Skip to research
PerpDEXIndependent research
RESEARCH · MONTHLY · VARIATIONAL
PERPDEX RESEARCH · OCTOBER 2026

Variational October 2026 Observation Report

Capital and open interest fell, leading markets recorded less volume, and the Treasury balance rose.

Published Nov 2, 2026 · Revised Nov 3, 2026 | October 2026 (UTC) | Data cutoff: Nov 1, 00:00
ContentsFindingsScaleParticipationMarketsTreasuryCostsData

October at a glance

  1. Capital and open interest fell back from early-October highs.
  2. Established markets’ volume readings fell 39.5% between the first and last weeks of October, with decreases concentrated in two US equity-index contracts.
  3. The Treasury balance rose by approximately USDC 2.60M from month-start to month-end. This is not confirmed revenue.
Month-end total value locked (TVL)USDC 223.75M-15.5%Month-end vs month-start | Variational API
Month-end reported two-sided OIUSDC 1.770B-13.2%Size of open contracts | month-end vs month-start
Estimated October volumeAbout USDC 78.15BEstimated from cumulative volume near month-start and month-endNot an exact calendar-month total; no full-month growth comparison
Net increase in Treasury balanceAbout USDC 2.60MMonth-end balance minus month-start balanceA balance increase is not confirmed revenue

All times are UTC. Coverage is stated beside each chart.

1. TVL and open interest fell back from their October highs

Both measures reached their highest recorded October levels on Oct 6. Month-end TVL was 19.2% below its monthly high, and reported two-sided OI was 23.8% below its own high.

How much value was locked each day on average?

USDC M 0 50 100 150 200 250 300 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 Oct 16–17 Missing daily values Daily average TVL 7-day mean
Unit: USDC M · Source: Variational official API historical snapshots · Oct 1–31 UTC; 29 complete days

Daily average TVL.

Evidence: View data · Calculation notes

How much open interest remained each day on average?

USDC B 0 0.5 1 1.5 2 2.5 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 Oct 16–17 Missing daily values Daily average two-sided OI 7-day mean
Unit: USDC B · Source: Variational official API historical snapshots · Oct 1–31 UTC; 29 complete days

Protocol-reported two-sided open interest.

Evidence: View data · Calculation notes

Daily average of the preceding 24 hours’ volume

USDC B 0 1 2 3 4 5 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 Oct 16–17 Missing daily values Daily average of past-24h volume 7-day mean
Unit: USDC B · Source: Variational official API historical snapshots · Oct 1–31 UTC; 29 complete days

The daily average of hourly past-24-hour volume readings, not trades executed that day.

Evidence: View data · Calculation notes

Source for all three charts: Variational API historical snapshots. Oct 1–31 UTC, with 29 complete days. Oct 16–17 are blank because hourly coverage is incomplete; the 7-day mean does not bridge gaps. TVL and OI are daily averages of hourly readings, rather than the month-end values in the cards.

Lower prices can also reduce dollar-denominated OI: Bitcoin and Ether fell 8.6% and 16.7% between readings near 23:00 on Oct 1 and Oct 31, but these readings do not identify positions opened or closed. The causes of the TVL change remain unverified.

View scale details and other comparison periods
MeasureMonth-startMonth-endChange from month-startMonthly average
TVLUSDC 264.76MUSDC 223.75M-15.5%USDC 248.71M
Reported two-sided OIUSDC 2.038BUSDC 1.770B-13.2%USDC 2.019B

The average past-24-hour volume across the month’s hourly readings was USDC 2.564B; adding these readings does not give monthly turnover. A separate comparison of Sep 24–30 with Oct 22–28 uses 549 markets with complete data in both periods: average volume readings fell 31.8% and OI fell 8.5%. This is not a full-month comparison.

2. Fewer new pools; existing addresses supplied most covered deposit amounts

Settlement pools hold funds for bilateral settlement between a user and the liquidity provider, OLP; pool counts are not user sign-ups.

October had 10,505 new settlement pools, 32.7% fewer than the full month of September. During Oct 1–30, an average of 15,717 pools had funding or return flows each day, peaking at 19,680 on Oct 29. The daily average for Oct 22–28 was 7.4% above Oct 1–7.

New settlement pools by day

Pools 0 200 400 600 800 1,000 1,200 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 New settlement pools
Unit: Settlement pools · Source: Arbitrum settlement-pool creation events

Oct 1–31 UTC; all 31 days covered. Each bar counts pools created on-chain that day, 10,505 in total; these are not sign-ups.

Evidence: View data · Calculation notes

Settlement pools with funding or return flows each day

Pools 0 3,000 6,000 9,000 12,000 15,000 18,000 21,000 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 Pools with funding/return flows
Unit: Settlement pools · Source: Daily funding/return-flow aggregates

Oct 1–31 UTC; 30 valid days. Each bar counts distinct settlement pools with the specified funding/return events that day, not active traders; Oct 31 is missing.

Evidence: View data · Calculation notes

Covered ordinary USDC deposits totalled USDC 211.34M. Existing addresses deposited USDC 174.89M, or 82.8% of that amount. “Existing” means user-side addresses already present in the data before October. This is a share of deposit amounts, not addresses or retention.

Deposits cover only the ordinary channel from Oct 6 and do not deduct withdrawals; they are not platform-wide net inflows.

View data and calculation notes

View daily deposits

How much USDC was deposited each day through the covered channel?

USDC M 0 3 6 9 12 15 18 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 Deposits before withdrawals
Unit: USDC M · Source: Verified ordinary USDC deposit events

Each bar is the USDC deposited through the covered channel that day, before withdrawals. Oct 1–5 are not covered; Oct 6 starts at 18:40; Oct 7–31 still cover only this channel (UTC).

Evidence: View data · Calculation notes

For example, deposits on Oct 7 totalled approximately USDC 10.30M. Repeated deposits are counted each time; this flow does not establish new capital, net inflows or TVL growth.

View depositing addresses and seven-day follow-up

Addresses depositing each day

Addresses 0 500 1,000 1,500 2,000 2,500 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 Addresses seen before October Addresses first seen in October 7-day mean
Unit: Addresses · Source: Verified ordinary USDC deposit events

Oct 6, 18:40–Oct 31 UTC; 25 complete days and one partial day. Each bar counts distinct addresses depositing through the covered channel that day, not active traders; the 7-day mean does not bridge gaps.

Evidence: View data · Calculation notes

The covered period includes 17,406 distinct depositing addresses; this is not the sum of daily counts. The median cumulative amount per address was USDC 499.25; 90% of addresses deposited no more than USDC 11,751.10 in total. The ten largest depositors accounted for 18.2% of covered deposits.

Deposits were recorded on one day for 10,891 addresses, on 2–7 days for 5,900, and on 8+ days for 615. Newer addresses have shorter follow-up, so this is not retention. Among 6,887 new pools observed for a full seven days with complete channel coverage, 4,998 (72.6%) had a qualifying deposit. The denominator is not all 10,505 new pools; shorter or incomplete follow-up is not counted as failure.

View deposit-day distribution · View deposit amount breakdown

3. Volume decreases were concentrated in two US equity-index contracts

Between Oct 1–7 and Oct 25–31, average past-24-hour volume for established markets fell by USDC 1.173B. US Non-Financial 100 and US 500 together decreased by USDC 804.81M, accounting arithmetically for 68.6% of the net decrease in established markets, without establishing its cause.

Which contracts had the largest changes in volume?

Change in average past-24h volume (USDC M) US Non-Financial 100 (US100S) US 500 (US500S) Bitcoin (BTC) Ethereum (ETH) Gold (XAUS) Alphabet (GOOGL) BNB Monero (XMR) Semiconductor Bull 3X ETF (SOXL) Hyperliquid (HYPE) -600 -500 -400 -300 -200 -100 0 100 -566.2 -238.6 -106.9 -70.7 -60.1 +3.1 +3.3 +4.0 +5.7 +22.7 0
Unit: Change in average preceding-24-hour volume (USDC M) · Source: Variational official market API historical snapshots

Compare the averages of the 23:00 UTC past-24-hour volume readings for Oct 1–7 and Oct 25–31. The five largest increases and five largest decreases are shown. Negative means less, positive more; not a cumulative monthly change.

Evidence: View data · Calculation notes

ContractOct 1–7 → Oct 25–31Change
US Non-Financial 100 (US100S)USDC 1.169B → USDC 602.96MDecrease of USDC 566.20M
US 500 (US500S)USDC 370.88M → USDC 132.27MDecrease of USDC 238.61M

The five leading contracts accounted for roughly three quarters of observed volume (76.0%).

The same five contracts are ranked by their combined volume readings near 23:00 across 29 valid October dates. The denominator is the sum of all market readings on those dates, not exact monthly turnover.

Across the same two seven-day periods, average open interest in established markets also fell 16.5%. Contract details and daily market mix are available below.

Which contracts had the largest falls in open interest?

Which markets had the largest falls in open interest?

Change in average open interest (USDC M) Bitcoin (BTC) US Non-Financial 100 (US100S) Ethereum (ETH) Zcash (ZEC) US 500 (US500S) Taiwan Index (TWIS) BNB Strategy (STRC) Alphabet (GOOGL) Hyperliquid (HYPE) -30 -20 -10 0 10 -29.9 -21.9 -21.6 -11.2 -10.9 +1.3 +2.3 +2.6 +3.5 +5.5 0
Unit: Change in average open interest (USDC M) · Source: Variational official market API historical snapshots

Oct 25–31 versus Oct 1–7 UTC; average the seven 23:00 readings in each window. Negative means less outstanding contract value, not losses or withdrawn capital; 551 comparable existing markets.

Evidence: View data · Calculation notes

OI is the size of contracts still open. This chart shows changes in market readings, not losses or withdrawals; the calculation notes explain how this differs from the headline protocol total.

How much volume and open interest belong to each contract type?

The 18 identified Swap contracts accounted for 57.1% of observed volume. The denominator is the sum of all market volume readings near 23:00 on 29 valid October dates, not exact monthly turnover.

What share of volume and open interest came from identified Swaps?

% 0 20 40 60 80 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 Identified Swaps / all volume Identified Swaps / all OI
Unit: % · Source: Variational official market API historical snapshots

23:00 UTC, Oct 1–31; 29 valid days. The lines show identified Swaps’ share of all observed markets’ volume and, separately, open interest. Other products remain unclassified; Oct 13 and 16 are missing.

Evidence: View data · Calculation notes

The two lines show identified Swaps’ shares of that day’s total market volume and open interest. The remainder of each measure is unclassified, not another confirmed product category.

View markets first observed this month

The 30 markets first appearing in this dataset in October accounted for 6.6% of observed volume. The denominator is again the sum of all market volume readings near 23:00 on 29 valid October dates, not exact monthly turnover. First observation is not an official listing date; contract names and observation dates remain in the market data.

Is volume concentrated in a few contracts?

Was each day’s volume concentrated in a few markets?

% 0 20 40 60 80 100 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 Daily top 1 / all volume Daily top 5 / all volume Daily top 10 / all volume
Unit: % · Source: Variational official market API historical snapshots

23:00 UTC, Oct 1–31; 29 valid days. Rerank markets each day and divide the top 1/5/10 markets’ preceding-24-hour volume by all markets’ volume that day. Different from the period-wide top-five share of 76.0%.

Evidence: View data · Calculation notes

This chart reranks markets each day. Its daily top-five share is different from the 76.0% combined share of the five markets ranked over the whole observed period above.

4. The Treasury balance increased by about USDC 2.60M over the month

The protocol Treasury balance rose from USDC 9.59M at month-start to USDC 12.19M at month-end, a net increase of USDC 2.60M. A balance increase is not confirmed revenue.

How much did the Treasury balance increase?

Balance (USDC M) Daily change (USDC M) 0 2 4 6 8 10 12 14 0 0.03 0.06 0.09 0.12 0.15 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 10-01 10-03 10-05 10-07 10-09 10-11 10-13 10-15 10-17 10-19 10-21 10-23 10-25 10-27 10-29 10-31 Balance that day Increase/decrease from previous day
Unit: USDC M · Source: On-chain wallet balance snapshots

Oct 1–31 UTC; 29 valid daily changes. The upper chart shows that day’s balance; the lower chart shows the increase or decrease from the preceding day. Gaps are not zero, and balance growth is not confirmed revenue.

Evidence: View data · Calculation notes

5. US Non-Financial 100 and US 500 had relatively low typical quoted costs in this sample

US Non-Financial 100 and US 500 had lower typical quoted costs than other established Swap contracts observed in this report. Gold and copper show other cost levels; this is not an estimate for every market or a comparison with other venues.

The table shows estimated one-sided quote costs in USDC for USDC 100,000 of trade notional. That is the nominal trade size, not margin. “Typical quoted cost” means the median of valid samples.

ContractTypical Long costTypical Short cost
US Non-Financial 100 (US100S)2.552.55
US 500 (US500S)4.134.13
Gold (XAUS)7.177.17
Copper (COPPP)88.8388.85

Each number measures the price difference between the USDC 100,000 quote and the midpoint of simultaneous USDC 1,000 buy/sell quotes. It is not a total fee or realized execution cost and excludes funding costs while holding a position. Source: valid open-market request-for-quote observations during Oct 1–31.

View data and calculation notes

View all quoted costs

The trade notional is USDC 100,000; all costs below are in USDC. “90% of samples at or below” is the sample’s 90th percentile, not its maximum or a guarantee about future executions. Actual coverage dates and sample counts by contract are in the calculation notes.

ContractLong: typical costLong: 90% of samples at or belowShort: typical costShort: 90% of samples at or below
China Index (CNA50S) First observed this month30.7232.3630.7332.36
Cocoa (COCOAP) First observed this month175.24224.16175.64225.85
Copper (COPPP)88.83119.8288.85124.29
Dollar Index (DXIP) First observed this month11.8113.3411.8113.70
Gasoline (GASOLP) First observed this month78.1579.6778.2179.67
Heating Oil (HOILP) First observed this month39.7855.1039.7755.10
Taiwan Index (TWIS)11.2415.3711.3115.38
Brent Crude Oil (UKOILP)17.3018.7017.3118.62
US Non-Financial 100 (US100S)2.552.692.552.69
US 2000 (US2KS) First observed this month7.928.387.928.35
US 30 (US30S) First observed this month2.993.843.004.07
US MidCap (US400S) First observed this month34.8336.1234.8536.13
US 500 (US500S)4.134.184.134.18
FANG+ Index (USFNGS) First observed this month34.2748.3334.4548.98
WTI Crude Oil (USOILP)18.9822.6519.1522.92
Volatility Index (VXIP) First observed this month545.13589.47545.13589.43
Silver (XAGS)34.7840.9534.7240.53
Gold (XAUS)7.177.437.177.42
View chart comparison

Typical and higher quoted costs for established markets

One-sided quoted cost (USDC) USDC 100K trade notional US Non-Financial 100 (US100S) US 500 (US500S) Gold (XAUS) Taiwan Index (TWIS) Brent Crude Oil (UKOILP) WTI Crude Oil (USOILP) Silver (XAGS) Copper (COPPP) 0 30 60 90 120 150 Long typical Long 90% threshold Short typical Short 90% threshold
Unit: USDC (USDC 100K trade notional, one side) · Source: Historical request-for-quote prices

Valid open-market request-for-quote prices during Oct 1–31 for eight established Swaps: typical costs and 90% sample thresholds. See the data for sample counts and actual coverage.

Evidence: View data · Calculation notes

Quoted costs for newly observed markets

One-sided quoted cost (USDC) USDC 100K trade notional US 30 (US30S) US 2000 (US2KS) Dollar Index (DXIP) China Index (CNA50S) FANG+ Index (USFNGS) US MidCap (US400S) Heating Oil (HOILP) Gasoline (GASOLP) Cocoa (COCOAP) Volatility Index (VXIP) 0 100 200 300 400 500 600 Long typical Long 90% threshold Short typical Short 90% threshold
Unit: USDC (USDC 100K trade notional, one side) · Source: Historical request-for-quote prices

Ten Swaps first seen in this dataset during October, with different coverage starts. Request-for-quote prices only, excluding funding while a position is held.

Evidence: View data · Calculation notes

View quote sample coverage

Of 7,044 standard open-market hours, 6,467 were valid (91.8%); 210 were missing locally and 367 lacked a valid source quote. Closed or unknown-schedule hours are separate. No stale quotes were detected in the sample, which does not establish their absence throughout the history. These gaps are not collectively evidence that the platform could not be traded.

Last month’s questions, this month’s progress

September questionOctober observation
Open-interest trendOpen interest fell from its early-October high. September’s coverage and measures differ, so this is not a full-month growth comparison.
Large-trade quotesThis report adds quote-cost observations for USDC 100,000 of trade notional; within-month medians alone do not establish stability across months.
User fundsDeposit and funding-flow observations have been added; complete net inflows remain unverified.

Links refer to the corresponding sections of the September report in Chinese.

What to watch in November

  • Whether capital and open interest recover.
  • Whether volume in the two US equity-index contracts recovers.
  • Whether funding activity persists and deposit/withdrawal coverage can be completed.

Data, downloads and related research

Each chart links directly to its data and calculation notes. Unknown intervals remain blank; full fields, sampling rules and reconciliation are available for verification.

View fields, sampling and full limitations

Protocol data contain 736 of 744 hourly readings; daily charts show only 29 complete days. Market structure uses 29 sets of readings near 23:00, missing Oct 13 and 16. The within-October comparison fixes 551 markets with complete data in both seven-day periods; the cross-month comparison has 549 common markets. Reported two-sided OI and user-side market OI remain separate and are not joined by mechanically dividing by two.

Daily funding-flow aggregates retain counts, not pool or address sets. Inbound provenance is not independently verified, so flows are not labelled official Core OLP funding. Automated settlement and transfers are not treated as user-initiated trades. Complete withdrawals, platform-wide net inflows, classified revenue, historical per-pool balances and complete commissions remain unverified.

PerpDEX Independent Research · Amounts in USDC; all times UTC.This print edition includes core charts. Detailed methods and the financing appendix are available through the evidence links.