You checked the APY yesterday and it said 4.1%. Today it says 3.7%. Nobody changed anything on your side, and no announcement went out. What happened?
Staking rewards are the output of a live formula. That formula reacts to how many coins are staked, how the validators are behaving, and how busy the network is. It's less like a savings-account rate and more like weather — always changing, sometimes calm, occasionally rough. Here's a plain-English tour of what's actually going on.
The total stake ratio does the heavy lifting
Most proof-of-stake networks issue a fixed pool of rewards each period. That pool is split across everyone who staked. When more people stake, each staker's slice shrinks. When people leave, the remaining slices grow.
On Ethereum, the reward per validator roughly follows a curve based on the square root of total ETH staked. So doubling the amount staked doesn't halve the reward — it reduces it by a factor of about 1.4. The math is deliberate: the network wants enough stake to be secure without paying more than needed.
Validator performance nudges the number
Even in a calm market, individual validators earn different amounts. Some run lean setups with high uptime; others miss slots or get penalised for late attestations. Those losses don't stay on the sloppy validator alone — pool-based staking spreads them across everyone.
- Missed attestations reduce the pool's yield slightly.
- Slashing events remove a chunk of stake and reduce future rewards.
- Well-run pools with vetted operators post steadier numbers than open pools.
This is one reason pool choice matters. See our risk rundown for how slashing plays out in practice.
Network activity adds a tip
On Ethereum, part of the reward comes from priority fees and MEV (the ordering economics of the blockchain). Busy weeks pay more; quiet weeks pay less. That's a genuine variable, not a marketing quirk.
Solana works differently — most of its yield is inflation-based, though vote credits and tips still nudge things. Every network has its own mix of "guaranteed" issuance and "performance-linked" income. Understanding your network's split explains why rates can shift week to week without anything dramatic happening.
Protocol fees set what actually lands in your wallet
| Reward source | Who earns it | Who takes a cut |
|---|---|---|
| Base issuance | Validator | Protocol fee, if using a pool |
| Priority fees / tips | Validator | Same |
| MEV rewards | Validator (via searchers) | Same |
Liquid staking providers commonly keep around 10% of everything, which is why liquid staking APY reads lower than gross. Different providers charge different amounts, and the fee is usually the single largest determinant of what "3.5%" versus "3.9%" means to your wallet. Read more in this liquid staking intro.
Inflation schedules quietly do a lot
Ethereum's issuance is capped and offset by fee burns. Solana's issuance follows a schedule that gently declines over years. Both directly shape long-term staking rewards.
The consequence: an APY that looks stable this year may drift lower over five years even if everything else stays the same. When you compare rewards across networks, compare the issuance model too, not just today's number. And remember these are token-denominated returns. If the token loses half its price, the yield doesn't rescue you. Our rewards calculator lets you model the compounding side without pretending price is fixed.
Putting it all together: why staking rewards APY changes
The number moves because the inputs move. More stakers push it down. Higher network activity pulls it up. Slashings and poor operators drag it lower. Protocol fees take their share. Inflation schedules set the long-term floor. A quoted APY is a snapshot of all of those forces at once. Expect it to change constantly, don't chase week-to-week highs, and pay attention to the trailing average rather than any single day's headline. That framing turns "why did my rate move" into a question with a boring, satisfying answer.