TycoonGlobal

Strategy

Reading the Order Book

Depth beats top-of-book, VWAP is the number that matters, and the self-churn trap that cost simulated agents $48 million in tax.

Strategy6 min readPublished 26 Oct 2026#strategy#markets

The spot exchange is a price-time priority limit order book with one-cent ticks. You can trade it through Smart Auto-Fill without ever seeing it, and most players will. If you do look, here is what repays attention.

Depth, not the top

A $58.00 ask on microchips tells you almost nothing. The questions are how many at that price and what does the next level cost.

Your real fill price on a market order is the volume-weighted sweep through however many levels your quantity consumes. On a thin Tier-3 book, a 300-unit order can clear four or five levels and land well above the posted best ask.

Smart Auto-Fill quotes you the sweep — "Buy 500 Copper Wire — $8,240 Instant" is the all-in number, not 500 times the top ask. That is why the button is genuinely useful rather than a simplification: it is answering the right question.

VWAP is the reference number

The trailing 24-hour volume-weighted average price does three jobs:

  • It anchors the B2B contract collar at ±20%.
  • It values your inventory for credit purposes, at a 70% haircut.
  • It is the fairest single answer to "what is this worth".

When you are deciding whether a contract price is good, compare it to VWAP, not to the current best bid. The bid is a snapshot of this moment's impatience.

Post rather than take

The 6% exchange tax is charged on seller proceeds whether you posted a resting limit order or crossed the spread with a market order. The spread, however, is only paid by the taker.

On a staple with a two-cent spread that is trivial. On a Tier-3 good where the spread can be several percent, consistently posting instead of taking is often worth more than the tax itself over a week of production.

The trade-off is fill risk. A resting order may not fill before your warehouse fills up or your perishables expire. Post when you have buffer; take when you have a deadline.

The self-churn trap

This is the single most expensive habit available to an automated or semi-automated player, and the simulation demonstrated it at scale.

Early runs showed traded volume exceeding produced volume for POWER and MICROCHIP. The cause: agents bought inputs up to a 3x hourly-need buffer, and sold anything above a 2x buffer. So they bought, crossed into the 2x—3x band, and immediately sold the same stock back — paying 6% each way, forever.

Aligning the two thresholds cut the spot-tax sink from $58.6M to $10.7M over the run.

The same thing will happen to a human player whose Auto-Fill threshold sits above the level at which the "sell surplus" banner fires. If you tune your own buffers, derive the buy trigger and the sell trigger from one number. A buy threshold above your sell threshold is a slow, invisible leak.

Liquidity is a design decision

Phase 1 runs as a single global realm with all 30 cities, and realm sharding does not begin until roughly 200,000 active companies. That is a liquidity choice: splitting the world halves every order book and halves the sample behind every VWAP, which makes the collar noisy and the market worse for everyone.

The practical consequence is that Tier-0 and Tier-1 staples trade thickly and Tier-3 goods trade thinly even in a healthy world. Expect wide spreads on laptops and tight spreads on wheat, and size your orders accordingly.

What never trades here

City retail runs on a clamped elasticity curve against a simulated population, not an order book.

Utilities rarely clear above the NPC ceiling of 2.0x reference, because the Central Utility Authority sells water, power and fuel without limit at that price. The book exists but the ceiling is effectively the price.

Recurring supply belongs in contracts at 1.5%. If you find yourself buying the same quantity of the same SKU every day on spot, you are paying four times what you need to.