VENUE
QCX LLC, trading under the assumed name Polymarket US. Designated contract market, designated July 9, 2025. Access is a direct account with identity verification plus your own key, and the public catalogue answers without one.
Read 2026-09-28. The CFTC's Designated Contract Markets list, under Industry Filings, records the assumed name against the QCX entity.
WATCHES
Each day, a check reads one published page and compares it with the read recorded here. Each row below says whether the page was unchanged, moved, or could not be read at the last check.
Splitrule checks 4 pages on this venue. This page shows 1 of them.
SCHEDULE
Dated reference observed 2026-09-17T14:24:51Z. This is not current fee authority. Formula estimates precede fill rounding, order caps, and volume rebates. These terms come from the venue's own published schedule, not from a summary of it. The fee is proportional to the price times one minus the price, so it peaks at 50¢ and falls toward the tails. Any flat per-contract quote is wrong on this venue by construction.
taker: 0.0695 x C x P x (1-P) 1.74¢ per contract at 50¢ maker: rebate of 0.0125 x C x P x (1-P) 0.31¢ per contract at 50¢
The selected schedule takes effect 2026-09-17. The schedule names no settlement, membership or withdrawal fee. Splitrule reads that silence as unstated rather than as zero.
FEE PER CONTRACT
An event contract pays a dollar or nothing, which makes the answer exact rather than approximate. A fee of one cent per contract is one percentage point of true probability you have to be right by, on top of the price. So the minimum edge a trade demands is not a matter of opinion or of market conditions. It is the fee schedule, read out loud.
Nothing below reads a market. The prices are a grid of whole cents chosen here, and each figure is a property of the published schedule this page names.
A taker who buys and later sells at the same price has paid the fee twice before being right about anything. That arithmetic decides whether a spread is an opportunity or a bill. The fee is proportional to the price times one minus the price, so it is largest at 50¢ and smallest at the edges.
| Price | Taker, held | Taker round trip | Maker rebate |
|---|---|---|---|
| 5¢ | 0.33¢ | 0.66¢ | 0.06¢ |
| 10¢ | 0.63¢ | 1.25¢ | 0.11¢ |
| 25¢ | 1.30¢ | 2.61¢ | 0.23¢ |
| 50¢ | 1.74¢ | 3.48¢ | 0.31¢ |
| 75¢ | 1.30¢ | 2.61¢ | 0.23¢ |
| 90¢ | 0.63¢ | 1.25¢ | 0.11¢ |
| 95¢ | 0.33¢ | 0.66¢ | 0.06¢ |
The takeaway. A taker round trip at 50¢ costs 3.48 points of edge on this venue. The maker column is a rebate at every price when its coefficient is negative. A zero coefficient gives no maker rebate. The amounts depend on the selected schedule.
edge = coefficient x P x (1 - P) x legs x 100, in cents per contract
none. The frontier runs on the unrounded rate. Rounding is per fill, so a per-contract figure with a rounding step would describe one fill size, not a rate
exact on the published terms. A $1 payout makes a cent of fee identical to a point of true-probability edge, so this is the edge the trade demands, not an approximation of it
§ permalink to this receiptedge = coefficient x P x (1 - P) x legs x 100, in cents per contract
none. The frontier runs on the unrounded rate. Rounding is per fill, so a per-contract figure with a rounding step would describe one fill size, not a rate
exact on the published terms. A $1 payout makes a cent of fee identical to a point of true-probability edge, so this is the edge the trade demands, not an approximation of it
§ permalink to this receiptedge = coefficient x P x (1 - P) x legs x 100, in cents per contract
none. The frontier runs on the unrounded rate. Rounding is per fill, so a per-contract figure with a rounding step would describe one fill size, not a rate
exact on the published terms. A $1 payout makes a cent of fee identical to a point of true-probability edge, so this is the edge the trade demands, not an approximation of it
§ permalink to this receiptThe schedule panel above carries the terms these figures come from, and the rounding rule that decides who is right on a small fill. Run your own numbers on any market’s own page, under the filter.
Public fee schedules and incentive programs from the venues' own documents and regulatory filings, checked October 5, 2026. Contracts differ between venues, so equal fees do not make two markets equivalent. A stated scenario cannot establish anyone’s eligibility or rewards.
Current. Takers pay feeBPS(p) = peakBPS × min(p, 1 − p) / 0.5 of the trade value. The documented peak is 1.5% at a price of 0.50. Each market publishes its own 100-bucket schedule. Makers pay no fee. No rounding rule is published.
Sources: Order Book (trading guide), read October 2, 2026; API Reference, read October 2, 2026.
Current. Each market sets three fractions of the trade value per side: liquidity providers, treasury and distributor. The guide states 0% to 2% on buys. Shares received depend on pool state.
Sources: AMM (trading guide), read October 2, 2026; API Reference, read October 2, 2026.
Current. The venue's own app adds $0.0085 per transaction for blockchain gas. Other interfaces can charge their own frontend fee from 0% to 5%.
Sources: AMM (trading guide), read October 2, 2026; API Reference, read October 2, 2026.
Active. Described as the arrangement at launch. No end date is published.
Eligibility: Makers whose resting limit orders fill on order-book markets.
Payout: 100% of taker fees revert to makers, paid periodically. The allocation rule, period and minimum are not published.
Limit orders near the midpoint also weigh more in weekly Points distributions. Points are not a cash rebate.
Sources: Order Book (trading guide), read October 2, 2026.
Current. Opening a position has no fee. Closing and settlement can carry a fee, but the base outcome rate is not published, and the HIP-4 overview says fees are currently zero for initial testing. A template outcome multiplies any base rate by scale + max(scale, 1). Makers get no rebates. Builder codes can add an interface fee in the quote token.
Sources: Fees: outcome tokens, read October 2, 2026; HIP-4: Outcome markets, read October 2, 2026; HIP-4 deployer actions (labelled testnet-only), read October 2, 2026.
Markets also shows stored market data from OG (Crypto.com), Limitless, Predict.fun, Novig, Kalshi, Opinion, and Pascal. It shows the odds that The Odds API, a data supplier, reports for ProphetX, BetOpenly, Betfair Exchange (UK), and Matchbook. Activity shows the dated public reports of Novig, ProphetX, and ForecastEx. Splitrule calculates no fee for these sources. A market row shows the source’s own fee text when the source gives one, and says so when it gives none.
Ask these questions through Splitrule’s research tools in Claude or ChatGPT
RULES DRIFT
Venues publish the rules a market settles by, and they edit them. Nobody publishes the history. This is 41 intervals of that history for one venue, read field by field.
One venue. Every figure below is Polymarket US and says nothing about any other venue. Nothing on this screen is folded together with a reading from anywhere else.
The axis runs from 2026-07-31 to 2026-09-13, joined as 41 intervals from 42 usable captures. The store contains 7 quarantined captures from failed walks. Across the full source history, 4 intervals span 7 quarantined captures. 0 quarantined captures precede the source history, and 0 follow it.
The census is rebuilt daily and this page quotes ONE build of it, named in every receipt.
WHAT MOVED
Over 886,138 carried market-intervals, 16316 carried a change of any kind. Wording changed, void language changed, deadlines moved and rules text changed length bracket. Not one market changed the source its outcome is read from.
| Value | What it is | Share |
|---|---|---|
| 13343 | Wording only | 100.0% of the longest bar, 13343 |
| 2511 | Void language | 18.8% of the longest bar, 13343 |
| 321 | Deadline moved | 2.4% of the longest bar, 13343 |
| 141 | Length bracket | 1.1% of the longest bar, 13343 |
The day-scale rate is 16152 changes over 775,542 carried market-intervals, in the 33 intervals at least 20 hours wide and measured end to end by one instrument: 2.0827%. An all-interval rate would mix 0.01-hour and 120-hour intervals under one denominator.
One day-scale interval crosses an instrument boundary. Its 144 changes over 15,693 carried market-intervals stay outside the single-era rate. Each interval contributes once, regardless of its transition count. From 2026-08-12T14:30:28.312Z to 2026-08-13T14:30:27.981Z, instrument era 1 to 2: 144 changes over 15,693 carried market-intervals. The registry transition at 2026-08-12T23:11:16Z, era 1 to 2, records VENUE-EDIT. The probe attributes the observed changes to the venue rather than to the instrument change.
16,152 / 775,542 = 2.0827%
four decimal places, display only; the counts are integers and are not rounded
a first reading over 41 intervals (2026-07-31 to 2026-09-13) on one venue, and the subset matters: the all-interval rate mixes 0.01-hour intervals with 120-hour ones under one denominator, so it is not quoted here. One day-scale interval crosses an instrument boundary. Its 144 changes over 15,693 carried market-intervals stay outside the single-era rate. Each interval contributes once, regardless of its transition count. Including these boundary counts, the rate is 2.0596%. The adjacent receipt names each capture interval and recorded transition.
§ permalink to this receiptan interval is a boundary crossing when its two captures fall in different instrument eras; eras come from verification/rules-versions/instrument-boundaries.json. Each day-scale crossing contributes once, even when it contains several registry transitions
none — counts
One day-scale interval crosses an instrument boundary. Its 144 changes over 15,693 carried market-intervals stay outside the single-era rate. Each interval contributes once, regardless of its transition count. From 2026-08-12T14:30:28.312Z to 2026-08-13T14:30:27.981Z, instrument era 1 to 2: 144 changes over 15,693 carried market-intervals. The registry transition at 2026-08-12T23:11:16Z, era 1 to 2, records VENUE-EDIT. The probe attributes the observed changes to the venue rather than to the instrument change. Probe verdicts describe recorded transitions, not a new probe of these capture endpoints. Boundary crossings stay outside the single-era rate regardless of attribution
§ permalink to this receiptone carried market-interval is one market present in BOTH captures of one interval, summed over the 41 intervals; a market added or removed inside an interval cannot have changed across it
none — a count
markets recur across intervals by construction, so this is market-intervals and never markets: the same market carried through all 41 intervals contributes 41. Added 385,196 and removed 306,136 are counted separately and never netted
§ permalink to this receiptSOURCE-SWAP fires when the set of source hostnames in the rules text moves with both sides non-empty; SOURCE-VANISHED and SOURCE-APPEARED read that same set emptying or filling from empty, each while the prose itself remains; VOID-LANGUAGE-CHANGE fires when the frozen void-marker set moves in either direction
none — counts
the extractor reads hostnames and frozen markers out of the prose, so a source that changed without its hostname changing would not fire. It is a floor on stability, not a proof of it
§ permalink to this receiptInterval by interval, on a date axis. In this chart, 4 intervals span 7 quarantined captures. Outside this chart, 0 intervals span 0 quarantined captures.
2 of the 28 intervals carried no rules change at all
| Interval (end, UTC, and width) | Markets changed | Note |
|---|---|---|
| 09 Aug 14:33 (24.05h) | 4 | |
| 10 Aug 14:30 (23.95h) | 13 | |
| 10–12 Aug 14:30 (2 days, 48h) | 50 | No usable capture between endpoints. 1 quarantined capture inside |
| 13 Aug 14:30 (24h) | 144 | Instrument changed |
| 14 Aug 14:30 (24h) | 1 | |
| 15 Aug 14:32 (24.04h) | 2 | |
| 16 Aug 14:31 (23.98h) | 9 | |
| 17 Aug 14:30 (23.99h) | 30 | |
| 18 Aug 14:30 (24h) | 18 | |
| 19 Aug 14:30 (24h) | 8 | |
| 20 Aug 14:30 (23.99h) | 5 | |
| 21 Aug 14:30 (24h) | 5 | |
| 22 Aug 15:00 (24.5h) | 2 | |
| 22–24 Aug 14:30 (47.5h) | 3 | No usable capture between endpoints. 1 quarantined capture inside |
| 25 Aug 14:30 (24h) | 61 | |
| 26 Aug 14:30 (24h) | 3 | |
| 27 Aug 14:30 (24.01h) | 1 | |
| 28 Aug 15:01 (24.5h) | 21 | |
| 28–30 Aug 14:31 (47.5h) | 0 | No usable capture between endpoints. 1 quarantined capture inside |
| 31 Aug 15:47 (25.28h) | 26 | |
| 01 Sep 14:30 (22.71h) | 22 | |
| 02 Sep 14:31 (24.02h) | 7 | |
| 03 Sep 14:31 (24h) | 5 | |
| 04 Sep 14:31 (23.99h) | 13 | |
| 05 Sep 14:31 (24h) | 8 | |
| 05–07 Sep 14:32 (48.01h) | 0 | No usable capture between endpoints |
| 08 Sep 14:32 (24.01h) | 12 | |
| 08–13 Sep 14:32 (5 days, 120h) | 15,766 | No usable capture between endpoints. 4 quarantined captures inside |
2026-07-31T22:03:53.939Z to 2026-09-13T14:32:12.073Z, joined as 41 consecutive intervals
the interval count is an integer and the dates are the capture stamps, to the day; interval hours below carry two decimal places, display only
The store contains 7 quarantined captures from failed walks. Across the full source history, 4 intervals span 7 quarantined captures. 0 quarantined captures precede the source history, and 0 follow it. In this chart, 4 intervals span 7 quarantined captures. Outside this chart, 0 intervals span 0 quarantined captures. The widest interval in the full source history is 120 hours.
§ permalink to this receiptsports rate 2.1239% over its own carried population, not over the venue's
four decimal places, display only
changes landed in 2 of the 10 categories (sports 16,174, politics 142), and the sports share is nearly uninformative on its own because sports is 85.9385% of the carried population. What survives is narrower: culture carried 23,908 market-intervals and changed none
§ permalink to this receiptWHAT THIS AXIS CANNOT SEE
A change is only visible on a market present in BOTH captures of an interval. Of the 367,994 distinct markets across 42 captures, 142,914 appear in exactly one, and that one is neither the first capture nor the last. A market seen once is never carried, so no edit to its rules text can fire here. A further 46,094 markets are seen once at one end of the window, where a single appearance says more about the window than about the market, and they are excluded from that count.
Per interval the same bound reads differently, over a different population: 385,196 of 1,271,334 rows in a later capture had no earlier row to compare against, 30.2986%. The first share counts distinct markets across the whole axis. The second counts rows inside 41 intervals. They are two measurements and they do not combine.
Where this venue edits. Every one of the 16316 changed market-intervals sits in 2 of the 10 categories the venue labels (sports 16,174, politics 142). The other 8 carried 36,803 market-intervals between them across the same 41 intervals and changed none. The category label is the venue's own. Splitrule derives no taxonomy of its own from the text.
A changed count does not say what changed. The two movers moved different things: sports — 13342 where wording changed, 2511 where void language changed, 321 where deadlines moved; politics — 141 where rules text changed length bracket, 1 where wording changed. A deadline shift replaces one date or clock token in text that otherwise holds. A length-bracket change is prose that crossed a size boundary while every tracked field held. The count column reads both as one changed market-interval, and the class column is what separates them.
| Category (venue label) | Carried market-intervals | Changed | Top class per change |
|---|---|---|---|
| sports | 761,534 | 16,174 | PROSE-EDIT-ONLY 13342, VOID-LANGUAGE-CHANGE 2511, DEADLINE-SHIFT 321 |
| politics | 87,801 | 142 | LENGTH-CLASS-CHANGE 141, PROSE-EDIT-ONLY 1 |
| culture | 23,908 | 0 | — |
| macro | 3,054 | 0 | — |
| finance | 3,003 | 0 | — |
| technology | 2,767 | 0 | — |
| crypto | 2,029 | 0 | — |
| climate | 1,161 | 0 | — |
| science | 477 | 0 | — |
| geopolitics | 404 | 0 | — |
When the changes happened. The 16316 changed market-intervals are not spread evenly across the 41 intervals: sports changed in 35 of them, politics in 1. All 142 politics changes fall in one interval, from 2026-08-12 to 2026-08-13. In the other 40 intervals politics changed none.
Which changes moved a term of the contract. A change counts here when it moved one of four tracked token sets. The four are a source host, a source reference, a deadline or clock token, and a void marker. Sports: 2,832 of 761,534 carried market-intervals. Politics: 0 of 87,801 carried market-intervals.
A zero there is not a finding that a category held still. Politics carries 142 of the 16316 changes in this corpus, and none of them moved a term this instrument reads as a term. Of those, 141 crossed a text-size bracket while all four term sets stayed byte-identical. Text that names no date, host, URL or void marker is consistent with boilerplate. It is equally consistent with an operative sentence this extractor cannot read. The corpus cannot tell the two apart. This is a lower bound on what changed, never an estimate of it.
In 41 capture intervals of Polymarket US rules text — 886,138 carried market-intervals between 2026-07-31 and 2026-09-13 — Splitrule recorded zero resolution-source swaps and zero sources removed or added. All 2,832 term-moving edits it did record sat in sports markets.
a change is TERM-MOVING when it moved one of four tracked token sets: a source host, a source reference, a deadline or clock token, or a void marker. Summed over the 41 intervals, 2,832 of the 16,316 changed market-intervals did, over 886,138 carried
none — counts
markets recur across intervals by construction, so this is market-intervals and never markets: a market that moved a deadline in two intervals is counted twice, and the distinct-market count is not this number and is not in the census. It is also a LOWER BOUND on materiality and never an estimate of it: a criterion rewritten in prose that names no new date, host, URL or void marker does not fire, and 141 of the 16,316 changes are exactly that shape: they crossed a text-size bracket with all four token sets byte-identical. That text is consistent with boilerplate and equally consistent with an operative sentence this extractor cannot read, and the corpus cannot tell the two apart
§ permalink to this receiptWhere the Splitrule instrument moved. The day-scale rate above counts 16152 changes over 775,542 carried market-intervals, and every one of them is sports. It contains no politics change. Category absence from that rate alone does not locate its changes in a boundary interval.
One day-scale interval crosses an instrument boundary. Its 144 changes over 15,693 carried market-intervals stay outside the single-era rate. Each interval contributes once, regardless of its transition count. From 2026-08-12T14:30:28.312Z to 2026-08-13T14:30:27.981Z, instrument era 1 to 2: 144 changes over 15,693 carried market-intervals. The registry transition at 2026-08-12T23:11:16Z, era 1 to 2, records VENUE-EDIT. The probe attributes the observed changes to the venue rather than to the instrument change.
a market is CARRIED when it is present in both captures of an interval, and a drift class fires only on a carried market; a market that appears in exactly one capture is therefore never carried and no edit to its rules text can ever fire: 38.8360%
four decimal places on the share, display only; the counts are exact
a market seen once in the FIRST or the LAST capture is an edge effect of where the window stops rather than a market that lived and died between two walks, so the 46,094 of those are counted and excluded from the value. This bounds the instrument, not the venue: the walk reads every open market the catalogue returns, and the same store carries rotating categories whose markets ARE caught when the venue lists them across two walks
§ permalink to this receiptsummed over every interval: rows in the later capture of the interval whose market was absent from the earlier one, over all rows in the later capture: 30.2986%
four decimal places on the share, display only; the counts are exact
this is a share of ROWS INSIDE INTERVALS and the never-carried share is a share of DISTINCT MARKETS over the whole axis. The two shares have different numerators, different denominators and different questions. Neither is a restatement of the other
§ permalink to this receiptBoth shares are Splitrule's measurement of its own reach, not a statement about the venue.
ADJUDICATED EVENTS
An event is one thing that happened at one venue, over one named interval, with the cause adjudicated rather than asserted. Each one has its own page, its own permanent address, and a derivation under every number on it.
This feed is a record, not an alert. It says what Splitrule saw and adjudicated. It does not promise to catch the next one, and it monitors no market for any reader. Subscribe at feed.xml.
13 Aug 2026
141 of the 1,047 carried politics markets changed their rules text across 24 hours. Adjudicated cause: VENUE-EDIT.
Events are rare by construction. Splitrule publishes one when a probe adjudicates a cause, and never to fill a page.
SPREAD BY HOUR
Splitrule keeps every scheduled read of the open catalogue. The band below shows the spread by hour of day as quantiles. An hour without a measurement shows as a gap.
The history charts need JavaScript. The numbers behind them are served as plain JSON at /api/history.json, which needs none.
CROSS-LISTED
A confirmed mark means that Splitrule found the same question on another venue. Hover over the count for its denominator, its row limit and its capture. This panel names no other venue and shows no market text or price. A confirmed mark does not establish hedge eligibility.
DISPUTES
These incidents come from an oracle system that Splitrule does not cover. The record is collapsed because it gives context, not a reading of this venue. Each entry says which of the two it is.
The dispute record needs JavaScript. It is editorial content with public citations.