The Harbor Master's Ledger

Galley / Chronometers / The Harbor Master's Ledger  Advanced

Prerequisites: Knot Tying · Episode 09 — The Rope Walk · 18 min read

I bought $890,000 worth of a mid-cap semiconductor stock. My account was $450,000. It used 2× margin. I didn't even know I had margin enabled. I was in a meeting when the order filled. I got the confirmation email and almost threw up... I lost $47,000. In 90 seconds. Because of a decimal point.

THE MEASUREMENT

Every trade has two prices: the one you see on the screen, and the one you actually get. The gap between them is execution cost — slippage, market impact, and latency. For a discretionary trader making ten trades a month, this gap is a rounding error. For a systematic trader making fifty thousand trades a year, it is the difference between a yacht and a lifeboat.

The Harbor Master's Ledger is the mathematics of that gap. We derive the square-root law of market impact, estimate slippage from the bid-ask spread, and calculate the annual friction that devours automated strategies. If you trade with size, frequency, or leverage, this is the math that determines whether you keep your edge or donate it to the market.

THE DERIVATION

Step 1: The Order Book

Before a trade reaches the tape, it passes through the limit order book (LOB). The LOB is a queue: buyers on the left (bids), sellers on the right (asks). The best bid and best ask define the spread. When you submit a market order, you cross the spread and pay the ask (if buying) or receive the bid (if selling). That spread is your first cost.

Step 2: The Square-Root Law

Slippage is only the beginning. When your order is large relative to the market's daily volume, your own buying (or selling) moves the price against you. This is market impact. Empirical research — from Almgren et al. (2005) to modern execution desks — shows that impact follows a square-root relationship with order size.

Convert impact to dollars:

Step 3: Total Execution Cost

The total cost of a single trade is the sum of slippage and market impact. We ignore commissions (broker-dependent) and latency (negligible for swing traders, critical for HFT).

Step 4: Annual Friction

For systematic traders, the true horror is compounding. A strategy that makes 50,000 trades per year with an average cost of 0.20% per trade loses 10,000 basis points — 100% of its expected alpha — to friction alone.

WORKED EXAMPLE

On March 14, 2018, Yuki's algorithm bought $890,000 of a mid-cap semiconductor stock. Let us calculate what the Harbor Master's Ledger would have predicted for that single trade — and what 50,000 similar trades would cost annually.

Step 1 — Shares: $890,000 / $50.00 = 17,800 shares

Step 2 — Slippage: ($0.05 / 2) × 17,800 = $445

Step 3 — Market Impact: 0.02 × √($890,000 / $100,000,000) = 0.02 × 0.0943 = 0.1886%. In dollars: $890,000 × 0.001886 = $1,678

Step 4 — Total Cost: $445 + $1,678 = $2,123 (0.24% of the order)

Step 5 — Annual Friction: Assuming an average trade size of $5,000 and the same 0.24% cost: 50,000 × $5,000 × 0.0024 = $600,000 per year in execution costs alone.

Yuki's $47,000 loss came from a position-sizing bug. But the Harbor Master's Ledger reveals a slower, quieter thief: even if every trade had been perfectly sized, his automation would have paid roughly half a million dollars annually just to cross the spread and move the market.

THE INTERACTIVE CHRONOMETER

Enter your trade parameters below. The Ledger calculates slippage, market impact, total cost, and annual friction. Test your own strategy. The numbers may surprise you.

INTERPRETATION

Execution cost is not a moral failing. It is physics. When your order is large, fast, or frequent, you leave footprints. The Ledger tells you how deep they are.

Healthy Cost < 0.10% — Your edge is safe. Typical for large-cap equities with small size.

Caution Cost 0.10% – 0.50% — Your edge is eroding. Consider reducing size, widening time horizon, or using limit orders.

Danger Cost > 0.50% — Your strategy is paying the market for the privilege of trading. Reduce frequency, shrink position, or find more liquid instruments.

The latency caveat: For traders holding positions for days or weeks, latency measured in milliseconds is irrelevant. For Yuki's 50,000 trades per year — many intraday — a 100ms delay can mean missing the quote, receiving worse fills, or being front-run by faster algos. The square-root law above captures size impact. Latency impact requires co-location, direct market access, and hardware optimization — topics for the next voyage.

Related Guides