How Long a Market Order Takes to Fill
Venue timestamp to consolidated feed, in microsecondsranking ·
2026-10-02 · 5×3
Median quoted spread in basis points, same sessionranking ·
2026-10-02 · 5×3
Prints in a pinned 30-minute window, five household namesranking ·
2026-10-02 · 5×4
Seconds between prints, by how often a name tradesranking ·
2026-10-02 · 5×3
What Exchange Colocation Costs Per Month
Quote updates in one hour, five names, 10:00 to 11:00 ET on 10 June 2026ranking ·
2026-09-30 · 5×3
SPY quote updates per millisecond, 10:00 to 10:05 ET on 10 June 2026ranking ·
2026-09-30 · 5×3
SPY trade messages per second by half hour, 10 June 2026series ·
2026-09-30 · 32×3
Mass Cancels and Purge Ports Explained
AAPL quoted spread in basis points, first hour of April 7, 2025series ·
2026-09-30 · 60×3
AAPL quote updates per second, volatile session vs quiet sessionseries ·
2026-09-30 · 54×4
How far AAPL travelled in each ten-minute bucket, same two sessionsseries ·
2026-09-30 · 54×3
Distinct option contracts that printed a trade on April 7, 2025ranking ·
2026-09-30 · 5×2
How Exchange Speed Bumps Work
Average spacing between national best quote updates, one fixed hourranking ·
2026-09-28 · 5×3
Quote messages per minute against distinct best-bid prices (NVDA)series ·
2026-09-28 · 60×3
Which venues held the best bid, half-hour slice (AAPL)ranking ·
2026-09-28 · 16×4
Latency Models in HFT Backtests Explained
Trade prints per second through one full session, ET clockseries ·
2026-09-24 · 78×2
How long a top of book quote survives, by latency budgetranking ·
2026-09-24 · 9×2
Average distance the mid price travels over each latency horizonranking ·
2026-09-24 · 4×2
Share of quotes replaced within 5 ms and 50 ms, by symbolranking ·
2026-09-24 · 5×3
Market Data Timestamps: SIP vs Exchange Clocks
Off exchange AAPL prints by reporting delay, 10 June 2026ranking ·
2026-08-14 · 5×3
The same twelve prints, ranked by venue clock and by tape clocktable ·
2026-08-14 · 12×7
SIP receive lag by venue, AAPL, 10 June 2026 (microseconds)ranking ·
2026-08-14 · 11×4
Prints that change bar when you switch clocks, by bar lengthranking ·
2026-08-14 · 4×3
PCAP Market Data: How Market Replay Works
Trade prints per second across the electronic day, June 10 2026series ·
2026-08-12 · 64×3
One minute of AAPL quotes, second by second, 09:30 ET on June 10 2026series ·
2026-08-12 · 60×3
Quote messages per trade, 09:30 to 10:00 ET on June 10 2026ranking ·
2026-08-12 · 5×3
AAPL quote clocks, SIP stamp minus exchange stamp, 09:30 to 11:00 ET on June 10 2026series ·
2026-08-12 · 9×4
SIP Feeds vs Direct Exchange Feeds
Which venues sat on the bid side of the quote record, KO on June 16, 2026ranking ·
2026-08-09 · 17×3
Quote update traffic across five household tickers, 10:00 to 11:00 a.m. ET on June 16, 2026table ·
2026-08-09 · 5×5
US equity exchanges that feed the consolidated tapetable ·
2026-08-09 · 18×4
Venue clock to consolidated tape clock, KO quotes on June 16, 2026series ·
2026-08-09 · 15×4
Venue timestamp to consolidated feed, in microseconds
Venue timestamp to consolidated feed, in microseconds
| symbol | median_report_microseconds | p95_report_microseconds |
|---|---|---|
| AAPL | 458 | 464971 |
| MSFT | 558 | 442909 |
| SPY | 608 | 357114 |
| NVDA | 7392 | 487876 |
| KO | 94509 | 501996 |
the exact SQL behind every number
SELECT
ticker AS symbol,
round(quantileDeterministic(0.5)(
toFloat64(dateDiff('microsecond', participant_timestamp, sip_timestamp)),
cityHash64(ticker, sequence_number)), 0) AS median_report_microseconds,
round(quantileDeterministic(0.95)(
toFloat64(dateDiff('microsecond', participant_timestamp, sip_timestamp)),
cityHash64(ticker, sequence_number)), 0) AS p95_report_microseconds
FROM global_markets.stocks_trades
WHERE ticker IN ('SPY', 'NVDA', 'AAPL', 'MSFT', 'KO')
AND sip_timestamp >= toDateTime('2026-09-16 14:00:00', 'UTC')
AND sip_timestamp < toDateTime('2026-09-16 14:30:00', 'UTC')
AND participant_timestamp > toDateTime('2026-09-16 00:00:00', 'UTC')
AND participant_timestamp <= sip_timestamp
GROUP BY ticker
HAVING count() > 100
ORDER BY median_report_microseconds
More from this analysisHow Long a Market Order Takes to Fill
Median quoted spread in basis points, same session
ranking 5×3
→
Prints in a pinned 30-minute window, five household names
ranking 5×4
→
Seconds between prints, by how often a name trades
ranking 5×3
→
Quoted spread across six household names, 11:00 to 11:30 a.m. ET on Sep 16, 2026
ranking 6×3
→
See all 3,256 queries →