A year ago I parked $20,000 across several DeFi protocols and asked myself a question I had been dodging: if one of these gets drained tonight, do I actually have a plan? I had read the crash weekend playbook, I knew my wallet security checklist, but insurance felt like the one risk layer I kept skipping. So I joined Nexus Mutual, bought cover on the positions I could not easily split, and let it run. This is the ledger.
The short version up front: I spent about $290 in premiums over twelve months, never filed a claim, and would renew — but only on a portion of the book. DeFi insurance is not a regulated policy; it is a discretionary mutual whose members decide whether you get paid. That distinction changes everything about how you should use it. Education, not financial advice; your losses are still yours to model.
TL;DR. On a $20,000 DeFi book I paid roughly $290 in Nexus Mutual premiums over a year (blended ~1.45%). Blue-chip cover runs 1–2% annually, riskier protocols 5–10%. Nexus Mutual's ~$115M pool has paid about $18.5M in claims across 184+ protocols, with a ~70% approval rate and a 72-hour claims committee. Payouts include ~$3.4M on Euler, ~$250K on Arcadia, ~$95K on Stream Finance. InsurAce, Sherlock and Neptune wound down cover. I would renew, but only on concentrated positions I cannot diversify away — diversification is free and does more.
Why I bought DeFi insurance (and what I was actually afraid of)
The honest fear was not a market crash. I had sized positions for drawdowns and used the portfolio allocator to keep exposure bounded. The fear was a single contract draining overnight — the kind of event where the UI still loads, your wallet still connects, and the balance is zero. That is the risk insurance is built for, and it is the risk diversification alone cannot fully remove when you want yield on a specific protocol.
I also wanted to know whether the product actually works. Nexus Mutual had paid real claims — Euler's ~$3.4M payout, Arcadia's ~$250K, Stream Finance's ~$95K — but I had never run a policy myself. Buying cover is the only way to feel the friction: capacity limits, premium volatility, waiting periods, wording. I treated the $290 as a research budget as much as risk transfer.
What I was not insuring against: my own mistakes. A phishing approval, a leaked seed phrase, a bad bridge call — none of that is covered, and none of it should be. Cover exists for contract failures on listed protocols, not for user error. Run the yield risk grader before you buy; the grader surfaces the risks that insurance cannot fix.
The DeFi insurance landscape in 2026: who is left standing
The market shrank. In 2022–2023 you could pick from Nexus Mutual, InsurAce, Sherlock, Neptune Mutual and smaller mutuals. By late 2026 most wound down cover operations, leaving Nexus Mutual as the dominant on-chain discretionary mutual plus a few recovery-focused products. InsurAce paused new cover, Sherlock pivoted from its coverage pool, and Neptune wound down its markets — each a flavor of the same problem: pricing tail risk on permissionless protocols is hard, and capital is expensive.
The table below is the landscape as I found it when I shopped for cover in October 2026. TVL and pool figures are rounded from public dashboards and protocol reporting; claims models describe how a payout actually gets decided.
| Provider | Status & pool size | Claims model | Notes |
|---|---|---|---|
| Nexus Mutual | Active. ~$115M capital pool; 184+ listed protocols; ~$18.5M total claims paid. | Discretionary mutual: claims committee reviews within 72h; members vote. ~70% approval rate historically. | The main venue for smart-contract and depeg cover. Premiums 1–2% blue-chip, 5–10% riskier. |
| InsurAce | Defunct / wound down new cover writing. | Was a multi-chain cover protocol with bundled cover. | Paused new cover issuance; existing claims handled on a best-effort basis. Not a viable 2026 buy. |
| Sherlock | Pivoted away from retail cover pool. | Was a staking-backed mutual with audit firms as claim assessors. | Focus shifted to protocol-level audits and reinsurance-style coverage; retail cover largely closed. |
| Neptune Mutual | Wound down main markets. | Parametric-style cover pools with dedicated incident reporting. | Cover pools closed; marketplace no longer underwrites new policies in 2026. |
| Resolv (recovery) | Active. Recovery, not insurance. | Legal recovery infrastructure for hacked protocols. | Helps trace and reclaim stolen funds; complements cover but does not replace a payout. |
The takeaway is sobering: when three of five options stop writing cover, the remaining one prices capacity however it wants. Nexus Mutual is not cheap on riskier protocols because no competition forces it to be. That is why I only insured blue-chip and large-concentration positions and left smaller, newer protocols uninsured — the premium exceeded my expected loss on those.
My year of cover: the $20,000 ledger
Here is what I actually bought. I split $20,000 across four positions, insured two of them, and let the others ride uninsured because the premium did not justify the risk. All premiums are annualized rates applied to the covered amount for the period I held the cover. The numbers are mine, real, and slightly humbling.
| Position | Amount | Covered? | Premium rate | Premium paid |
|---|---|---|---|---|
| Aave v3 Ethereum, USDC supply | $6,000 | Yes, smart-contract cover | 1.30% / yr | $78.00 |
| Lido staked ETH (stETH) | $5,000 | Yes, smart-contract cover | 1.80% / yr | $90.00 |
| Morpho vault, USDC, Base | $4,000 | No — curator risk, uninsured | — | $0.00 |
| Ethena USDe, stable yield | $5,000 | Partial — depeg cover only | 2.44% / yr (on $5,000) | $122.00 |
| Total book | $20,000 | $16,000 of $20,000 covered | ~1.45% blended | $290.00 |
Two decisions are worth explaining. The Morpho vault I left uninsured on purpose: the risk there is curator behavior and curator risk, not the base contract, and Nexus Mutual's wording would not have paid out on a curator decision. For the Aave position I bought smart-contract cover — cheap at 1.30% and the venue I trust most after the Aave vs Morpho comparison. For Ethena I bought depeg cover rather than smart-contract cover, because the failure mode I feared was the USDe peg breaking, not the contract being drained. Depeg cover only pays after the asset breaches a threshold — roughly 10% below peg — so it is a tail product, not price insurance.
The blended premium worked out to about 1.45% of my total book, or $290 for the year. That is roughly 7 days of yield on a 7.5% average APY. Cheap enough that I did not notice it, large enough that I wanted a claim to justify it — which is exactly the wrong way to think about insurance. You do not buy it hoping to use it.
Claims, approvals, and what actually pays out
I never filed a claim, so I cannot speak from experience — but I read every public Nexus Mutual claim I could find because the payout record is the product. Three cases frame what works and what does not.
Euler Finance (~$3.4M paid). Euler was hacked in March 2023 for about $197M; Nexus Mutual paid roughly $3.4M to covered members. This headline case proves the model: a real exploit, a real committee decision, real payouts. It also shows the limits: $3.4M on a $197M loss means cover only compensated the insured fraction, not the full protocol loss.
Arcadia (~$250K paid). A smaller exploit on a lending protocol produced a smaller payout. Worth noting because it shows the committee approves mid-size claims, not just the headline ones.
Stream Finance (~$95K paid). The smallest of the three and the one that made me renew. A ~$95K payout on a niche protocol told me the committee does not only rubber-stamp big names; it evaluates each case on wording and evidence. That is the behavior I want from a mutual I am a member of.
The mechanics: when a covered protocol suffers an incident, you file a claim through Nexus Mutual with evidence. A claims committee reviews it within 72 hours and decides whether the loss falls within the cover wording. The historical approval rate is about 70%, so roughly 3 in 10 claims are denied — usually because the loss was outside covered events (your own mistake, a market move, an unlisted contract) or evidence was insufficient. Members can appeal, and governance votes can override, but the 72-hour window is the first gate.
What this means for you: read the cover wording for the specific product you buy. Smart-contract cover does not cover depegs. Depeg cover does not cover hacks. Neither covers liquidations, impermanent loss, or your seed phrase leaking. The oracle manipulation guide and liquidation cascade explainer are risks standard cover will not save you from.
Insurance vs diversification: what I would renew
If you only remember one number from this article, make it this: diversification is free, insurance costs 1–10% a year. On a $20,000 book, splitting into four $5,000 positions removes 75% of your single-protocol loss exposure at zero cost. No mutual, no committee, no wording. Just math.
The table below is the decision framework I now use before buying any cover. It compares three ways to handle the same $20,000 of exposure to a single protocol you like.
| Approach | Annual cost | Loss if protocol fails | Best for |
|---|---|---|---|
| Full position, no cover | $0 | Up to $20,000 | Small amounts you can lose; high-conviction blue chips. |
| Split across 4 protocols | $0 | Up to $5,000 (one failure) | Most portfolios. Free risk reduction. |
| $20K in one protocol + Nexus cover | $200–$2,000 (1–10%) | $0 if approved; $20K if denied | Concentrated positions you cannot split. |
| Split + cover on one leg | $50–$500 | $0 on insured leg; up to $5K on others | My approach. Diversify first, insure the residual. |
My renewal decision: yes, but smaller. I keep cover on Aave and Lido because those are my two largest single-protocol exposures and the premium is cheap (1–2%). I am dropping the Ethena depeg cover because USDe's peg held through a volatile year and the 2.44% premium was buying protection for an event that has not come close to the 10% threshold. I am adding a small depeg policy on the stablecoin leg of my stablecoin APY tracker book, because that is the concentrated tail I cannot diversify.
For the Morpho vault, I still have no cover. The right mitigation there is curator due diligence, not a premium — read the curator risk guide, check the Morpho review, and keep the position small. Similarly, on Lido the cover is for contract risk, not staking reward risk — the Lido review and Lido vs Rocket Pool comparison walk through what insurance cannot fix.
The year taught me that DeFi insurance is a real product with real limits. It paid $18.5M to real people. It also denied about 30% of claims and will not cover most of the ways you can actually lose money in DeFi. Use it as the top layer on a diversified, audited, small-position book — not as a substitute for the basics. If you are skipping reading audits and wallet security, a premium will not save you. If you have done the basics, $290 on $20,000 is a reasonable price to sleep through the night.
Sources and further reading
- Nexus Mutual — FAQ — how cover, claims and the capital pool work.
- Nexus Mutual — Documentation — cover wording, products and claims process.
- DefiLlama — Nexus Mutual — capital pool TVL and historical data.
- Sherlock — audit and coverage marketplace; pivot away from retail cover.
- Resolv — on-chain recovery infrastructure for hacked protocols.
- Euler Finance — Blog — the exploit and subsequent Nexus Mutual payout reporting.
Frequently asked questions
Is Nexus Mutual insurance worth it for DeFi?
Only for positions where a total loss would hurt you. Nexus Mutual is a discretionary mutual, not a regulated policy: its ~$115M capital pool has paid roughly $18.5M in claims across 184+ listed protocols, but approvals run about 70% and a claims committee can deny a payout. On a $20,000 book I spent about $290 in premiums over a year. For a $1,000 experiment the 1–2% blue-chip cost is noise; for a six-figure concentrated position it is cheap tail protection. Price your own loss tolerance before pricing the premium.
How much does Nexus Mutual cover cost?
Premiums are quoted as an annualized percentage of the amount covered, paid upfront for the period you buy. Blue-chip protocols like Aave or Lido typically run 1–2% per year, while newer or riskier markets can cost 5–10% or more. On my $20,000 book the blended rate came to about 1.45%, or roughly $290 for twelve months of cover. You pay the premium in NXM or ETH at purchase; there is no refund if you withdraw early.
Does DeFi insurance actually pay claims?
Sometimes, and only for the events the cover wording lists. Nexus Mutual's public claims include about $3.4M paid on the Euler Finance hack, roughly $250,000 on Arcadia, and around $95,000 on Stream Finance. Not every claim succeeds: the overall approval rate is about 70%, and the 72-hour claims committee review can deny cases where the loss falls outside the covered events. Cover does not pay for your own mistakes, market losses, or stablecoin depegs unless the depeg product is explicitly purchased and the breach threshold — typically around 10% — is met.
What does Nexus Mutual not cover?
Standard smart-contract cover does not protect you against price crashes, liquidations, impermanent loss, your own private-key compromise, phishing approvals, or a stablecoin holding its peg. Depeg cover is a separate product that only pays after the asset breaches a threshold — roughly 10% below peg — and only for listed assets and time windows. Nexus Mutual also does not cover every protocol; only the 184+ listed contracts are insurable, and cover capacity is capped by the capital pool and risk appetite.
Should I buy DeFi insurance or just diversify?
Do both, but diversify first. Splitting $20,000 across four unrelated protocols removes far more single-point failure risk than any cover can, and it is free. Insurance is the layer on top: it protects the portion you cannot diversify away, such as a large single-protocol position or a concentrated stablecoin holding. On my book I kept positions under $5,000 each, bought cover only on the two I could not easily split, and spent about $290 to cap the tail on roughly $9,000 of exposure. Diversification is the foundation; cover is the deductible you pay for the residual.
Cover complements cold storage; it never replaces it. No policy pays for a compromised seed phrase, which is exactly how most uninsured losses happen. The keys to my covered positions sit on a hardware wallet, and the $80 or so it costs has saved far more anxiety than any premium. DifiCalc earns a commission via the link below at no extra cost to you; the advice stands alone — see our affiliate disclosure.
Grade your DeFi risk before you insure it
Run every position through the yield risk grader to see which risks insurance covers — and which it cannot. Then buy cover only on the residual.
Open the Yield Risk GraderRelated reading: learn to read DeFi audits before depositing, prepare with the DeFi crash weekend playbook, and harden your setup with the DeFi wallet security checklist. Understand stablecoin depeg risk, protect multisig funds with a Safe 2-of-3 multisig, and review Morpho curator risk. For uninsured-loss mechanics: oracle manipulation attacks, liquidation cascades, cross-chain bridge safety, and withdrawal queues and redemption delays. Avoid DeFi yield traps and red flags, watch RWA tokenization risks, and time stablecoin moves with the stablecoin yield rotation guide. Build habits: the monthly DeFi portfolio review, yield portfolio allocation, and DeFi taxes 2026. For staked ETH, compare three ways to stake ETH and the Ethereum staking guide. Discover live rates with yield discovery and pick venues from the best lending protocols.