Crypto compound interest calculator

A savings calculator has one moving part. You put in dollars, a rate is applied to dollars, and dollars come out. A staking position has two. The rate is paid in tokens and grows a token balance. The dollar value of that balance is the token count multiplied by a price that moves for reasons the rate knows nothing about. Most calculators that rank for this query were built for a deposit account and quietly treat the two as one thing.

Separating them changes what the output means. The token leg is close to arithmetic. If a validator pays a rate and you restake what it pays, your balance rises by that rate and the only uncertainty is whether the rate holds. The dollar leg is an assumption you supply, and the calculator gives it no weight of evidence at all. It is your number, carried forward, not a forecast.

The place where the two legs meet is a single figure that a dollar only calculator cannot show. For every rate there is an annual price decline that exactly cancels it, and past that point the dollar value falls while the token count rises. On the largest pool in this build snapshot, STETH on lido, the rate is 2.318 percent and the price change that cancels it is -2.27% a year. The cushion a rate buys is always about the size of the rate itself, and a single digit rate sits well inside the ordinary yearly movement of the assets it is paid in.

The calculator above reports that crossover for whatever rate, term and contribution you enter. With a recurring contribution the break even shifts, because later tokens spend less time compounding, so the general form solves the whole plan rather than the rate alone.

Two details in that set are worth stating plainly. The period count is floored, because a partial compounding period pays nothing, and a calculator that credits a fraction of a period is inventing a payment. The contribution is added once per compounding period, which is the assumption the closed form requires. A monthly contribution against a daily compounding schedule is a different and messier calculation, and this page does not pretend the two are the same.

The break even line is the one to read twice. It divides the tokens you put in by the tokens you end with, takes the annual root, and subtracts one. When the contribution is zero it reduces exactly to the price decline that cancels the rate. When there is a contribution it accounts for the shorter compounding life of each later addition.

The rates in the selector are not typed constants. This build read 16,744 pools from the DefiLlama yields endpoint and kept 1,952 that hold at least two million dollars, which is the floor below which a published rate says more about a thin pool than about a market. The largest is STETH on lido at 2.318 percent across 24.85 billion dollars locked.

A rate has parts, and the parts do not behave the same way. The base rate comes from the work the protocol charges for, staking rewards or borrower interest. The reward rate comes from token emissions a team decided to pay and can decide to stop. Of the pools kept in this snapshot, 360 publish a reward component above zero and 219 of those draw more than half their headline rate from it. Compounding an emissions rate over five years assumes a budget decision holds for five years, and that is the failure mode this page is here to name.

Stability is checkable too. The endpoint publishes a 30 day mean beside the current rate. In this snapshot 1,515 pools carry a usable 30 day mean and 168 of them sit more than a quarter above their own recent average, which is the shape of a spike rather than a rate you can plan around. Another 437 pools publish no usable mean at all, and for those there is nothing to compare against, so the page says so rather than filling the gap. The selector shows the split for whichever pool you pick, so the number you compound arrives with its own provenance attached.

One more thing the incumbents get wrong by default. A quoted APY is already an effective annual rate with compounding inside it. Feeding that figure into a calculator set to compound monthly counts the compounding twice and inflates the result. Choosing a pool here sets the frequency to once a year for exactly that reason, and the field stays editable for a nominal rate that genuinely needs compounding applied.

A five year curve is persuasive because nothing in it has to survive contact with a real year. The table under the calculator puts your chosen asset and Bitcoin next to what they actually did over the windows the source publishes, which are 24 hours, 7 days, 30 days, 200 days and 1 year. There is no 90 day figure on this endpoint and none is shown. Where a window is empty the cell reads not available, because a missing measurement is information and a substituted one is not.

The wider picture from the same fetch. Of 212 non stablecoin rows in the universe, 186 carry a 1 year price change, and 146 of those are lower than they were twelve months ago. The median 1 year change across them is -47.1%. Set that against any of the rates in the selector above and the ordering is clear. Price movement dominates the outcome, and the rate is a modest edge applied to something much larger and much less predictable.

Prices in this build were cross checked. 184 rows matched a second independent source within 0.899 percent at worst, and 1 rows were dropped for failing a range check rather than repaired. Our methodology page describes the checks in full, and the scorecard covers the fundamentals that decide whether a token is worth holding long enough for any of this compounding to matter.

An APY is a snapshot of a rate that was being paid at the moment of the fetch. It is not a term rate and nobody has promised it for the length of your projection. Validator sets change, emissions schedules end, borrowing demand falls, and the rate moves with them.

Compounding assumes the reward is restaked without cost. In practice claiming and restaking costs gas, some protocols pay on a schedule rather than continuously, and unbonding periods mean the tokens are not always available when you want them. On a small balance those frictions can consume most of a low single digit rate. The calculator models none of them, and treating its output as achievable is the first mistake to avoid.

Tax is ignored. In several jurisdictions a staking reward is income at receipt, which means the compounding balance carries a liability the token count does not show. Slashing is ignored, and so is smart contract failure, bridge failure and the possibility that a liquid staking token trades below the asset it represents. Impermanent loss is ignored, which matters for any pool holding more than one asset.

The price assumption is yours. This page supplies no forecast and treats a flat price as the default because a flat price at least states its own assumption openly. A long horizon at a high rate produces a large number that says more about the exponent than about any outcome. If a fifty year projection at eighty percent looks like a plan, the arithmetic is working and the model is not.

For the trading side of a position, entry and exit fees, break even price and loss recovery live on the crypto profit calculator. For what we currently rate as worth research attention, see opportunities.