Discrete Batching vs. Continuous Liquidation Triggers

Maximizes liquidation yield: Aggregates multiple underwater positions into a single transaction. This reduces per-liquidation overhead, allowing keepers to profitably liquidate positions with smaller health deficits (e.g., 1-2%). Ideal for protocols like MakerDAO's auction system, where maximizing collateral recovery is paramount.

Controlled execution windows: Liquidations occur at scheduled intervals (e.g., every hour). This allows protocols to batch transactions, smoothing out network demand and avoiding gas wars during volatility. Used by Compound v2 and Aave v2 to provide predictable costs and reduce failed transactions from front-running.

Sub-second risk mitigation: Positions are liquidated the instant they breach the safety threshold via on-chain or oracle-based triggers. This near-real-time action, as seen in dYdX (order book) and Aave v3's e-mode, virtually eliminates the accumulation of undercollateralized debt, crucial for high-leverage or volatile markets.

Eliminates coordination complexity: No need for sophisticated batching logic or timing coordination among keepers. This lowers the barrier to entry for keeper networks, increasing decentralization and resilience. Protocols like Synthetix and Euler (pre-hack) used this model to ensure liveness from a broader set of participants.

Maximizes liquidation proceeds: Liquidations are queued and executed in discrete, periodic batches (e.g., every 8 hours). This allows for Dutch auctions (e.g., MakerDAO's Flop auctions) or batch auctions to discover optimal clearing prices, potentially yielding better recovery rates for the protocol. This matters for protocols with large, illiquid collateral positions where price impact is a primary concern.

Controllable network load: By scheduling liquidation events, protocols like Compound v2 and early Aave versions can offload work to keeper bots during off-peak times, avoiding gas wars and unpredictable spikes during market volatility. This simplifies infrastructure planning for keepers and reduces the risk of failed transactions due to sudden base fee surges on L1s.

Near real-time solvency protection: Positions are monitored and can be liquidated the moment they fall below the health threshold (e.g., Aave v3, Compound v3). This minimizes the protocol's exposure to undercollateralized debt, especially critical for volatile assets or during rapid market downturns. This matters for maximizing the safety of user deposits and maintaining protocol solvency.

Higher frequency opportunities: Continuous systems create a constant stream of potential liquidation transactions. This enables sophisticated keeper networks (e.g., using Chainlink Automation, Gelato) to run optimized MEV strategies, improving the reliability and competitiveness of the liquidation market. Higher keeper profits lead to more robust and responsive defense for the protocol.

Vulnerability to price gaps: The batch interval (e.g., 8 hours) creates a window where an underwater position cannot be touched. A sudden price drop just after a batch closes leaves the protocol exposed until the next batch. This matters for protocols with highly correlated or volatile collateral, increasing the risk of bad debt accumulation.

Persistent gas competition: Every block becomes a potential liquidation opportunity, leading to persistent gas auctions among keepers. This can erode keeper margins through MEV extraction and increase network congestion. The cost is ultimately borne by the protocol via larger liquidation penalties or passed to users. This matters for cost-sensitive applications on high-fee networks.