Direct answer
A staking rate tells you how fast your token count grows. It says nothing about how fast the total number of tokens grows underneath it. Those are two separate rates, and your share of the supply is the first divided by the second — not the first minus the second. Between the advertised rate and your balance sit two more mechanisms: whether the number is an APR or an APY, which are not interchangeable, and what share of the rewards a validator or platform keeps before they reach you.
- The condition that matters most
- A share of the token supply is a share of a token count. It is not money value, not purchasing power, not governance weight and not a claim on protocol revenue. You can hold a growing share of a supply and lose money, and hold a shrinking share while the price rises.
- What this does not tell you
- This article explains four mechanisms and how they combine. It recommends no token, protocol, validator or platform, publishes no rate, no supply figure and no price, and takes no view on whether any yield is sustainable or whether staking is worthwhile. Supply dilution is not evidence about price in either direction, and nothing here forecasts one.
- Last reviewed
Two rates, and only one of them is advertised
A staking page shows a percentage. Sometimes it is large. Whatever it is, it is answering one question:
How fast does your token count grow?
There is a second rate, and it is almost never on the same page:
How fast does the total token count grow?
Both matter, because owning tokens is only ever owning a fraction of the tokens that exist. If your balance grows 12% in a year while the total supply grows 8%, your balance is a larger fraction than it was. If your balance grows 12% while the supply grows 20%, you finish the year holding more tokens and a smaller fraction of them.
That second sentence is the one this article exists for. It describes a perfectly ordinary outcome, it is not a trick, and nothing about a headline rate hints at it — because a headline rate was never measuring it.
Four mechanisms sit between the number on the page and what actually happened. They are separate, they combine in a specific order, and three of them are routinely collapsed into one.
Mechanism one: APR and APY are not the same number
An APR is a nominal annual reward rate. It is a rate of accrual, and on its own it is not yet growth: what it does over a year depends entirely on what happens to the rewards as they arrive.
- Left alone, they accrue against the balance you started with. 12% APR over a year on 1,000 tokens is 120 tokens, and those 120 tokens earned nothing themselves.
- Restaked, each reward joins the balance and earns in every period after it. 12% APR compounded monthly is 1% a month, twelve times, which is 12.6825% — not 12%.
An APY is an effective annual growth rate. It has a compounding convention inside it already. 12.6825% APY is the result of 12% APR compounded monthly; the two describe the same year.
Which means the single most common error in staking arithmetic is to take an APY and compound it again at some frequency. That reinterprets it as an APR and inflates the result. The correction is not subtle and it is not optional: an APY is applied directly, once per year of the period, and nothing else is done to it.
An APR needs a compounding assumption before it becomes growth. An APY already has one, and adding a second is not a refinement — it is an error.
Mechanism two: a reward fee is a share of the rewards
A validator commission is a share of the rewards the validator produced. So is a platform reward fee. Neither is a charge against the tokens you already hold.
This distinction has a sharp consequence that is worth stating before it comes up: at a 100% reward fee, your balance stops growing and stays exactly where it started. It does not fall to zero. A fee that is a share of rewards cannot reach the balance, because the balance is not what it is a share of.
The mirror-image error is more common and more expensive. Applying a reward commission as though it were a percentage of the whole balance — 10% of 1,000 tokens rather than 10% of the 120 tokens of reward — overstates the cost by roughly a factor of the rate itself. On a 12% APR that is being wrong by about eight times.
Whether the rate you were shown is already net
Reward fees create a second problem, and it is one of presentation rather than arithmetic. Some published rates are gross — before the validator takes their share. Some are already net — after it. Nothing about the number tells you which, and the two are frequently displayed identically.
If you deduct a commission from a rate that is already net of it, you have deducted it twice. If you fail to deduct one from a gross rate, you have not deducted it at all. There is no way to tell from the figure, and no calculator can tell for you: it is a fact about the page you read it on.
The awkward middle case is real and worth naming: a rate quoted net of the validator’s commission but before a separate platform fee. The answer is to treat it as a gross rate with respect to the fee you still need to model, and to enter only that remaining fee.
Why taking a fee out of an APY is not a multiplication
If a rate is quoted as a gross APY and a commission comes out of it, the
tempting shortcut is net APY = gross APY × (1 − fee). It is wrong, and the
reason is mechanical rather than pedantic.
A fee is taken out of each reward as it arrives. A smaller reward is added to the balance, so a smaller base earns in every subsequent period. The fee changes the path of the compounding, not just its endpoint, and scaling the annual figure does not reproduce that.
Doing it properly means going backwards first: convert the APY to the periodic reward that would produce it, take the fee out of that, then compound the result. On a 12.6825% gross APY with monthly rewards and a 10% commission, the implied monthly reward is 1.0000%, the net monthly reward is 0.9000%, and the year’s growth is 11.3510%. The shortcut gives 11.4143%. The gap is small here, and it is a gap in the direction that flatters the product.
Mechanism three: total supply is a moving denominator
Most proof-of-stake networks pay staking rewards by issuing new tokens. Some burn tokens as well. The net effect on the total supply is a rate in its own right, and it is set by protocol rules, not by anything you did.
That rate is not the staking rate, and treating them as the same figure is a third distinct error. Rewards are what a staker receives. Issuance is what everyone’s denominator does. The two are related — the first is often funded by the second — but they are not equal, and the gap between them is where the whole question lives.
Three things this article deliberately does not claim about supply:
- Supply growth does not mean the price must fall. Demand, utility, liquidity and what the tokens are used for are all outside the arithmetic.
- Supply contraction does not mean the price must rise, for exactly the same reason. A burn reduces a denominator. It does not create demand.
- Token-supply dilution is not consumer-price inflation. They share a word and nothing else. One changes how many tokens exist; the other changes what a currency buys. If you want the second, that is a different calculation with a different input.
Mechanism four: the answer is a division
Here is the part that is easiest to get almost right.
Your share of the supply changed by:
(1 + net token growth) ÷ (1 + supply growth) − 1
Not:
net token growth − supply growth
The subtraction is an approximation. It is a good one when both rates are small, which is why it survives: at 3% against 2% it is off by six-hundredths of a percentage point and nobody notices. It degrades as either rate grows, and it degrades fastest exactly where people rely on it most — a large advertised rate against a large supply expansion.
Two worked cases, both produced by the model behind the calculator:
| Scenario | Token growth | Supply growth | Subtraction says | Division says |
|---|---|---|---|---|
| 12% APR restaked monthly, 10% reward fee, 8% supply growth, one year | 11.35% | 8.00% | 3.35% | 3.10% |
| 15% APY already net of fees, 20% supply growth, two years | 32.25% | 44.00% | −11.75% | −8.16% |
In the second row the subtraction overstates the loss of share by more than three and a half percentage points. In both, the shortcut and the answer disagree about a figure someone would use to make a decision.
There is a reason the division is the right form and it is not arbitrary. A share is a ratio. Your tokens are the numerator and the supply is the denominator, and when both change you divide the new ratio by the old one. Subtracting two growth rates would be correct if shares were differences, and they are not.
The threshold: what it takes to stand still
The boundary is simple and worth naming, because it reframes what a staking rate has to clear:
Your share of the supply is unchanged when your balance grows by exactly as much as the total supply over the same period.
If the supply grows 8% a year, a net staking yield of 8% a year leaves you with exactly the fraction you had. Everything below that is a shrinking share; everything above it is a growing one.
This is a mathematical boundary between two numbers, and it is emphatically not:
- a required return, or a yield you should be aiming for;
- a rate any protocol has offered or could offer;
- a measure of whether staking was worthwhile;
- a claim that a shrinking share is a loss, or a growing one a gain.
It is a level, and the only thing it tells you is which side of it a scenario sits on.
More tokens, less ownership: the whole point
Put the four mechanisms together and the outcome the headline rate cannot show becomes ordinary rather than surprising.
A network pays a large advertised rate. It funds that rate by issuing tokens. A staker takes the rate, loses a share of it to a commission, and finishes the year with substantially more tokens than they started with — and a smaller fraction of a much larger supply.
Nothing went wrong. Nobody was defrauded. The advertised rate was accurate. It was answering the question it said it was answering, which was about a token count, and the question about ownership was never asked.
That is why “what is the staking APY?” is a weaker question than it sounds, and why “what is the net issuance?” belongs beside it.
What a share of the supply is not
The share figure is precise, and it is narrow. It is worth being explicit about what it does not establish, because every one of these is a substitution a reader makes naturally.
- It is not money value. Your share can rise while the price falls far enough to leave you worse off in every currency. It can fall while the price rises enough to leave you better off. The two are separate lenses and adding them together produces a number that answers neither.
- It is not purchasing power. Nothing in this calculation is adjusted for what anything costs.
- It is not governance weight. Voting power depends on what is staked, what is delegated, what participates and how the mechanism is designed — not on a share of total supply.
- It is not a claim on revenue. Whether a protocol earns anything, and whether holders receive it, is a question about the protocol’s economics.
- It is not protection against dilution in any general sense. It measures dilution. It does not resist it.
When the calculator is worth using
The staking calculator is useful in a narrow set of situations, and it is honest about which:
- you have a rate and you know whether it is an APR or an APY;
- you can find out whether it is quoted before or after the fees you care about;
- you have a view — your own — on what the total supply will do;
- and you want to see the two rates against each other rather than one of them alone.
It is not useful if any of those is a guess you would rather not make. A supply assumption you invented produces a share figure that is precisely as invented as the assumption. The calculator does not fetch a rate, a price or a supply figure from anywhere, and that is deliberate: a number retrieved silently is a number nobody wrote down, and a scenario nobody can reproduce.
What none of this covers
The mechanisms above are the arithmetic. They are not the risk, and the gap between the two is large.
- Slashing. A validator that misbehaves or goes offline can cost you principal. That is a probability question with a protocol-specific answer, and it is nowhere in a yield calculation.
- What funds the reward. Issuance, transaction fees, MEV, a treasury subsidy: these have very different implications for whether a rate persists, and the rate itself does not distinguish them.
- Whether the rate persists at all. Reward rates move with participation, emissions schedules and governance decisions. A calculation over ten years at a constant rate is a scenario, not a projection.
- Liquid staking. A liquid-staking token trades against the asset it represents, at a rate that is not fixed. That exchange rate is its own risk and is not modelled by treating the position as a token balance.
- Lock-ups. A lock changes what you can reach and when. It does not change the arithmetic of a reward rate — and specifically, a longer lock does not raise a rate. Some products offer a higher tier for a longer commitment; that is a different rate you would have to enter yourself.
- Counterparty and contract risk. A custodial platform can fail. A contract can be exploited. Neither appears as a percentage anywhere.
- Tax. Staking rewards are treated very differently between jurisdictions, often at the moment of receipt. Every figure discussed here is before tax.
The one sentence
If you take nothing else from this:
More tokens is not more ownership, and neither of those is more money.
Three separate questions, three separate answers, and a headline rate answers only the first.
Sources
UBWHY's own work
Calculations and reconstructions produced by UBWHY, recorded so the method can be examined. Not independent evidence, and not verification of the records they are built from.
UBWHY Staking Yield after Fees and Token Dilution calculation model
UBWHY Tool Blueprint — Staking Yield after Fees and Token Dilution, blueprint version 1.0, calculation model version "Staking Yield after Fees and Token Dilution v1.0", last reviewed 5 August 2026. Held in the UBWHY repository and not published as a document.
Supports: UBWHY’s own calculation model: the separation of a nominal APR from an effective APY and the refusal to compound an APY a second time; the treatment of every reward fee as a share of rewards rather than as a charge against the balance; the two-state rate fee basis and the consequent unavailability, rather than nullity, of a fee impact under a rate already quoted net; the inversion of a gross APY to its implied periodic reward before a commission is deducted and recompounded; the effective annual net total-token-supply factor; the multiplicative relative token-supply-share identity as the net token factor divided by the supply factor, and the explicit prohibition of the subtraction form; the annualised restatements of both the token yield and the share change; the optional user-entered token-price scenario held separate from the share result; and the lock-up comparison that alters no factor.
UBWHY’s own working specification, not independent evidence, and filed as such. Its arithmetic is verified twice against the published test vectors: once by an independent specification verifier and once by the production model itself. It contains no staking rate, token price, token supply, validator record, protocol parameter, provider fee schedule or market data of any kind, and it retrieves none at build time or at run time.
Figures that UBWHY calculates, and the conclusions drawn from them, are UBWHY's own work and are labelled as such in the text. They are not claims made by any source above.