AG6 archive
One collector's capture, mirrored unmodified. Six of its hours have been audited against on-chain fills; no hour has a coverage verdict.
third-party/ag6/, 2026-08-01T13 to 2026-08-15T09,
one Parquet file per UTC hour, named
polymarket_orderbook_YYYY-MM-DDTHH.parquet. 136
objects, 58.1 GiB, measured 2026-08-29.
We serve these bytes. We do not vouch for them. They were collected by AG6 and mirrored here exactly as received. One machine recorded them, so nothing cross-checks them, and no hour has a coverage verdict. As for checking, the quality audit was planned in advance and has now run, and what it measured is published with this corpus; the results are at the bottom of this page.
Why it is here
It closes part of a gap nothing else could. PMXT's archive ends at
2026-08-10T00 and that final hour is truncated - it
holds 59,132,258 rows against AG6's
86,519,038 for the same hour, because PMXT's capture stopped
partway through. AG6's copy of that hour is whole.
Its run also begins before PMXT's last hour rather than after it, so the
two overlap instead of meeting end to end. Where they overlap you have two captures of the
same hour by two different parties, the only such overlap in this archive outside
v3/.
It does not close the whole gap. 68 hours between
2026-08-15T10 and 2026-08-18T05 are still
missing, with no known source for the venue as a whole;
a contributed tier holds crypto Up-or-Down markets for
those hours only. If you can help us fill the gap, please DM
x.com/PendulumFlow.
Columns
The same sixteen columns as pmxt/v2/. The two schemas were read from the
footers of the served files and compared field by field, including nullability.
timestamp_receivedtimestamp[ms, tz=UTC]never null
When the exporter ingested the event. The hour partition key and the sort key. Milliseconds, so rows tie constantly.
timestamptimestamp[ms, tz=UTC]never null
Polymarket's own timestamp for the event. Can be stale on a quiet market, so it is not a substitute for the receive time.
marketfixed_size_binary[66]never null
Condition id, 0x and 64 hex characters. Cast to text on read.
event_typestringnever null
price_change (~99.7% of rows), book,
last_trade_price, tick_size_change.
asset_idstringnever null
Outcome token id, as a decimal string.
bidsstringnullable
Resting bid depth on book rows, as a JSON string
you have to parse. Null on every other event type.
asksstringnullable
Resting ask depth on book rows, as a JSON string. Null elsewhere.
pricedecimal(9,4)nullable
On price_change and last_trade_price.
sizedecimal(18,6)nullable
On price_change and last_trade_price.
sidestringnullable
On price_change and last_trade_price.
best_biddecimal(9,4)nullable
On book rows. Frequently null even there in the
files we serve, so recompute the touch from bids rather than relying on it.
best_askdecimal(9,4)nullable
On book rows, and null as often as best_bid is.
Recompute it from asks.
fee_rate_bpsuint16nullable
On last_trade_price.
transaction_hashstringnullable
The on-chain hash, on last_trade_price.
old_tick_sizedecimal(9,4)nullable
On tick_size_change.
new_tick_sizedecimal(9,4)nullable
On tick_size_change.
16 columns. The types above were read from
the footer of 2026-08-15T08,
not copied from documentation, so they are what your own reader will report.
nullable says whether the schema permits a null at all, which is not the same as
how often one appears.
The reading notes on the V2 overview apply here unchanged,
including that best_bid and best_ask are frequently null even on
book rows, and that a touched price is not a fill, see A fill, from a price you touched.
What is wrong with this data
AG6 is missing hours, and a few of the markets that traded.
- It holds 136 of the 333 hours in its span. 2026-08-01T13 to 2026-08-15T09, 41% hour density. The 197 hours it does not hold are absent outright. The hours it does hold are as full as their neighbours. hour census over the served AG6 objects, 2026-08-31
- Traded coverage is nearly whole. Ratio 0.9977 across the six AG6 hours picked in advance for the audit, 47 traded markets missed in total. On the hour read in depth, four markets traded with no book, with fill counts of 171, 5, 4 and 2, and no phantom fills at all. the audit's check against on-chain fills, 2026-08-31
- Its market universe is broader than pmxt/v2's. On the seven hours they share, AG6 lacks 0.55% of the markets v2 holds, and v2 lacks 3.09% of the markets AG6 holds. the audit's comparison of the market sets on the seven shared hours, 2026-08-31
- On the hours they share, AG6 is more complete than pmxt/v2 on the markets that traded. Both corpora measured against the same set of on-chain fills across all seven shared hours: AG6 0.9972, pmxt/v2 0.9896. The split behind that: markets present in v2 and absent from AG6 are 47 traded and 4,298 untraded; markets present in AG6 and absent from v2 are 246 traded and 24,688 untraded. So most of what AG6 lacks never traded, and only 47 markets that did trade in those seven hours are in v2 and absent from AG6. That is a different count from the 47 traded markets missed in the six drawn hours above, though the two happen to match. No phantom fills appear in any of the seven hours. That is expected: phantom fills ended on 2026-06-19 and these hours are all in August. It is a consistency check passing, not evidence that the column is never true. on-chain fills for the 7 shared hours, joined to both corpora, 2026-08-31
Pendulum Flow Discord
Something wrong with an hour, or a file you need that is not here? Say so in the Discord, where the people who run the site and prediction market data geeks compare notes.
Join the Discord