On This Page
Key Takeaways
- Quote stuffing targets the processing capacity of exchanges or market participants.
- A queue builds when incoming messages exceed the consumer’s processing capacity.
- The example receives 5,040 quote messages and processes 2,000, leaving 3,040 queued.
- At 40 new messages per second, the system clears that queue in 1.55 seconds, and its quotes lag the market until it does.
- Quote events describe book updates; trade prints record completed executions.
What is quote stuffing?
Quote stuffing is the use of rapid bursts of order or quote messages to burden exchange or participant systems. The mechanism is processing load: when arrivals exceed a consumer’s capacity, messages accumulate in a queue and wait to be processed. Regulators and exchanges treat intentional overload of quotation systems as a disruptive practice.
Quote feeds carry the book events each data product defines, such as new orders, changes and cancellations. Trade feeds record completed executions. A stuffing burst floods the quote feed while the trade feed barely moves, and the quote-to-trade ratio catches that gap.
Formula
End-of-burst backlog Q = max(0, M − C × T). Catch-up time = Q / (C − λ), for constant λ < C.- M is the quote-message count during the burst.
- C is processing capacity in quote messages per second.
- T is burst duration in seconds.
- The backlog formula assumes the queue starts empty and the consumer stays busy for the whole burst; add any messages already waiting to M.
- λ is the quote-message arrival rate after the burst.
- Catch-up time is how long the consumer runs behind the market before the queue clears.
- Quote-to-trade ratio = quote messages / trades in the same interval; the burst lifts it from 10 to 1,260.
Quote stuffing example: a 5,040-message burst
At baseline, a quote-processing stream receives 40 messages per second while the trade tape records four trades per second. For one second, quote arrivals rise steadily to 5,040 messages. The consumer begins with an empty queue and processes 2,000 quote messages per second. After the burst, arrivals return to 40 per second.
- Baseline quote-to-trade ratio: 40 / 4 = 10
- Burst ratio: 5,040 / 4 = 1,260
- Quote messages processed during the burst: 2,000
- End-of-burst backlog: 5,040 − 2,000 = 3,040 messages
- Net catch-up rate: 2,000 − 40 = 1,960 messages per second
- Catch-up duration: 3,040 / 1,960 = 1.551 seconds
The consumer ends the burst 3,040 quote messages behind. With 40 new messages arriving each second, its 1,960-message net processing rate clears the backlog in 1.55 seconds, and every quote it shows until then lags the market.
The queue clears in 1.55 seconds. For that long the consumer works through the backlog while new messages keep arriving, so every update it shows lags the market.
How regulators and exchanges treat quote stuffing
CFTC guidance includes orders or cancellations intended to overload quotation systems among its examples under the statutory spoofing prohibition. That provision requires proof of intent. Cboe Rule 8.23 also addresses intentional overload and intentional or reckless messages with disruptive potential.
Surveillance evaluates message patterns, context and intent together. Market-data products differ in detail: order-level feeds carry order references, and some attributed messages also identify the submitting market participant.
What research found about intense quoting episodes
Egginton, Van Ness and Van Ness studied intense quoting episodes in 2010 U.S. equity data. Their research found at least one episode in more than 74% of exchange-listed securities and measured lower liquidity, higher trading costs and greater short-term volatility during those episodes.
Quote messages vs completed trades
Market depth shows displayed prices and quantities, where a quote burst lands. Time and sales shows completed trades. Completed contract activity appears in options flow, which records executions rather than quote messages.
Limitations of quote stuffing analysis
- A burst leaves the book stale. The example’s consumer works through a 3,040-message queue for 1.55 seconds, and every quote it shows in that time lags the market, so use limit orders until it catches up.
- Fast markets lift the ratio too. Ordinary quoting rises as prices move, so read a 1,260 quote-to-trade ratio beside the price action before you call it stuffing.
- Intent is the legal test. The CFTC provision requires it, and surveillance reads message patterns and context to find it.
- Expect worse fills during a burst. Egginton, Van Ness and Van Ness measured lower liquidity, higher trading costs and greater short-term volatility during intense quoting episodes in 2010 data, so size down or let the burst pass before you cross the spread.
- Capacity sets the delay. A consumer that processes 5,040 messages a second takes the same burst with no queue, and a queue that already holds messages when the burst starts takes longer than 1.55 seconds to clear.
Bottom line
Quote stuffing overloads processing capacity, so a message burst outruns a system and leaves updates waiting in a queue. Read it through message rates, queue size and the quote-to-trade ratio, and use limit orders while the book catches up. The rules turn on intent or recklessness, which surveillance reads from message patterns and context.
Clear answers
Frequently Asked Questions
Sources and Further Reading
- CFTC Interpretive Guidance and Policy Statement on Disruptive Practices, 78 FR 31890 (May 28, 2013)
- Cboe Exchange, Inc. Rulebook, Rule 8.23 Disruptive Practices
- FINRA Rule 5210: Publication of Transactions and Quotations
- Egginton, Van Ness and Van Ness, Quote Stuffing, Financial Management 45(3), 2016
- SEC: Market Information Data Analytics System
How entries are sourced, calculated and updated: editorial methodology.