The recent incident on Oct. 10 marked the largest liquidation event ever recorded in the crypto sector, with over $19 billion getting liquidated, as per CoinGlass data. This led to a $65 billion drop in open interest. This event overshadows prior liquidation instances, like the COVID-19 crash, which saw $1.2 billion, and the FTX collapse with $1.6 billion in liquidations.
Following the event, investigators reached a preliminary consensus that it was partly triggered by weak pricing oracles on the Binance platform. The collateral values for three pegged crypto tokens—USDE, bnSOL, and wBETH—were derived from Binance’s internal orderbook rather than an external oracle. This exposes users of the “Unified Accounts” feature to liquidation risks during market volatility.
While it’s possible this vulnerability was exploited in a planned attack on Oct. 10, the evidence is not definitive. The token USDE notably played a role in the liquidation wave, racking up around $346 million in volume, compared to wBETH at $169 million and bnSOL at $77 million. The massive withdrawal of buy-side liquidity in a stablecoin pair is especially concerning.
Utilizing detailed data obtained from our partners at the AI-oriented market analytics startup Rena Labs, Cointelegraph Research analyzes the unusual maneuvers in the USDE/USDT trading pair in this article.
A Mass Liquidity Meltdown
Rena’s anomaly detection system identified one of the most dramatic and intricate market disruptions ever witnessed in stablecoin trading. This is unexpected, given there were no worries about the integrity of USDE’s collateral, unlike prior issues with UST and USDC. Mints and redemptions of USDE kept functioning normally. Nevertheless, professional market makers dramatically withdrew liquidity from the pair. Some of this withdrawal can be credited to automated risk-assessment systems, which activated defensive quote withdrawals to mitigate exposure.
Before the market fall, the average total liquidity of USDE was at $89 million, with a balanced mix of buy and sell orders. Between 21:40 and 21:55 UTC, the pair’s liquidity on Binance plummeted by nearly 74%, dropping to around $23 million. By approximately 21:54, market depth had virtually vanished, with total liquidity dwindling to a mere $2 million, and market-making activity nearly ceased. As a consequence, bid-ask spreads skyrocketed to 22%.
The market lost its structural stability during the crash. Trading volume surged by a staggering 896 times as the ask-side depth fell by 99%. This imbalance drove USDE’s price down to $0.68 on Binance’s spot market, while it remained close to its peg on other exchanges.
During the 10-minute crisis, trade intensity heightened nearly 16 times relative to the standard rate of 108 trades per minute, peaking at almost 3000 trades per minute, with 92% of those being sell orders. Many transactions were likely caused by panic selling, stop-loss activations, and forced liquidations.
Evidence of Anomalous Market Activity
However, Rena’s anomaly detection system identified unusual activity well ahead of the USDE liquidity crisis. Around 21:00 UTC, it detected 28 anomalies, a rate four times higher than during the prior hour. These anomalies comprised unexpected spikes in trading volume, prices, or trade intensity, alongside suspicious patterns such as bursts, clusters, and sequences of trades. It also included recognizing activity typical of various types of order spoofing.
Three distinct surges of large orders occurred just before the crisis within the order book’s size profile. These transactions took place when BTC had started its downturn on major exchanges but prior to USDE initiating a liquidity squeeze.
This event underscores the fragility and leverage still prevalent in the crypto market, where cascading liquidations can erase what appear to be secure trades. Similar to the 99% drawdowns observed in several altcoins during the crash, the USDE depeg illustrates that the market for many tokens lacks sufficient organic demand to sustain it. Without significant market makers like Wintermute, the orderbooks of numerous crypto assets have shown limited resilience.
This article does not provide investment advice or endorsements. Every trading or investment decision carries risk, and readers are encouraged to do their own research before proceeding.
This article serves only for informational purposes and should not be interpreted as legal or investment guidance. The opinions expressed here are solely those of the author and do not necessarily reflect the views of Cointelegraph.
Cointelegraph does not endorse the content of this article or any products mentioned. Readers should conduct their own research before taking any actions related to products or companies mentioned and are fully responsible for their decisions.
