Tape & Order Book Simulator

A continuous double auction with price–time priority, quoted across eight venues. Orders you send are matched against the same book as everyone else's — market, limit, stop and stop-limit. Large orders walk the book and print at several prices.

Symbol
SYNTH
NBB
—
NBO
—
Spread
—
Last
—
Volume
0
VWAP
—

Level 2 — depth of book

VenueBid szBidAskAsk szVenue

Tape — time & sales

TimePriceSizeVenueCond

Your orders & fills

TimeSideTypeQtyPxStatus

No orders yet. Send one from the ticket on the left — try a 5,000-share market order to watch it walk the book.

How this works — and what it gets right

It is a continuous double auction, not a Dutch auction

The distinction matters because it changes what the book is. In a Dutch auction an auctioneer calls a descending price until someone accepts, and participants trade with the house. US equities during regular hours run a continuous double auction: both sides post orders continuously, participants trade with each other, and a trade happens the instant two orders cross.

The intuition behind "Dutch auction" is not wrong, it is just pointed at the wrong session. The opening and closing crosses are single-price call auctions — orders accumulate over a window and everyone executes at one clearing price. That uniform-price property is what a Dutch auction shares. Continuous trading, which is what this page simulates, does not work that way.

Price–time priority

Resting orders are ranked by price first, then arrival time. A better price always goes ahead; at the same price, whoever queued first fills first. This simulator holds a genuine FIFO queue at every price level, so your resting order sits behind the size that was already there — which is why a limit order at the inside does not fill instantly.

Executions print at the resting order's price, never the incoming order's. The passive side sets the price; the aggressive side pays it.

Walking the book

An order larger than the size at the inside quote consumes that level, then the next, then the next — printing at each price it touches. That is why a large market order shows up on the tape as several prints at worsening prices, and why the quote moves after it. Consuming liquidity moves the price; that is the whole mechanism.

NBBO across venues

Every order carries a venue. The National Best Bid and Offer is the highest bid and lowest offer across all of them, and the two sides routinely come from different exchanges. The strip at the top shows which venue is setting each side.

Rules this enforces

RuleWhat it means here
Rule 612 — minimum pricing increment Quotes are in $0.01. The 2024 amendment adding a $0.005 tick for tick-constrained names has a compliance date of November 2027, so a penny is still correct today.
Rule 610(e) — locked and crossed markets The book can never display a bid ≥ an offer. Any order that would lock or cross executes instead. (The SEC proposed rescinding this in June 2026; exchanges would keep their own rules.)
Round lots Sizes are shown in shares. Prints under 100 shares are marked as odd lots.

The tape has no side flag

Real consolidated tape does not tell you whether a print was a buy or a sell. Traders infer it: a print at or above the offer is buyer-initiated, at or below the bid is seller-initiated, in between is uncertain. This tape colours prints on exactly that convention rather than inventing a flag the real feed does not carry.

What this deliberately does not model

  • Hidden and iceberg orders, midpoint pegs, and dark pools — all real, none visible on a Level 2 montage.
  • Order Protection (Rule 611) routing between venues: here the book is consolidated, so a sweep never has to route.
  • Auctions, halts, LULD bands, and the odd-lot quote changes in the 2024 amendments.
  • Fees, rebates, and queue position economics — the reason much of real routing exists.

Background order flow is synthetic: simulated participants post and cancel around the mid, and occasionally take liquidity. The price moves because they trade, not because a price series is being replayed.

Originally published June 2025 at this address; the matching engine was rebuilt in September 2026 — the earlier version filled only against the inside quote and refilled consumed levels at the same price, so taking liquidity never moved the market.
Educational simulation. Synthetic data, no market feed, and no relationship to any real security.