Typically metrics are computed by taking into account all coins/tokens.
For some metrics, it makes sense also to compute a derivation of them on the subset of coins/tokens that have been moved at least once no longer than X days/years ago.
UTXO blockchains naturally define the age of a coin, but this is not true for account-based blockchains. To define the age of a coin in account-based blockchains we developed our own Coin Age Model
Timebound metrics names are formed from the original metric name plus a timebound suffix. Available timebound suffixes are:
_1d- 1 day
_7d- 7 days
_30d- 30 days
_60d- 60 days
_90d- 90 days
_180d- 180 days
_365d- 365 days
_2y- 2 years
_3y- 3 years
_5y- 5 years
_10y- 10 years
Note: There are metrics that support only part of the timebounds. For example, the mean_coin_age has only
- mvrv_usd_30d - The MVRV metric is computed by considering only the coins/tokens that were active in the past 30 days.
- circulation_3y - The number of tokens transacted at least once in the past 3 years. If a coin/token is considered dead/lost (sent to graveyard address, the owner lost private key, etc.), such circulation can approximate total supply minus lost/dead coins.
circulation_1d metric counts the number of coins/tokens that participated
in on-chain transactions in the past 24 hours.
On one particular day, Alice sends 20 ETH to Bob, Bob sends 10 ETH to Charlie and Charlie sends 5 ETH to Dean.
1 2 3 4 5 6
Alice -- 20 ETH --> Bob | 10 ETH | v Dean <-- 5 ETH -- Charlie
In this scenario the transaction volume is 20 + 10 + 5 = 35 ETH, but the ETH in circulation is 20 ETH.
This difference can be explained as having twenty $1 bills. Alice sends all of them to Bob, Bob sends 10 of the received bills to Charlie, and Charlie sends 5 of them to Dean. There are 20 dollars in circulation total in this case.
One of the most valuable properties of circulation is that it is immune to mixers and gives a much better view of the actual amount of tokens that are being transacted on-chain.