77 / 97
Quote Staleness
Working definition
The age of a displayed quote relative to the current state of the market it claims to represent; a stale quote advertises a price that is no longer executable.
A price feed is a claim: this is where the market is now. Quote staleness measures how far behind “now” that claim runs. Every feed carries some latency — aggregation, transport, and display all take time — so staleness is a spectrum, not a binary. The question is whether a venue’s staleness is an unavoidable few milliseconds of plumbing or a structural property that changes what its clients can know.
Staleness has several distinct causes. Infrastructure latency delays honest quotes. Throttled update cadence — republishing the feed at a coarser interval than the underlying market moves — leaves each displayed quote representing an increasingly historical state between updates. Aggregation across liquidity providers can hold quotes while a composite is recalculated. And at the deliberate end of the spectrum sits tick suppression during fast markets, which is censoring: staleness as policy rather than physics.
The consequences run in two directions. For the trader, decisions keyed to a stale quote are decisions about a market that no longer exists; the resulting fills arrive with systematic slippage, concentrated in exactly the volatile moments when the displayed price lagged most. In the other direction, a persistently slow venue is exploitable by anyone with a faster view of the same market — the raw material of latency arbitrage — and venues aware of this tend to respond with defensive mechanics such as last look, asymmetric rejection, or further throttling, each of which degrades execution for ordinary clients.
Staleness is measurable from the feed itself. Inter-tick arrival distributions, update cadence under calm versus volatile regimes, and timestamp comparison against a multi-venue reference during the same windows together characterise how current a feed actually is. None of that requires the venue’s cooperation — only a recorded feed and its timestamps.
Commonly confused with
Neighbouring concepts that get used interchangeably, and the distinction that actually separates them.
- Quote fade
A stale quote should have moved and did not — it is a price that stopped updating. A faded quote was current and then withdrawn as you reached for it. One is a feed running behind; the other is a feed reacting to you.
- Latency
Infrastructure latency is one cause of staleness and not the same thing as it. Aggregation delays, throttled update cadence and outright tick suppression produce stale quotes on fast connections. A venue can be quick and stale at once, which is why measuring the network tells you little.
- Censoring
Censoring names the same thinning viewed as a property of the record rather than of the clock: a feed whose updates are dropped or rate-limited through the fastest intervals. Whether any instance is plumbing or policy is not observable from the tape, which is why the honest question is what the feed preserved rather than what anyone intended.
- Liquidity void
A void is real absence of depth — the market genuinely has nothing there. Staleness is a claim about depth that has stopped being updated. A stale feed can display comfortable depth into a void, which is the worst of both.
How to measure it in your own data
A definition you cannot test is a definition you have to take on trust. This is the shortest honest route from the concept to a number you computed yourself.
- Records you need
A recorded feed with its own timestamps, and a multi-venue reference over the same windows. Neither requires the venue's cooperation, which is what makes this measurable at all.
- What you compute
Inter-tick arrival distributions; update cadence compared between calm and volatile regimes; and timestamp comparison against the reference during the same windows. The calm-versus-volatile split is the informative one — a single average cadence conceals the behaviour that matters.
- What the answer tells you
Every feed carries some latency, so staleness is a spectrum rather than a binary, and a few milliseconds of plumbing is not a finding. What is a finding is cadence that coarsens exactly as the market speeds up: a feed republishing at a slower interval than the underlying moves leaves each displayed quote representing an increasingly historical state, precisely when accuracy matters most.
If this has already cost you
How current a feed actually is can be established from its own timestamps against a reference, without the venue’s cooperation.
- Feed Fidelity Assay“Is the feed my terminal shows me behaving consistently?”Will not establish: A verdict on your broker. One exported log from one terminal measures your feed as you received it — venue-side behaviour, other account tiers and intent are all outside what this data can carry.
Intake is not open yet, so none of these can be commissioned today. They are listed here so you know the measurement exists and what it would and would not settle — the launch list hears first.
Questions and answers
Is quote staleness just a slow internet connection?
No, and assuming so is how the question gets dropped. Your connection affects when you receive the feed; staleness is about how current the feed was when it was published. Throttled update cadence, aggregation delays and tick suppression all produce stale quotes that arrive at you instantly.
How stale is too stale?
There is no universal threshold, because the relevant comparison is against the market's own speed rather than against a clock. A cadence that is perfectly adequate in a calm session can be badly behind during a release. That is why the measurement compares regimes rather than reporting a single average.
Can I detect staleness without the venue's cooperation?
Yes. Inter-tick arrival distributions, cadence under calm versus volatile conditions, and timestamp comparison against a multi-venue reference all come from a recorded feed and its timestamps. Nothing in that requires access to the venue's systems or its agreement to be measured.
Why would a venue slow its own feed deliberately?
A persistently slow venue is exploitable by anyone with a faster view of the same market, so slowing or throttling can be a defence against being picked off. The difficulty is that the defence is not selective: the same throttling that frustrates a latency arbitrageur also degrades execution for ordinary clients, who were not the threat.
Related terms
Derived from the links this entry makes and the entries that link back to it.
Where the term is used
Instrument pages whose published copy uses this term. Each page states what it measures and what it does not establish.
In the research
Quote Staleness comes up in two research notes on this site.