$130 Billion — DeFi Is Quietly Rebuilding While Nobody Is Watching
There is a pattern I have noticed across multiple crypto cycles. During bull runs, everyone talks about DeFi. TVL charts make the rounds on social media, yield farming strategies fill YouTube, and every protocol launch gets breathless coverage. During drawdowns, the conversation shifts entirely to price action and macro. DeFi keeps building, keeps accumulating value, and keeps innovating — but almost nobody pays attention until the next price surge forces them to look again.
We are in one of those quiet rebuilding phases right now. Total DeFi TVL across all chains currently sits at approximately $130-140 billion, depending on which aggregator you use and how they handle double-counting of liquid staking derivatives. That number, by itself, might not seem exciting. It is well below the all-time high of roughly $180 billion reached in late 2024. But context matters: the post-FTX bottom was approximately $50 billion. DeFi has nearly tripled from its cyclical low, and the composition of that TVL is far healthier than what we saw during the yield farming mania of 2021.
I have spent the past two weeks going through protocol-level data for the top 20 DeFi platforms by TVL. What follows is my analysis of the three protocols that I believe will define the next phase of DeFi evolution: Lido, Aave, and EigenLayer.
Lido — $27.5 Billion and the Liquid Staking Monopoly Question
Lido remains the largest DeFi protocol by TVL at approximately $27.5 billion. That represents roughly 21% of all DeFi value locked across every chain. To put it in corporate terms, Lido has a market share that would make most Fortune 500 companies envious. The product itself is straightforward: you deposit ETH, receive stETH (a liquid staking token that accrues staking rewards), and can use your stETH across the broader DeFi ecosystem while your original ETH earns validator rewards.
The simplicity is both Lido’s strength and its vulnerability. On the strength side, stETH has become a de facto money primitive in DeFi. It is the most widely accepted collateral on Aave, MakerDAO, and dozens of other lending protocols. It trades with deep liquidity on Curve, Uniswap, and centralized exchanges. The network effects are formidable — the more protocols that accept stETH, the more valuable holding stETH becomes, which drives more deposits to Lido, which increases stETH liquidity, which convinces more protocols to accept it. This flywheel has been spinning for three years and shows no signs of slowing.
The vulnerability is concentration risk. Lido currently accounts for approximately 29% of all staked ETH. The Ethereum community has repeatedly expressed concern about any single entity controlling more than 33% of stake, which would give it theoretical power over block finality. Lido’s governance has taken steps to address this by diversifying its node operator set and implementing a “self-limiting” governance framework, but the fundamental tension remains. If Lido’s market share continues growing, I expect regulatory or community pressure to intensify.
My personal position: I hold stETH as roughly 40% of my ETH exposure. The staking yield (currently around 3.8% APR) provides a meaningful return floor, and the DeFi composability means I can simultaneously use my stETH as collateral on Aave to borrow stablecoins for other opportunities. The capital efficiency is excellent. I would be more concerned about concentration risk if Lido were controlled by a single entity, but its DAO structure and distributed operator model give me reasonable comfort — for now.
Aave — $27 Billion and the Evolution From Lending to Financial Infrastructure
Aave sits just below Lido at approximately $27 billion in TVL, and the gap has been closing. What impresses me most about Aave is not the raw number but the strategic transformation the protocol has undergone over the past 18 months. Aave v3 was already a significant upgrade with its cross-chain portals and efficiency mode. But the recent launches of GHO (Aave’s native stablecoin) and Aave v4’s modular architecture signal something bigger: Aave is positioning itself not just as a lending protocol but as a full-stack financial infrastructure layer.
GHO is the piece that most analysts underestimate. A lending protocol that also issues its own stablecoin creates a closed-loop system where borrowing demand directly supports the stablecoin’s utility, which drives more borrowing demand. MakerDAO proved this model works with DAI, but Aave has the advantage of starting with an already massive user base and deep protocol integrations. GHO supply has grown to approximately $800 million from near zero 14 months ago, and I expect it to cross $2 billion by year-end if Aave continues its current integration pace.
The risk for Aave is competition and regulatory uncertainty. Morpho, a peer-to-peer lending optimizer built on top of Aave and Compound, has been quietly eating into Aave’s market share on certain pairs. And the SEC’s ongoing scrutiny of DeFi lending protocols — while not yet resulting in enforcement actions against Aave specifically — creates an overhang that dampens institutional participation. Still, Aave’s revenue generation (approximately $450 million annualized protocol revenue) gives it the financial resources to weather regulatory challenges that would sink smaller competitors.
EigenLayer — $13 Billion and the Restaking Revolution
EigenLayer is the youngest of the three major protocols but arguably the most innovative. At approximately $13 billion in TVL, it has grown from zero to top-five DeFi protocol status in under 18 months. The core concept — restaking — allows ETH stakers to opt in to securing additional protocols and services beyond Ethereum itself, earning extra yield in exchange for accepting additional slashing conditions.
I was initially skeptical of restaking. The pitch sounded like a leveraged yield play that would add systemic risk to Ethereum’s security model. And to some extent, that concern remains valid. If a major Actively Validated Service (AVS) secured by restaked ETH suffered a catastrophic failure that triggered mass slashing, the cascading effects could temporarily destabilize Ethereum’s own staking economics. This is not a theoretical risk — it is a mathematically modeled scenario that the EigenLayer team has acknowledged and attempted to mitigate through slashing caps and insurance mechanisms.
But as I have studied the ecosystem more deeply, I have come to appreciate what restaking actually enables. It allows new protocols to bootstrap economic security without creating their own token and hoping enough people buy and stake it. Oracle networks, data availability layers, cross-chain bridges, and keeper networks can all tap into Ethereum’s massive staking pool for security. This dramatically reduces the cost and time required to launch new infrastructure, which should accelerate the overall pace of crypto innovation.
The AVS ecosystem has grown to over 20 live services, with another 40+ in development. EigenDA — EigenLayer’s own data availability service — has become the data availability provider for several L2 rollups, directly competing with Celestia. The revenue generated by AVS fees flows back to restakers, creating a yield premium above standard ETH staking. Current restaking yields range from 5.2% to 8.7% APR depending on the AVS mix, compared to 3.8% for vanilla ETH staking.
Ethereum’s 68% DeFi Dominance — Can It Hold?
One macro trend worth noting: Ethereum still accounts for approximately 68% of all DeFi TVL. This dominance has actually increased slightly over the past six months, despite aggressive growth from Solana DeFi and the emergence of new L1 contenders. The reason is straightforward — the biggest DeFi protocols (Lido, Aave, MakerDAO, EigenLayer, Uniswap) are all Ethereum-native, and their combined TVL dwarfs the entire DeFi ecosystem on any other chain.
Solana’s DeFi TVL sits around $12-14 billion, driven primarily by Jito (liquid staking), Marinade, and the Jupiter DEX aggregator. That is impressive growth from roughly $1 billion post-FTX, but it is still less than Lido alone. BNB Chain has approximately $8 billion, mostly in PancakeSwap and Venus. Avalanche, Polygon, and Arbitrum each sit in the $3-6 billion range.
I do not expect Ethereum’s DeFi dominance to erode significantly in 2026. The composability advantages — where stETH can be deposited in Aave, which can be borrowed against in GHO, which can be deployed in Uniswap liquidity pools — create an ecosystem density that no other chain has replicated. Cross-chain DeFi exists, but it adds complexity, latency, and bridge risk that institutional capital finds unattractive. Most large DeFi allocators I have spoken with explicitly prefer single-chain strategies on Ethereum for these reasons.
The Recovery in Context — From $50B Post-FTX to $130B Now
The recovery trajectory itself tells an important story about DeFi maturation. The post-FTX bottom was approximately $50 billion, driven by a combination of genuine value destruction (Luna/UST), contagion fear (which funds held FTX/Alameda exposure?), and reflexive deleveraging across every DeFi protocol. What was notable about the recovery was how gradual and organic it has been. Unlike previous cycles where TVL spikes were driven by yield farming ponzinomics and recursive leverage, this recovery has been built on genuine demand for lending, staking, and exchange services.
Aave’s TVL growth, for example, has been closely correlated with actual borrowing demand rather than incentive farming. Lido’s growth reflects genuine staker preference for liquid staking over running solo validators. EigenLayer’s growth, while partly speculative (airdrop farming was real), has also been driven by legitimate AVS demand for economic security. The quality of TVL in 2026 is markedly higher than in 2021, when $100 billion of the $180 billion peak was arguably circular and unsustainable.
My Outlook: EigenLayer Restaking Drives the Next DeFi Innovation Wave
If I had to place one bet on which force will most shape DeFi in the next 12 months, it would be EigenLayer and the restaking ecosystem. Not because restaking itself generates the most revenue — Aave and Lido are far ahead on that metric — but because restaking reduces the bootstrap cost for new protocols to near zero. Any team that wants to launch a decentralized service now has access to billions of dollars of economic security on day one, without needing to convince anyone to buy and stake a new token.
The second-order effects of this are enormous. More protocols launching faster means more experimentation, more innovation, and more competition. Some of those protocols will fail, but the ones that succeed will create new DeFi primitives that we cannot even imagine yet. The restaking thesis is not about EigenLayer’s TVL number — it is about the combinatorial explosion of new services that shared security enables.
My portfolio positioning reflects this view. I hold ETH (including stETH through Lido) as my base layer exposure. I have a meaningful EigenLayer restaking position across three AVSs that I have personally vetted for slashing risk. I hold AAVE tokens as my bet on lending infrastructure dominance. And I maintain a watchlist of early-stage AVS tokens that I plan to accumulate if their post-launch prices reach my target levels. The DeFi recovery from $50B to $130B was phase one. The restaking-driven innovation wave is phase two, and I believe it takes total DeFi TVL above $200 billion by mid-2027.