How to Choose a Cardano Stake Pool: Comparing Fees, Performance, and Decentralization Metrics

Last week, CardanoScope shipped a staking analytics module: a stake pool ranking, a saturation calculator, and a lookup tool that lets any delegator paste in a pool ID and see exactly where it sits on three axes — saturation, margin, and fixed cost. I built it because I kept running into the same gap: ADA holders can see their balance and their last reward, but almost nothing tells them whether the pool behind that reward is still a good place to be.

So before anything else: if you are delegating, go check your pool. Not because I assume your pool is bad — most aren't — but because saturation, margin, and fees change every epoch, and "set it and forget it" is the wrong default for a metric that moves.

What "Healthy" Actually Means

A pool's health is not its APY. APY is mostly a function of network-wide parameters and short-term luck in block production; it tells you almost nothing about the operator's behavior. Three numbers do:

• Saturation — live stake relative to the network's saturation threshold. As of epoch 639, that threshold sits at roughly 77.4 million ADA per pool, with k=500 as the protocol's optimal pool count and 2,940 pools actually live on-chain. Above the threshold, rewards per ADA delegated start to decline — the protocol is designed to push delegators toward decentralization, not just yield.
• Margin and fixed cost — the operator's cut. A pool charging more than roughly 3% margin, or more than the protocol's minimum fixed cost of 340 ADA, isn't necessarily predatory, but it's a worse deal than the median operator offers for comparable security.
• Pledge — not a verdict on its own, since large multi-pool operators split pledge across pools, but worth checking as context.

CardanoScope's pool table flags a pool as crowded above 85% saturation and oversaturated above 100%, and surfaces a rotating daily list of pools sitting in the 30%–85% saturation band with reasonable fees and real pledge — not a blind delegation recommendation, just a starting point of pools that aren't an obviously bad choice on those three axes.

Why Cardano Never Needed a Liquid Staking Token

This is also a good moment to clear up something that confuses people coming from other chains, Ethereum especially: Cardano never built a separate liquid staking dApp, because it never had the problem liquid staking exists to solve.

On Ethereum, running a validator means locking 32 ETH into the protocol itself, with an exit queue to get it back. That illiquidity is exactly why Lido, Rocket Pool, and similar protocols exist — they're third-party applications that issue a derivative token (stETH, rETH) representing locked stake, so holders can keep using that value elsewhere in DeFi while the underlying ETH stays locked. The derivative token is the product; the lock-up is the problem it solves.

Cardano's delegation was designed differently from the start. Delegating a stake key never moves or locks the ADA. You can spend it, trade it, or move it to another wallet the same epoch you delegate, and you keep earning rewards, because rewards are computed from a stake snapshot, not custody. There's no unbonding period, no derivative token, and no extra smart-contract risk layered on top of staking — because there was never a liquidity problem to engineer around. For a non-technical user, this is arguably the friendliest staking design among major proof-of-stake networks: delegate in one transaction, never lock anything, never touch a third-party contract.

The honest tradeoff sits on the other side of that design. Liquid staking derivatives don't just solve a UX problem — they also became deep DeFi collateral, and the lock-up itself feeds a supply-side narrative some investors price in. ADA staking has no equivalent "removed from circulating float" effect, because nothing about it is ever actually removed. Cardano's staking ratio — roughly 56% of active supply as of epoch 639, computed directly from CardanoScope's own db-sync query — is high by any standard, but it doesn't tighten liquidity the way locked-and-restaked ETH does, because the ADA underneath that 56% never stopped being liquid in the first place.

A Thin Market Either Way

Here's the part that matters more than the mechanism: even if Cardano's staking model produced the same kind of lock-up narrative Ethereum's does, there isn't much on-chain liquidity right now for that narrative to act on. DeFiLlama puts total Cardano DeFi TVL at roughly $91 million today, and total stablecoin supply on Cardano at roughly $49 million. Set that against a 56% staking ratio on tens of billions of dollars of ADA, and the picture is consistent: capital on Cardano overwhelmingly sits in delegation, not in DeFi. I made a related point about this capital-activation gap in an earlier piece, and the fresh numbers here say the same thing — the absence of a staking lock-up effect is a smaller cost today than it would be on a chain with deep, active on-chain markets. There isn't much liquidity being held back from circulating, lock-up or no lock-up.

Some in the community call this a "ghost chain," and the TVL and stablecoin numbers above are exactly why that label sticks, however unflattering. I'd put it more precisely: Cardano has built a real structural equivalent to most of the other ecosystems' DeFi stacks — DEXs, lending markets, a privacy sidechain in Midnight — without yet producing the one application that pulls users in from outside the existing Cardano-native crowd. Every chain that broke into mainstream usage did it behind one application that people outside crypto actually used. Cardano doesn't have that yet.

That's not a reason for discouragement about the chain's fundamentals — the staking layer alone, as this piece argues, is genuinely well designed and genuinely friendly to the people using it. It's a reason to keep building toward the application that closes the gap, instead of assuming infrastructure alone will do it. If you're delegating, check your pool's health today. If you're building, the gap I just described is still open.

--- Sources:

• CardanoScope staking analytics, computed from cardano-db-sync (epoch_stake, pool_update, pool_retire, epoch_param), accessed 2026-06-22
• DeFiLlama, Cardano chain TVL, accessed 2026-06-22: https://defillama.com/chain/cardano
• DeFiLlama, Cardano stablecoin circulating supply, accessed 2026-06-22: https://defillama.com/stablecoins

*[email protected]*