Perp DEX funding stability, 24h stddev of ETH funding ranked
24h standard deviation of ETH funding rate, in basis points per 24h, across the perp-funding cohort. Lower means funding stays in a tight band, the carry trader's preferred signal.
TL;DR. As of , Vest posts the lowest 24h ETH funding stddev at 0.00000% (24h avg) across 17 ranked venues. Source: OpenChainBench, https://openchainbench.com/benchmarks/perp-funding-stability.
Read this carefully
Ranked only with 24 hours of funding history (288 samples) and a stddev above zero: a young listing, or a venue whose rate has not moved since listing (Ondo Perps at launch, Polymarket in July), reads artificially stable and shows as Provisional, never as the leader. Stability is not directionality: a persistently positive but tight band ranks above a rate that flips every settlement.
This benchmark ranks perp venues by how tightly their ETH funding rate clusters over the trailing 7 days. Average funding tells you which side gets paid; stability tells you how reliably. A venue whose 24h hold cost wanders 15 bps in a week is harder to carry trade than one that holds within 3 bps even if both average the same number. The bench reads the same normalized series as perp-funding (per-venue hold cost in bps per 24h) and exposes stddev_over_time of that series on a 24 hour window for ETH. Cohort is 19 venues across two sources: 8 polled directly by the harness (Hyperliquid onchain, Bybit, dYdX v4 Cosmos appchain, Binance, OKX, Paradex Starknet L2, Polymarket, Aster BNB Chain) and 11 sourced through Mobula's CEFI funding aggregator (Ondo Perps, Kalshi, Vest, StandX, ApeX Omni, Bitget, Gate, Kraken, KuCoin, Lighter, MEXC). Lower is better.
Methodology
The bench reuses the normalized 24h ETH funding series from perp-funding (bench № 036). For each venue it publishes the 7 day rolling standard deviation of the ETH hold-24h gauge, in bps. Sign is preserved on the underlying series; the ranking metric is the magnitude of the swing, not its direction. Lower means the venue's funding stayed in a tight band, the cleanest signal for funding-rate arbitrage and basis trades. Cadence and freshness inherit from the upstream funding scrape (60 second poll, 30 second Prom scrape).
- A stddev of zero is a property of the window, not of the venue. Most of the cohort quotes the same baseline when open interest is balanced, 0.01 percent per 8 hours, which normalizes to 3.00 bps per 24h hold: on 2026-10-21 nine of the twenty sat exactly there on ETH. Any venue holding the baseline for a full 24 hours reads zero and goes unranked, whatever its age. All of them move on a longer window, Polymarket between -0.26 and 3.00 bps over 30 days.
- Source. Same upstream as perp-funding: each venue's public funding endpoint polled every 60 seconds, normalized to bps per hour and per 24h hold, sign preserved.
- Metric. stddev_over_time of the 24h hold cost on ETH over a 24 hour window, per venue, on whichever of the two funding metrics that venue is read from. A day is short enough that one settlement regime dominates, so the figure answers what the band looked like today rather than this week.
- Sign handling. The underlying series is signed (positive means longs pay). Stability measures the swing, so the ranking uses the unsigned stddev. Average direction and level are the perp-funding bench.
- Cohort. Nineteen venues, two sources. 8 polled directly (perp_funding_hold_24h_bps): Hyperliquid, Bybit, dYdX v4, Binance, OKX, Paradex, Polymarket, Aster. 11 via the Mobula CEFI aggregator (perp_venue_funding_24h_bps): Ondo Perps, Kalshi, Vest, StandX, ApeX Omni, Bitget, Gate, Kraken, KuCoin, Lighter, MEXC. Both normalize to bps per 24h hold, sign preserved. Deribit, edgeX, Extended, Pacifica and Lighter RH are ranked on perp-funding-cost-30d and are not on this bench.
- Success. Same freshness signal as perp-funding: perp_funding_venue_ok_unix advancing within the last 5 minutes. A venue whose scrape stalls is tagged stale on the row.
- Cadence. The upstream funding harness polls every 60 seconds; Prometheus scrapes every 30 seconds. The 7 day stddev refreshes on every scrape.
- Failures. A venue whose upstream funding feed errors keeps its last gauge value; the leaderboard tags the row stale once the success gauge drops to 0.
- Reproducibility. The harness emits perp_funding_hold_24h_bps with labels venue and asset; the bench query is the stddev over 7d on that series with asset=ETH pinned.
Results: 19 venues measured, 17 ranked by 24h ETH funding stddev (24h avg)
| № | Venues | p50 | Success | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Vestzk perp exchange, hourly funding | 0.00000% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 2 | HyperliquidOnchain order book perps, funding settled hourly | 0.0011% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 3 | ParadexStarknet L2 perps, 8h funding accrued continuously | 0.0040% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 4 | KuCoinCEX, USDT perpetuals via Mobula aggregator, 8h funding | 0.0044% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 5 | LighterLighter L2 order book, hourly funding read from /funding-rates | 0.0047% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 6 | BitgetCEX, USDT perpetuals via Mobula aggregator, 8h funding | 0.0048% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 7 | AsterBNB Chain perps DEX, Binance compatible API, 8h funding | 0.0061% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 8 | GateCEX, USDT perpetuals via Mobula aggregator, 8h funding | 0.0062% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 9 | BinanceLargest CEX perp book, 8h funding (4h on some pairs) | 0.0067% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 10 | MEXCCEX, USDT perpetuals via Mobula aggregator, 8h funding | 0.0084% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 11 | OKXUSDT swaps, interval derived from settlement timestamps | 0.0095% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 12 | StandXDUSD-margined perps on BNB Chain, hourly funding | 0.010% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 13 | BybitUSDT linear perps, funding interval per instrument | 0.015% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 14 | ApeX OmniOmnichain orderbook perps, hourly funding | 0.016% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 15 | KalshiCFTC-regulated US exchange, 8 h funding | 0.022% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 16 | KrakenRegulated CEX, perpetuals via Mobula aggregator | 0.026% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 17 | dYdX v4Cosmos appchain order book perps, funding settled hourly | 0.058% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Measured, not ranked: sample below the ranking floor | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| not ranked | PolymarketPolygon perps by Polymarket, 1h funding, public info API | 0% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| not ranked | Ondo PerpsSGX off-chain matching, custody on Ethereum and Arbitrum, hourly funding read from the venue API | 0% | 100.00% (n=2880) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Frequently asked
What does funding stability measure?
It is the 24 hour standard deviation of each venue's ETH funding rate, expressed in basis points per 24 hour hold. The metric measures how tightly funding clusters around its own 7 day mean, ignoring whether the rate sits positive or negative. A low value means traders carrying a position on that venue paid (or received) a predictable number every settlement; a high value means the rate swung a lot over the week.
Why is lower better?
Carry trades, basis trades and funding-rate arbitrage all assume the rate behaves predictably over the holding window. A venue with a noisy funding curve forces wider stop-losses and reduces the achievable Sharpe even when the average rate looks attractive. Stability is the second-order property that turns an average rate into a trade you can size.
How does this differ from perp-funding (bench 036)?
perp-funding ranks venues by the current normalized 24h funding cost on ETH; it answers 'who is cheapest to hold right now'. perp-funding-stability ranks venues by the 24 hour standard deviation of that same series; it answers 'whose rate is most predictable'. The two are complementary: a cheap-but-noisy venue is still risky to carry; a stable-but-expensive venue is honest about its cost.
Why ETH only?
ETH is the deepest contract on every venue in the cohort, so the funding series has the strongest signal-to-noise ratio for cross-venue comparison. BTC and SOL are also published by the upstream harness as gauges; future revisions of this bench may expand the dimension.
How fresh are the numbers?
The upstream funding gauge is scraped every 30 seconds. The stddev refreshes on every Prom evaluation, so the leaderboard reflects the most recent 24 hour window minus at most 30 seconds of lag.
Source code github.com/ChainBench/OpenChainBench/tree/main/harnesses/perp-cohort-stats