DeFi lending losses about $3 per $10,000 on EVM, Solana

Keyring Network founder Alex McFarlane calculated that lenders on Ethereum Virtual Machine chains and Solana lost roughly $3 for every $10,000 deposited in DeFi lending markets over the trailing 12 months. McFarlane derived the figure from DefiLlama records on May 17 after isolating lending markets and excluding bridge incidents. His tally runs through May 16.

McFarlane measured $30.9 million in gross non-bridge lending exploits against an average total value locked of $99.6 billion, a rate that equates to about 3.1 basis points gross and roughly 3 basis points net after recoveries. He wrote that spreading $10,000 across the largest EVM and Solana lending markets produced an annualized hack-loss expectation of about $3 over the past year. The calculation excludes bridge risk, oracle failures and protocol-specific bugs, and assumes a deposit did not fall inside a market that suffered a tail event.

DefiLlama full-history records show $7.75 billion in gross losses across DeFi and $4.52 billion when bridge incidents are removed. Crypto thefts totaled $606 million in April, the largest month since a 2025 exchange breach; Kelp DAO and Drift incidents accounted for about 95% of April losses.

The loss data are skewed: a small number of large events make up most cumulative damage while most incidents are small. On a logarithmic scale the pattern approximates a lognormal distribution. Most exploits targeted a single component inside a market rather than draining an entire protocol, and larger markets tended to absorb a smaller percentage hit when incidents occurred.

Recoveries reduced headline losses. Across all DeFi protocol losses tracked by DefiLlama, capped recoveries amounted to about 8% of gross damage. For EVM and Solana lending excluding bridges, recoveries rose to roughly 20%. Euler Finance is a noted example, after a 2023 flash-loan exploit the attacker returned all stolen funds.

On code and design, Morpho contributor Merlin Egalite argued that ‘minimalism is the dividing line between safe and unsafe lending markets.’ Several lending projects continue to draw the bulk of new capital.

McFarlane included a risk comparison, writing that a probability of 3 in 10,000 is approximately equal to the rate of Americans who die from slipping and falling.

The $3 per $10,000 figure reflects realized, historical losses over the past year. It excludes bridge-related incidents and other concentrated tail risks and does not project future results.