Data MethodologyAvailable

How Amonhen Data is structured and evaluated.

Transparency into the sourcing, indexing, calculation, and coverage of Robinhood Chain datasets.

This page summarizes what is described in the Data application's own documentation. Where a detail is not documented, it is stated as such rather than assumed.

A

Data Sources & Provenance

Amonhen Data is delivered by an indexing layer that sits above the chain's retained raw data. It does not replace raw blockchain data. It decodes it, links values to their source evidence, and publishes defined measurements. The information it organizes includes:

  • Blockchain transaction and event data from Robinhood Chain
  • Smart contract interactions, decoded for supported venues (currently Uniswap-V3 and native/bonding-curve trades)
  • Token transfers and token metadata
  • Onchain market activity, derived from decoded trades

Raw blockchain information

Blocks, transactions and events as recorded on the chain and retained in the raw store.

Processed or normalized datasets

Decoded tables such as tokens, transfers, swaps, pools, balances and blocks.

Derived metrics and estimates

Calculated values such as token statistics, rolling volume, minute bars, momentum, and USD display values.

Directly observed values are those decoded from onchain records, such as a transfer or a swap. Calculated metrics are computed from those records and depend on the definitions and assumptions described below.

The Data documentation does not describe any third-party market-data feed. Prices it describes are derived from decoded onchain trades. Upstream infrastructure providers are not named on this page.
B

Indexing & Data Freshness

Freshness is communicated so users can judge how current a value is, rather than assuming it is current.

  • Latest indexed position. The Data application's API reference lists serving-metadata and coverage endpoints that describe indexing status.
  • Update behavior. The documentation states that the indexer publishes on each processed block, and a live stream pushes new blocks and trades as they are indexed. Amonhen does not commit to a specific indexing latency.
  • Rolling windows. Metrics such as 24-hour volume use exact elapsed time, anchored to the processed watermark rather than the wall clock.
  • Reorganizations. Recent blocks on any blockchain can be reorganized. Users should allow for this when relying on the most recent data. Specific reorganization handling is not described on this page.
  • Historical availability. History begins at each dataset's indexing coverage start. Earlier activity may not be represented.
  • Dataset differences. Not every dataset necessarily updates at the same frequency. Derived datasets such as minute bars and momentum are built from accepted events and may trail the underlying ledger.
C

Metric Definitions

The table lists metrics whose definitions are documented in the Data application. Definitions are summarized, not reproduced; follow the links for the authoritative text.

MetricDefinitionData sourceCalculation notes
TokensRegistry of tokens with lifecycle status and decimals status.Tokens datasetIf decimals cannot be verified, the value is shown as unknown. A default is never assumed.
TransfersDecoded ERC-20 transfer ledger with raw integer values and source identity.Transfers datasetRaw values are stored as integers. Scaling uses verified decimals.
Swap activityTrades decoded into source-linked buys and sells, with signed amounts and a quote-unit price.Swaps datasetCovers supported venues only. Each trade carries a price and adapter status.
Token pricesQuote-per-base prices. USD is shown on a named USDG 1:1 basis.Decoded swapsNot a validated USD feed. See Pricing Methodology.
Trading volumeRolling-window volume, for example over 24 hours.Token statistics (derived)Uses exact elapsed time anchored to the processed watermark.
Minute barsImmutable 1-minute OHLC bars in quote units per approved market.Minute-bar datasetBuilt from the accepted event set.
Momentum (r1 / r5)An observed return measure using the last eligible trade strictly before each anchor.Momentum datasetA stale or missing anchor yields no value, with a reason. This is an observed metric, not a forecast.
Balances and holdersRunning wallet balances with a coverage status.Balances datasetReflect transfers from the indexing coverage start only. Provably incomplete balances are flagged.
PoolsFactory-created pools with token pair, fee and creation evidence.Pools datasetPool records are documented. A separate liquidity-depth metric is not.
Not documented publicly at this time: precise definitions for transaction counts, token counts, wallet or address activity, and liquidity metrics. They are not defined here to avoid inventing them.
D

Pricing Methodology

  • How prices are presented. Prices are quote-per-base, in the quote unit of each market.
  • Observed versus calculated. A trade price is observed from a decoded swap. Rolling statistics, bars and USD figures are calculated from those trades.
  • USD conversion. USD values are shown on a named USDG 1:1 basis. This is an assumption, not a validated USD price source.
  • Provisional pricing. Prices from native or bonding-curve trades are tagged as provisional evidence where the venue adapter is not yet approved for interpretation.
  • When pricing is unavailable. If a token's decimals cannot be verified, its price is withheld rather than guessed. Missing or stale values appear as unavailable.
  • Limitations. Prices are not independently verified, and they may not reflect prices on other venues.
Certain prices and derived valuations may be provisional, incomplete, or dependent on assumptions. Users should review applicable data-quality notices and methodology before relying on these values.
E

Coverage & Limitations

Coverage depends on what is currently indexed and supported. It is not complete coverage of Robinhood Chain or of all tokenized assets.

  • Supported categories and datasets. See the dataset catalog and the coverage status on the Data product page.
  • Supported venues. Trades are decoded for supported venues only. Contracts and protocols without a supported adapter are not decoded.
  • Historical availability. Balances and holder data reflect the indexing coverage start.
  • Token metadata. Metadata, including decimals, may be unknown for some tokens.
  • Missing or partial data. Unknown values are shown as unavailable. Incomplete balances are flagged and not corrected silently.
  • Observed versus total activity. Figures describe activity that has been indexed and decoded. They are not total network activity.
  • Query limits. SQL results are time-limited and capped at 10,000 rows. A capped result is marked as such.
F

Data Quality Controls

The Data documentation describes these principles for handling quality. They describe how values are treated, not guarantees of accuracy or completeness.

  • Unknown stays unknown. Values that cannot be established are returned as unavailable, never as guesses or defaults.
  • Corrections are appended. A correction creates a new revision. Conflicting payloads are quarantined, and a served row is not silently overwritten.
  • Venues are enabled after testing. A trading venue is enabled only after its adapter passes tests.
  • Evidence and coverage records. Coverage, runs, publications and open issues are kept as records, and values trace back to their source block and revision.
  • Stream position. The contiguous-complete position is kept separate from the highest block seen.
Not documented publicly at this time: treatment of duplicate records, automated anomaly detection, and a formal methodology-versioning policy. No procedures are claimed for these.

Amonhen is an independent data and forecasting platform and is not affiliated with, endorsed by, or sponsored by Robinhood Markets, Inc.