Most stable USD-pegged stablecoin
Absolute deviation from $1.00 in basis points, computed from a liquidity-weighted median across USD-quoted venues.
This benchmark answers the question that decides whether a stablecoin is actually $1 today. how far has the per-minute aggregated price drifted from $1.00, on which venue, and for how long. Most "stablecoin tracker" sites display CoinGecko or Coinmarketcap averages refreshed every few minutes. That smooths out the venue-level discrepancies that actually matter, the USDC/USDT cross-venue gap on Binance when USDC briefly trades 25 bps below its Kraken quote, the USDe flash to $0.65 the night of October 10 2025, the DAI Curve 3pool reverse swap rate diverging from forward. We poll Binance, Kraken, Bitstamp (CEX REST tickers, 5 second cadence) and the Curve 3pool on-chain (12 second cadence, matches Ethereum block time), aggregate per-venue prices into a per-minute liquidity-weighted median per USD-quoted stable, and surface five live numbers per stablecoin: instantaneous deviation in basis points, cross-venue gap (max, min across venues), time outside the ±50 bps band over 24 h, time below peg vs time above peg (Circle redeems above-peg only, so the split tells you which side is structurally rare), and a depeg event flag set when the stable has been outside [$0.97, $1.03] for ≥5 consecutive minutes.
Methodology
We measure live stablecoin peg deviation by pulling mid-price ticks from every venue that exposes a USD-quoted pair for a given stable, then aggregating per-minute into a liquidity-weighted median. Deviation is `|aggregated_price, 1.00|` in basis points, recorded as both a gauge (latest) and a histogram (per-minute samples). p99 over 24 h via `quantile_over_time` is the headline leaderboard number. USDT-quoted pairs (Binance USDC/USDT, FDUSD, USDe) are deliberately excluded from the primary ranking because USDT's own peg deviation would contaminate every other stable's measurement; they are exposed on a separate `peg_deviation_usdt_anchored` metric instead. DAI's CEX coverage is essentially dead in 2026 (Binance zero book, Kraken thin, Bitstamp dead), so the only honest signal is on-chain via Curve 3pool's `get_dy`. We poll the pool both forward (USDC → DAI) and reverse (DAI → USDC) to surface directional asymmetry. A binary `peg_depeg_event_flag` goes high when a stable spends ≥5 consecutive minutes outside [$0.97, $1.03] and clears after 30 minutes back inside; conservative so it does not flap on normal stress.
Frequently asked
Which stablecoin is the most stable right now?
DAI currently leads the leaderboard at 0.019% (p99 deviation, 24 h) across 3 measured stablecoins on USD-quoted venues. The leaderboard re-sorts every minute against fresh Prometheus samples, so the answer reflects the last 24 hours of measured per-minute aggregated price, not a static white-paper figure.
What is stablecoin peg deviation in basis points?
Peg deviation is the absolute distance between a stablecoin's price and $1.00, expressed in basis points where 1 bp = 0.0001 = 0.01%. A stablecoin trading at $0.9978 has 22 bps of deviation; at $1.0050 it has 50 bps. The industry uses bps because the deviations are small. tens of bps is normal noise, hundreds of bps is stress, thousands of bps is a depeg event. The benchmark records both per-minute deviation and the worst-1% tail (p99) over 24 h, the integration-grade number for any product that holds inventory in the stable.
How does USDC compare to USDT on stability?
USDC returns 0.025% (p99 deviation, 24 h) on USD-quoted venues (Kraken, Bitstamp). USDT returns 0.12% on USD-quoted venues (Kraken, Bitstamp). USDC's peg is anchored by Circle's same-day USD redemption with US banks, USDT's by Tether's larger but slower-clearing redemption window. The leaderboard surfaces the live gap and the cross-venue spread per minute so the comparison is grounded in measured data rather than a sentiment-driven preference.
Did USDe really depeg to $0.65 in October 2025?
Yes, on Binance USDEUSDT specifically, the night of October 10 2025. Ethena's USDe is anchored by a delta-neutral basis trade rather than fiat reserves, and a sharp liquidation cascade on perp positions briefly cleared the orderbook on Binance well below peg. The benchmark exposes this kind of event live via the `peg_depeg_event_flag` (set when a stable has been outside [$0.97, $1.03] for ≥5 consecutive minutes) and via the USDT-anchored secondary metric. USDe's USD-anchored primary deviation cannot be measured because no major USD venue lists a deep USDe/USD pair; the USDT-anchored value is the honest live signal.
How is the stablecoin peg measured on OpenChainBench?
We poll Binance, Kraken and Bitstamp REST tickers every 5 seconds for every (stable, venue) pair, and the Curve 3pool `get_dy` on-chain every 12 seconds (matches Ethereum block time). Per-venue samples are grouped into 60-second buckets and aggregated into a liquidity-weighted median per stable (weight = 24h USD volume). Deviation is `|aggregated_price, 1.00|` in basis points, recorded as both a gauge and a histogram. p99 over 24 h via Prometheus `quantile_over_time` is the headline. USDT-quoted pairs are exposed on a separate secondary metric so USDT's own peg deviation does not contaminate the USD-anchored primary.
Why ignore CoinGecko or CoinMarketCap prices?
Aggregator prices (CoinGecko, CoinMarketCap, DefiLlama) are themselves liquidity-weighted medians of the venues this benchmark already polls directly, refreshed every 1-5 minutes. Re-aggregating an aggregator adds latency, hides the venue-level cross-spread that actually matters during stress (USDC briefly $0.9978 on Kraken while $1.0003 on Coinbase), and inherits whatever venue selection the aggregator chose. The benchmark goes to the source venues directly so the cross-venue gap is visible and the depeg event flag fires within 5 minutes of the price actually breaking the band.
Source code github.com/OpenChainBench/OpenChainBench/tree/main/harnesses/stablecoin-peg
