You deposit 100 ETH into a lending protocol as collateral and borrow against it. The system says your position is safe — well above the liquidation threshold. Then a sudden price drop pushes your collateral ratio below the minimum. The protocol automatically sells your ETH at a discount to repay your loan. That sale pushes the ETH price down further. Other borrowers who were also using ETH as collateral now fall below their thresholds. They get liquidated too. More ETH floods the market. The price drops again. The cycle accelerates.
This is a liquidation cascade — a self-reinforcing loop where each liquidation causes further price drops, triggering even more liquidations. Unlike a simple market correction, a cascade is amplified by the mechanical structure of overcollateralized lending. The larger the total borrowed amount and the tighter the collateral ratios, the more destructive the cascade.
BLUF: A liquidation cascade is a chain reaction where forced liquidations in DeFi lending protocols push asset prices down, which triggers further liquidations, creating a downward spiral. The mechanics are simple: when collateral value falls below a protocol’s liquidation threshold, the protocol automatically seizes and sells the collateral — often at a discount — to repay outstanding debt. If many borrowers share the same collateral asset, these forced sales flood the market, crashing the price further and pushing more borrowers into liquidation territory. The 2022 crypto crisis demonstrated this at scale: billions in cascading liquidations across protocols cascaded through the entire DeFi ecosystem. Detecting cascade risk on-chain involves monitoring health factors, total borrowed amounts per asset, and concentration of near-threshold positions. Defense requires maintaining high collateral ratios, avoiding correlated collateral, and understanding protocol-specific liquidation parameters.
What Is Liquidation in DeFi?
Liquidation is the process by which a lending protocol automatically repays a borrower’s debt when their collateral value falls below a required minimum. Protocols like Aave, Compound, and MakerDAO require borrowers to maintain an overcollateralization ratio — meaning the value of deposited collateral must always exceed the value borrowed.
For example, if you deposit $10,000 worth of ETH into Aave and the protocol sets a collateral factor of 80%, you can borrow up to $8,000. If the value of your ETH drops to $9,500, your loan-to-value ratio rises. Once it crosses the liquidation threshold, the protocol allows anyone — typically specialized liquidation bots — to repay your debt in exchange for a portion of your collateral plus a liquidation bonus.
Key Liquidation Parameters
| Parameter | What It Means | Typical Range |
|---|---|---|
| Collateral factor | Maximum % of collateral you can borrow | 50–85% |
| Liquidation threshold | Ratio at which liquidation is triggered | 70–85% |
| Liquidation bonus | Discount on collateral for liquidators | 5–15% |
| Close factor | Max % of a position that can be liquidated per transaction | 25–100% |
| Health factor | Ratio of collateral value to debt value (must stay > 1.0) | Target > 1.5 |
The health factor is the single most important metric. It is calculated as:
Health Factor = (Collateral Value × Liquidation Threshold) / Total DebtA health factor above 1.0 means the position is safe. At 1.0, the position can be liquidated. Below 1.0, liquidation is imminent or already underway.
How the Liquidation Process Works
When a borrower’s health factor drops below 1.0, anyone can call the protocol’s liquidation function. The liquidator repays part or all of the borrower’s debt (up to the close factor) and receives an equivalent amount of collateral plus the liquidation bonus. This bonus — typically 5–15% — is what incentivizes liquidators to monitor positions and act quickly.
The process is fully on-chain and transparent. Every liquidation is recorded in smart contract event logs, making it possible to trace exactly when, where, and how much was liquidated using a block explorer.
What Makes Liquidations Cascade
A single liquidation is a normal, healthy mechanism — it ensures protocols remain solvent by closing undercollateralized positions. The problem arises when liquidations cluster.
The Cascade Mechanism
Here is the step-by-step chain reaction:
- Trigger event — A market sell-off, oracle price update, or flash crash pushes an asset’s price down by 10–20%.
- First wave of liquidations — Borrowers near their thresholds get liquidated. Liquidators seize their collateral and sell it on decentralized exchanges.
- Price impact — The forced selling on DEXs pushes the asset price down further due to limited liquidity pool depth.
- Second wave — Borrowers who were previously safe now fall below their thresholds. More liquidations occur.
- Acceleration — Each wave of selling pushes the price down more, triggering the next wave faster. The process accelerates until either the selling pressure exhausts itself or external buyers step in.
The key amplifier is that liquidators must sell the seized collateral to realize their profit. This selling happens on the same DEX pools where the price is already falling, compounding the downward pressure.
Why Correlated Collateral Makes It Worse
Most DeFi borrowers use the same few assets as collateral: ETH, WBTC, and major stablecoins. When ETH drops, it affects every borrower using ETH collateral simultaneously. But the problem is even worse when borrowers use correlated assets — for example, using stETH (staked ETH) as collateral to borrow ETH. If stETH depegs from ETH even slightly, the position’s health factor deteriorates on both sides: the collateral loses value while the debt stays constant.
This is exactly what happened during the stETH depeg crisis in June 2022, when stETH traded as low as 0.93 ETH, triggering cascading liquidations across multiple lending protocols.
Historical Cascade Events
The 2022 DeFi Crisis (Celsius, Three Arrows Capital)
The most destructive liquidation cascade in DeFi history occurred during the summer of 2022. It followed a multi-chain chain reaction:
| Event | Date | Trigger | Cascade Effect |
|---|---|---|---|
| Terra/Luna collapse | May 2022 | UST algorithmic stablecoin depeg | $40B wiped out, triggering panic |
| stETH depeg | June 2022 | Luna aftermath + Celsius selling pressure | stETH fell to 0.93 ETH |
| Celsius withdrawals frozen | June 2022 | Insolvency from stETH positions | $12B in user funds frozen |
| 3AC liquidation | June 2022 | Overleveraged cross-protocol borrowing | $10B+ in defaults across lenders |
The cascade started with a single failure — Terra’s UST losing its peg — and rippled through the ecosystem because multiple protocols and funds shared exposure to the same assets. Celsius had parked hundreds of millions in stETH, using it as collateral to borrow more ETH. When stETH depegged, their positions became undercollateralized, forcing liquidations that pushed stETH even lower.
The Aave v2 Liquidation Event (November 2022)
In November 2022, an attacker used a flash loan to manipulate the CRV token price on decentralized exchanges, then targeted a large CRV-collateralized position on Aave v2. By artificially depressing the CRV price, the attacker triggered a massive liquidation of Avi Eisenberg’s position — approximately $60 million in bad debt for the protocol.
This event demonstrated that liquidation cascades can be intentionally triggered through price manipulation, not just by organic market movements.
The March 2020 Black Thursday (MakerDAO)
On March 12, 2020, Bitcoin and ETH crashed by over 50% in 48 hours. MakerDAO’s system allowed liquidators to bid on seized collateral using a Dutch auction model. But when gas prices spiked to hundreds of gwei, many liquidators could not submit transactions in time. A few opportunistic liquidators won collateral auctions with bids of zero — paying nothing for $4.5 million worth of ETH. This “zero-bid” liquidation bug resulted in the protocol’s first-ever debt shortfall.
Detecting Cascade Risk On-Chain
Understanding liquidation cascade risk requires analyzing several on-chain metrics. Here is what to monitor.
1. Protocol-Wide Health Distribution
The most important metric is the distribution of health factors across all borrowers in a lending protocol. If a large number of positions have health factors between 1.0 and 1.2, even a small price move can trigger mass liquidations.
You can query this data using Dune Analytics or by parsing liquidation-related events directly from the protocol’s smart contracts:
-- Example Dune query: Positions near liquidation threshold
SELECT
borrower,
collateral_asset,
total_collateral_usd,
total_debt_usd,
health_factor
FROM aave_v3_positions
WHERE health_factor BETWEEN 1.0 AND 1.3
ORDER BY total_debt_usd DESC2. Total Borrowed Amount per Asset
If $500 million is borrowed against ETH collateral, a 15% ETH price drop could trigger liquidations on tens of millions in positions. Check the protocol’s TVL breakdown by asset to understand exposure.
3. Collateral Concentration
A protocol where 70% of collateral is in a single volatile asset is far more cascade-prone than one where collateral is diversified across stablecoins, ETH, and BTC.
4. Oracle Freshness and Type
Protocols using TWAP (time-weighted average price) oracles are more resistant to flash-crash-triggered cascades because TWAP smooths out short-term price spikes. Protocols using spot price feeds from centralized exchanges can trigger liquidations more aggressively during momentary wicks.
5. Liquidation Queue Depth
Some newer protocols (like Aave v3’s efficiency mode) allow you to see pending liquidation activity. A growing queue signals cascade risk.
On-Chain Indicators of an Active Cascade
When a cascade is actively occurring, you can observe these patterns on-chain:
| Indicator | What You See | Where to Check |
|---|---|---|
| Spike in liquidation events | Hundreds of LiquidationCall events per block | Block explorer, Dune |
| Plummeting health factors | Average health factor drops rapidly | Protocol dashboard |
| DEX volume spike | Unusual trading volume on collateral/asset pairs | DEX analytics |
| Gas price surge | Liquidation bots bidding up gas | Mempool monitor |
| Price divergence | Asset trades below prices on other venues | Cross-exchange comparison |
| Stablecoin outflows | Users withdrawing stablecoins from the protocol | Protocol TVL tracker |
How to Protect Yourself
For Borrowers
- Maintain a high health factor — Keep your health factor above 1.5 at minimum. A 33% price drop brings a 1.5 health factor to liquidation territory.
- Diversify collateral — Do not use a single volatile asset as your only collateral. Mix in stablecoins to create a buffer.
- Avoid correlated collateral — Using stETH to borrow ETH means a depeg hurts you on both sides. Borrow a different asset.
- Monitor oracle type — Understand which oracles your protocol uses and how quickly they update during volatile periods.
- Set up alerts — Use on-chain monitoring tools to get notified when your health factor drops below a threshold.
For Liquidity Providers and Traders
- Watch cascade indicators — If you see liquidation volumes spiking for an asset, expect continued downward pressure.
- Avoid providing liquidity during active cascades — Impermanent loss amplifies during cascade events.
- Use limit orders instead of market orders — If you must trade during a cascade, use limit orders to avoid slippage from the cascade-driven price impact.
Limitations of On-Chain Liquidation Analysis
On-chain analysis has blind spots when it comes to liquidation cascades:
- Cross-protocol exposure — A borrower with positions on three protocols may appear healthy on each individually, but their aggregate risk is much higher. On-chain data does not always link positions across protocols.
- Off-chain leverage — Funds borrowing on DeFi protocols and then using centralized exchanges for additional leverage are invisible on-chain until the cascade hits.
- OTC and private arrangements — Over-the-counter lending and private borrowing arrangements between institutions (as seen with 3AC) are not visible on-chain.
- Timing uncertainty — You can see who is near liquidation, but predicting the exact trigger event (what will cause the initial price drop) is fundamentally unpredictable.
Related Topics
- Learn how flash loan attacks can artificially trigger liquidation cascades by manipulating collateral prices.
- Understand oracle manipulation — a common attack vector that can deliberately trigger liquidations.
- Read about stablecoin depegging — a frequent catalyst for liquidation cascades when stablecoin collateral loses its peg.
- Explore our DeFi protocol red flags guide to identify protocols with poor liquidation mechanisms.
- Use our address risk scoring tool to analyze addresses involved in liquidation events.
This article is for educational purposes only and does not constitute financial advice. On-chain analysis can help you understand risks but cannot predict market movements. Always do your own research before participating in DeFi protocols.