control the exchange, but you can control your setup.
- Spread costs scale with position size — A 1-cent spread on 100 shares is $1; on 10,000 shares, it's $100. This is why larger traders focus on the most liquid stocks.
- You can estimate slippage in advance — Calculate it as (bid-ask spread) + (estimated market move during latency). Measure actual slippage and compare it to your estimates to improve your execution.
The two sources of slippage
Slippage component 1: The bid-ask spread. When you buy at market, you pay the ask. When you sell at market, you receive the bid. The difference is the spread, and it's your cost of immediate execution. A stock with a 1-cent spread in 1,000-share blocks costs you $10 per round-trip trade. A stock with a 50-cent spread costs you $500 per round-trip. This spread is unavoidable—it's the price of liquidity.
Slippage component 2: Market movement during latency. Your order leaves your computer, travels across the internet to your broker, your broker routes it to the exchange, the exchange processes it, and the result comes back to you. During this time—typically 50–300 milliseconds for retail traders, 5–50ms for DMA traders—the market is moving. If you're buying and the price is rising, your market order will fill at a higher price than when you hit submit. If you're selling and the price is falling, you'll receive a lower price.
Slippage equation (simplified)
Total Slippage = Bid-Ask Spread + (Market Move × Likelihood of Adverse Move)
Example: A stock has a 2-cent spread. Your latency is 100ms. During 100ms, the stock typically moves 1–2 cents (in volatile stocks, more; in quiet stocks, less). If you're buying, you might hit the ask at $50.11, but by the time your order processes, the price is $50.13. You paid 2 cents more than you expected (beyond the spread). Total slippage: 2 cents (spread) + 2 cents (adverse move) = 4 cents, or $40 per 1,000 shares.
Measuring latency: hardware, network, and software
Hardware latency: The time your computer takes to submit the order. A fast CPU and SSD reduce this to milliseconds. A slow or overloaded computer adds tens of milliseconds.
Network latency: The time for your order to travel from your computer to your broker and then to the exchange. Fiber internet reduces this; DSL increases it. Distance to the broker's servers matters too. A trader in New York with a broker in New York has lower latency than a trader in Los Angeles with the same broker.
Platform latency: The time your trading platform takes to receive your click/keystroke and send the order to the broker. Some platforms are optimized for speed; others prioritize features and are slower. A professional trading platform like Thinkorswim or Interactive Brokers' TWS can be optimized for low latency. A web-based platform or mobile app has much higher latency.
Broker routing latency: The time the broker takes to validate your order and route it to the exchange. This varies widely. A broker with DMA has lower routing latency because orders go directly to the exchange. A broker routing through market makers adds additional latency.
Latency checklist
Typical latencies for different setups:
Retail broker + DSL + mobile app: 300–800ms
Retail broker + cable + web platform: 100–300ms
Retail broker + fiber + desktop app: 50–150ms
DMA broker + fiber + optimized platform: 5–50ms
HFT firm + co-located servers: <1ms
Why spreads blow up on fast moves
During normal market conditions, a stock might have a 1–2 cent spread. When the market moves fast—earnings announcement, Fed announcement, market-wide sell-off—spreads blow up. The same stock that trades on a 1-cent spread might jump to a 10–50 cent spread in seconds because market makers pull their bids and asks (to avoid getting hit with bad inventory) and bid-ask size shrinks.
When spreads blow up, your slippage on market orders skyrockets. You intended to exit a losing position for a 2% loss; instead, the 50-cent spread, combined with adverse movement, turns it into a 5% loss.
This is why professionals use limit orders during volatile times. You sacrifice execution certainty to protect your price. Your limit order might not fill, but if it does, you get your price.
Calculating expected slippage
Before placing a trade, you can estimate slippage:
-
Check the bid-ask spread. Look at level 2. If there's a 3-cent spread, that's 3 cents of guaranteed slippage.
-
Estimate market move during latency. Your latency is likely 50–200ms. In 200ms, a typical stock in moderate volatility moves 1–3 cents. Volatile stocks move more; quiet stocks, less. Add 1–5 cents to your estimate depending on volatility.
-
Multiply by position size. If your slippage estimate is 4 cents and you're trading 1,000 shares, that's $40 of slippage per trade.
Example: You want to buy 5,000 shares of a stock at $50.00. The bid-ask is $50.10–$50.12 (2-cent spread). Your latency is 100ms. You estimate a 1-cent adverse move during latency. Total expected slippage: 3 cents = $150. If your expected profit on this 5,000-share trade is $500, your net expected profit is $350 after slippage. If your expected profit is only $75, slippage eats 50% of your edge—probably not worth the risk.
Real-world examples
Example 1: The cost of poor execution. A trader uses a retail broker with a web platform and DSL internet. Their total latency is 250ms. They day trade 1,000 shares, 15 trades per day, 250 trading days per year. Average slippage per trade (including spread and adverse movement): 5 cents = $50. Total slippage per year: 15 × 250 × $50 = $187,500. They switch to a DMA broker with fiber internet and optimized platform. New latency: 50ms. New average slippage: 2 cents = $20 per trade. Total slippage per year: $75,000. They saved $112,500 per year by upgrading their execution infrastructure—and that's the slippage they eliminated, not the profit they gained. This is why pros invest in fast setups.
Example 2: Slippage during a flash crash. A stock is trading at $50.00 with a 1-cent spread. An algorithmic sell order causes a flash crash, and the stock plummets to $40.00 in 2 seconds. A retail trader with 200ms latency tries to exit their long position. By the time their market sell order reaches the exchange, the price is already $45.00, not $50.00. Their slippage: $5.00 per share, or $5,000 on a 1,000-share position. A DMA trader with 30ms latency hits the exit button as soon as they see the move starting. Their order reaches the exchange at $49.50. Their slippage: $0.50 per share, or $500 on the same position. The faster trader loses $4,500 less.
Example 3: Tight edge scenario. A trader identifies a stock that should trade up 2 cents in the next 30 seconds based on tape reading. They want to buy 2,000 shares and scalp the 2-cent move for a $40 profit. The spread is 2 cents. Using a standard broker, their latency is 150ms, creating an estimated 2-cent adverse move. Total cost: 4 cents = $80. This exceeds their $40 expected profit by 2×. The trade is unprofitable. Using DMA, their latency is 30ms, creating only a 1-cent adverse move. Total cost: 3 cents = $60. Still slightly unprofitable. Using a co-located HFT system, latency is <1ms, adverse move is <0.25 cents. Total cost: 2.25 cents = $45. Still unprofitable. The edge is too small. This trader needs to find larger moves or improve their prediction.
Minimizing slippage: actionable steps
Step 1: Choose the right stock. Trade highly liquid stocks with tight spreads. The difference between scalping Apple (spread: <1 cent) and scalping a microcap (spread: 50+ cents) is $500+ per trade in slippage alone.
Step 2: Get the fastest internet. Upgrade from DSL to cable or fiber. This cuts network latency by 50–75%.
Step 3: Use a professional trading platform. Web-based platforms and mobile apps add 100–500ms of latency. Professional platforms like Thinkorswim, TWS, or Ninjatrader reduce it to 10–100ms.
Step 4: Use DMA if your trading style needs it. If you're a position trader or swing trader, DMA doesn't help. If you're a day trader, DMA is critical. If you're a scalper or high-frequency trader, DMA is non-negotiable.
Step 5: Use limit orders when possible. On high-conviction trades or when you have time, use limit orders instead of market orders. You sacrifice execution certainty but eliminate slippage from market movement (though you still pay the spread in the form of a wider limit that's needed to get filled).
Step 6: Scale into positions. Instead of buying 5,000 shares at market in one order, buy 1,000 shares four times. This reduces the market impact of your order and, if price trends in your favor, you get better average fills.
Decision tree
Common mistakes
Mistake 1: Ignoring slippage in profit calculations. A trader plans to scalp a 3-cent move on 1,000 shares = $30 profit. They forget the spread (2 cents = $20) and adverse move (1 cent = $10). Actual profit: $0. They lose money and don't know why.
Mistake 2: Trading illiquid stocks and paying huge spreads. A trader wants to scalp microcaps because they move fast. A microcap has a 50-cent spread. One round-trip trade costs them $500 in spread alone. They need to move the stock 5+ cents to break even. This is a house-edge game.
Mistake 3: Not measuring actual slippage. Traders estimate slippage but never measure it. They make 100 trades and never look back at actual execution prices. Measuring actual slippage helps you identify whether your setup is slow or whether you're just unlucky.
Mistake 4: Thinking DMA solves all execution problems. DMA reduces latency but doesn't eliminate slippage. You still pay the bid-ask spread. You still risk adverse movement. DMA is one tool, not a silver bullet.
Mistake 5: Using market orders on wide spreads. During low-volume or volatile periods, spreads widen. A trader uses market orders anyway, accepting 10-20 cent slippage. Use limit orders during these times. You might not fill, but at least you don't overpay.
FAQ
How do I measure my actual latency?
Some trading platforms show latency stats. You can also measure it by placing test orders during market hours and comparing your expected price (from level 2 at the moment you hit submit) to your actual fill price. The difference is roughly your latency × typical volatility.
Does latency matter for swing traders?
No, not significantly. If you hold a position for hours, a 100ms lag is irrelevant—the stock might move 10 cents in that period, dwarfing latency effects. Latency only matters if you're holding positions for seconds to minutes.
Can I reduce slippage by using limit orders?
Yes, but with a trade-off. You don't pay spread + adverse movement, but you might not fill. Use limit orders when you're willing to wait (or not fill) in exchange for price certainty.
Should I avoid trading before major economic data?
If you're a scalper, yes. Before Fed announcements or jobs data, spreads can blow up 5–10×, and volatility spikes. If you're a swing trader, data releases don't matter much. Your latency isn't the limiting factor.
Do I need to worry about slippage on options or futures?
Yes, the same principles apply. Options have wider spreads, so slippage is often larger. Futures have better liquidity in major contracts, so slippage is lower.
Is there a slippage cost when I use a limit order that doesn't fill?
No, because the order doesn't execute. The only cost is opportunity cost (you didn't get the fill you wanted, so you missed the move).
Related concepts
- What Is Direct Market Access? — DMA reduces execution latency, which is the primary factor in minimizing slippage.
- Level 2: Seeing the Order Book — Check the bid-ask spread on level 2 before trading to estimate slippage.
- Order Types for Active Trading — Choose between limit (no slippage from execution, but no execution guarantee) and market orders (guarantee fills, but pay slippage).
- Trading Edges Explained — Slippage is the cost that reduces your edge; understanding slippage helps you calculate whether an edge is real.
Summary
Slippage—the difference between your expected fill price and actual fill price—is the invisible cost that eats active traders' profits. It comes from two sources: the bid-ask spread (unavoidable) and market movement during latency (partially avoidable with faster execution). Execution speed (latency) is determined by your internet connection, trading platform, broker routing, and whether you use DMA. A trader with 250ms latency on 15 daily trades can lose $187,500 per year to slippage compared to a trader with 50ms latency. Minimizing slippage requires trading liquid stocks, upgrading your internet and platform, using DMA if appropriate, and measuring your actual slippage so you can identify areas for improvement. For day traders and scalpers, slippage is often the difference between profitability and consistent losses.