Skip to main content

Sum

The sum of a metric is the total value of all the data points in the metric. This can be useful for calculating total sales, volume, or other metrics.

As we have seen earlier, we calculate sum by memtioning it in the fields that support summation.

amount(calculate: sum)

Summation with criteria​

We can also caculate sum by restricting the results with additional criterion. Let's look at an example: The query will return the sum of the amounts of tokens transferred to and from the specified address, in USD


query ($network: EthereumNetwork!, $address: String!, $from: ISO8601DateTime, $till: ISO8601DateTime, $limit: Int!, $offset: Int!) {
  ethereum(network: $network) {
    transfers(
      date: {since: $from, till: $till}
      amount: {gt: 0}
      any: [{receiver: {is: $address}}, {sender: {is: $address}}]
      options: {limit: $limit, offset: $offset, asc: "currency.symbol"}
    ) {
      sum_in: amount(calculate: sum, receiver: {is: $address})
      sum_out: amount(calculate: sum, sender: {is: $address})
      sum_in_usd: amount(in: USD, calculate: sum, receiver: {is: $address})
      sum_out_usd: amount(in: USD, calculate: sum, sender: {is: $address})
      currency {
        address
        symbol
        tokenType
      }
    }
  }
}

{
  "limit": 10,
  "offset": 0,
  "network": "ethereum",
  "address": "0xb5d85cbf7cb3ee0d56b3bb207d5fc4b82f43f511",
  "from": "2022-06-17",
  "till": "2022-06-24T23:59:59",
  "dateFormat": "%Y-%m-%d"
}

By mention the address as a sender or receiver we are calculating:

  • The total amount of ETH that was transferred into and out of the address, as sum_in and sum_out fields
  • The total amount of ETH that was transferred into and out of the address as the sum_in_usd and sum_out_usd fields

Sum is the aggregation you reach for when only the total matters: how much volume moved, how many tokens flowed in aggregate, or what fees added up over a window. It complements row-level listings: you might still use limit for samples, but calculate: sum answers headline KPIs in one response. Pairing sum with filters (time range, address, currency, amount thresholds) defines the population being totaled; pairing it with grouping at the query level (for example, per-currency transfer rows) yields subtotals per bucket instead of one global figure.

Practical Tips​

  • Trading and liquidity metrics: Use sum on trade or transfer amount fields to report total volume, turnover, or notional over a period for a pair, pool, or protocol.
  • Treasury and flow analysis: Combine sum with sender/receiver criteria (as in the example) to separate inflows, outflows, and USD equivalents for an address or contract.
  • Fees and costs: When the schema exposes fee-related amounts, summing over a date range gives protocol revenue or user cost totals without exporting raw rows to a spreadsheet.
  • Filtered totals: Apply the same predicates you use for listing events so the sum reflects only “large transfers,” “whitelisted tokens,” or other business rules—not the whole chain history.
  • Per-dimension subtotals: When your query returns one row per currency, token, or day (depending on schema support), each row’s sum is that slice’s total; verify how your selection groups data so you do not misread a global aggregate as a per-asset one.