Position Sizing on DEXs: How Slippage Scales with Size

Why Position Size and Slippage Are Not Linearly Related

One of the most common misconceptions in DEX trading is that slippage scales linearly with position size. Traders assume that if a $1,000 trade costs 2 bps in slippage, a $10,000 trade will cost 20 bps, and a $100,000 trade will cost 200 bps. This assumption is wrong — and the direction of the error is almost always in your favor at small sizes and against you at large sizes. Understanding the true non-linear relationship between position size and slippage is one of the most practically valuable things you can learn as a DEX trader.

The non-linearity arises from the structure of order books and AMM liquidity pools. Both systems have a finite amount of liquidity concentrated at any given price level. As your order size grows, it consumes progressively deeper and thinner layers of liquidity, and each additional unit of size is filled at a worse price than the last. This accelerating cost curve has profound implications for how you should size positions, structure entries, and think about optimal trade splitting.

Real Data: How Costs Scale from $1K to $1M

Based on LiquidView's continuous measurement data across DEX perpetual platforms in Q1 2026, the following cost curves illustrate the non-linear scaling for a BTC-USD perpetual position on Hyperliquid, the most liquid DEX perpetual as of this writing.

• $1,000 order: Total execution cost approximately 3.0–3.5 bps. At this size, taker fee dominates entirely and price impact is effectively zero. The marginal cost of each additional dollar in the order is flat — you are in the linear zone.
• $10,000 order: Total execution cost approximately 3.5–4.0 bps. Still dominated by the taker fee. Spread contributes roughly 0.5 bps. Price impact remains negligible. Cost curve is still nearly flat.
• $100,000 order: Total execution cost approximately 4.5–6.0 bps. Price impact begins to emerge meaningfully. You are now consuming multiple layers of the order book, and each layer is slightly thinner than the last. The cost is no longer flat — it curves upward.
• $500,000 order: Total execution cost approximately 8–12 bps. Price impact is now the dominant cost component. You are moving deep into the order book, pushing price significantly away from the mid. This is where the non-linearity becomes starkly visible: you are paying 2–4x the per-unit cost of a $10,000 order.
• $1,000,000 order: Total execution cost approximately 15–25 bps depending on market conditions. At this size, your own order is the market. You are consuming enough liquidity to move prices measurably, and the marginal cost of each additional dollar in the order is accelerating. A $1M BTC buy on Hyperliquid can shift the mid-price by several basis points on its own.

These figures are for BTC-USD on the most liquid DEX perpetual available. For ETH-USD, expect similar curves but with price impact appearing at slightly lower size thresholds. For altcoin perpetuals — SOL, AVAX, ARB, and especially smaller tokens — the curve inflects much earlier. A $50,000 SOL perpetual position on some platforms may already exhibit significant price impact.

Why the Relationship Is Non-Linear: Order Book Depth Explained

To understand why slippage is non-linear, you need to understand how orders are filled in a limit order book. When you submit a market buy order, the exchange matches it against existing sell orders starting from the lowest asking price and working upward. The order book typically looks like a pyramid: there is significant quantity sitting at prices very close to the current mid, and progressively less quantity sitting at prices further away.

A small order fills entirely within the first few price levels — the most liquid part of the book — and your average fill price is very close to the mid. A large order exhausts those first levels and then fills the next layers at progressively worse prices. As you reach deeper into the book, you are dealing with liquidity provided by market makers with wider pricing to compensate for their inventory risk, and at the deepest levels, by opportunistic sellers who only sell at significant premiums.

On AMM-based DEXs (which underlie some perpetual platforms indirectly), the same principle applies through the mathematical pricing curve. Each marginal unit of a purchase pushes the pool's price ratio slightly higher. The first dollar spent moves the price barely; the millionth dollar spent moves it far more than the average of all the dollars before it. This is the constant-product formula (x * y = k) in action, and it produces inherently convex price impact.

Finding Your Optimal Position Size on Each Exchange

The optimal position size is not simply "the biggest you can afford" or "the smallest that still makes sense." It is the size at which your expected P&L, after accounting for all execution costs, is maximized. Since execution cost accelerates non-linearly with size, there is a sweet spot where you are capturing enough notional to make the trade worth doing but not so much that execution cost is consuming an unreasonable share of your expected edge.

A practical framework for finding your optimal size on a given exchange and token: first, query LiquidView's size comparison tool to see the execution cost curve for your specific pair. Second, identify the size threshold where cost starts accelerating — this is typically where you transition from fee-dominated to price-impact-dominated execution. Third, set your maximum position size at roughly 70–80% of that inflection point to maintain a buffer against intraday liquidity variation.

• BTC-USD on Hyperliquid: inflection point typically around $200K–$400K. Optimal single-order size for cost-efficient execution: $150K–$250K.
• ETH-USD on Hyperliquid: inflection point typically around $100K–$200K. Optimal single-order size: $75K–$150K.
• SOL-USD on Hyperliquid: inflection point around $50K–$100K. Optimal single-order size: $40K–$75K.
• BTC-USD on Paradex: similar depth to Hyperliquid for BTC, with comparable inflection thresholds. ETH and altcoins have earlier inflection due to lower market depth.
• gTrade (synthetic): execution model uses oracle pricing, so price impact from order book consumption does not apply in the same way. Price impact is instead governed by the platform's spread variable, which increases with open interest in a given direction. Effective inflection for most pairs occurs around $500K–$1M.

Splitting Large Orders to Minimize Slippage

When your required position size exceeds the optimal single-order threshold, order splitting is the primary tool for reducing total execution cost. By breaking a large order into multiple smaller pieces executed over time, you allow the order book to replenish between fills, effectively resetting the cost curve and achieving a better average fill price.

The tradeoff is time. Splitting an order over 10 minutes versus executing it instantly means your entry price exposure is spread across time. If the market moves in your intended direction before all tranches are filled, you end up paying more for the later tranches. The optimal splitting strategy balances market impact reduction against timing risk based on your assessment of market momentum.

• Time-weighted average price (TWAP): Splits the order into equal chunks executed at regular intervals over a defined window (e.g., ten $50K orders over 50 minutes). TWAP is simple and predictable but does not adapt to price movement.
• Volume-weighted splitting: Executes larger tranches when liquidity is deeper (typically during peak trading hours, 14:00–22:00 UTC) and smaller tranches when the book is thinner. More sophisticated but better for very large orders.
• Opportunistic splitting: Places limit orders slightly inside the spread and waits for liquidity to come to you. Each limit fill is at maker rates (sometimes with rebates) rather than taker rates, reducing cost per tranche significantly. Requires patience and acceptance of partial fills.
• Cross-exchange splitting: Splits the order across multiple exchanges simultaneously to access aggregate liquidity. A $1M BTC position split $400K on Hyperliquid, $350K on Paradex, and $250K on Lighter will typically achieve a better aggregate fill than $1M on any single platform.

Using the LiquidView Size Comparison Tool

The LiquidView size comparison tool is purpose-built to solve the non-linear slippage problem. Rather than requiring you to manually model the cost curve for each exchange and pair, it continuously queries order book depth across all supported platforms and computes the estimated execution cost at any size you specify — from $1,000 to $1,000,000.

To use it: select your token pair (BTC-USD, ETH-USD, SOL-USD, or any supported perpetual), enter your intended order size, and select your direction (long or short). The tool returns the estimated all-in execution cost in basis points for each exchange, broken down into fee component, spread component, and price impact component. You can then see the total for each size tier side by side — making the non-linear relationship immediately visible and actionable.

For traders managing larger positions, the tool also provides the estimated execution cost for a TWAP split (5 tranches, 10 tranches, or 20 tranches) versus a single block — letting you immediately quantify the dollar savings from splitting before you commit to either approach.