Skip to main content
The On-Chain Dollar Rate is what depositors actually earned lending USDC and USDT in DeFi vaults over the last 90 days, net of fees and excluding reward programs. Webacy publishes it daily as a USDC rate, a USDT rate, a composite of the two, and the same rate by Webacy grade. You can see it on the Yield page in DD.
The dollar rate is an indicative index for information only. It is not a regulated benchmark and is not meant for contractual use.
This page describes methodology version dollar-rate-v0.1.2. Every published day records the version it was computed with, and the full history has been computed under this version.

​
What’s in it

A vault is in scope when its single underlying token is canonical USDC or USDT, matched by contract address on each chain, never by ticker:
  • USDC: native Circle USDC on Ethereum, Base, Arbitrum, Optimism, Polygon and Avalanche. Bridged USDC.e is a different token and is not included.
  • USDT: the USDT each chain’s DeFi uses. That is native USDT on Ethereum and Avalanche, and the Tether-endorsed bridged token on Arbitrum (USDT0), Optimism and Polygon.
Synthetic or yield-bearing dollars (USDe, sUSDe) and other fiat-backed dollars (USDS, PYUSD) are not part of the rate.

​
Eligibility

Each in-scope vault is checked every day against these rules, in this order. A vault that fails a rule is counted under the first rule it fails, which is what the “How the rate is built” breakdown on the Yield page shows. A vault listed twice on the same day is counted once.

​
The measure

Each vault contributes its realized 90-day yield, read from its own share price on-chain:
A share price is net of the vault’s fees, and reward programs such as Merkl campaigns are not in it, so the rate is what a depositor earned from the vault itself. See Vault Yield & Fees for how share prices are read for each protocol.

​
Weighting

  • TVL-weighted, with a 25% cap. Within the USDC rate and within the USDT rate, each vault is weighted by its TVL, but no vault can carry more than 25% of the rate. The excess weight of a capped vault is spread across the others in proportion to their TVL. In the USDT rate the largest vaults are often held at the cap, so the rate reflects the market as a whole rather than one dominant vault.
  • At least 5 vaults. With fewer than 5 eligible vaults, no rate is published for that asset that day.
  • Composite. The USDC and USDT rates are blended by their eligible TVL, so the composite always lies between the two.
On the Yield page, each vault’s weight is its share of its USDC or USDT rate after the cap. Multiplying each weight by the vault’s realized 90-day yield and summing gives back the published rate.

​
Rate by grade

The same capped rate is also computed within each Webacy grade (A to F), using that day’s vault grade. A grade with too few vaults shows its count and TVL but no rate. Days before March 25, 2026 have no rate by grade, because grades are never applied backwards from today.

​
History and restatement

  • Daily: the rate is computed once a day. The rate summary reports the latest value, the change over 30 days and a year, and the average, high and low over the last 365 days.
  • Backfilled history: days before October 23, 2026 are reconstructed from on-chain share prices and total assets, with USDC and USDT valued at $1. Deposit status isn’t known for those days, so every otherwise eligible vault is included. Vaults that had already closed before the history was built are not included.
  • Versioned: every published value records its methodology version. A past day only changes when the methodology changes, and then it is recomputed under the new version.

​
Known limitations

  • Concentration: a few large vaults hold most USDT deposits. The 25% cap keeps any one vault from setting the rate, so the rate is not a plain TVL average.
  • Survivorship: backfilled days include only vaults that still existed when the history was built.
  • Deposit status: backfilled days don’t know whether deposits were open, so they can include vaults a new depositor couldn’t have entered.