Method

Definitions are versioned. If one changes, the readings taken under the old definition are kept rather than recomputed, and the change is recorded here with the date it took effect. Nothing is revised silently.

Mission Ratio

The share of Robinhood Chain that is tokenised equity. Numerator is the sum across every canonical stock token of multiplier adjusted supply times price. Denominator is Robinhood Chain TVL as reported by DefiLlama for chain ID 4663.

We compute the numerator ourselves rather than quoting a third party. Supply is read from totalSupplyUI(), never totalSupply(). After a corporate action those differ and the raw figure is wrong.

Other publications have used a different TVL denominator, so our headline will not always match a number you read elsewhere. We use DefiLlama because it is free, stable and independently reproducible.

Robinhood has deployed 203 stock tokens. We know that because we read the deployment events emitted by the issuer's own factory contract, rather than asking an indexer to search for them. Until 20 August 2026 we did the latter, that search paginates in a loop, and it is ordered by market cap, so the tokens it never reached were the small ones. We reported 95 and the true figure was 203. The count on this site now comes from the chain.

A sum over some of the tokens divided by the whole chain understates the ratio by exactly the share we could not price. That is acceptable only if the shortfall is stated, so the rule is a floor plus a disclosure. The floor is weighted by holders rather than by token count: the Mission Ratio is published only when the tokens carrying a price account for at least 95 per cent of all holder positions, and wherever the figure appears both the token count and that percentage appear with it.

Weighted by holders because a count is misleading here in the honest direction. 95 of the 203 tokens carry a price on our source, which sounds as though half the figure is missing. The tokens without a price hold about one per cent of all holder positions between them. A count floor would silence a working metric over that tail; a holder-weighted floor still goes quiet the moment a genuinely used token loses its price, which is the case the rule exists for.

Nine of the 203 are not indexed by our data source as tokens at all. We keep them in the count anyway, because the alternative is a figure that describes an indexer's coverage rather than what Robinhood issued. No price is ever estimated, carried forward or substituted to close a gap.

Inputs also have a maximum age. The TVL denominator must have been observed within six hours, and every token supply and holder reading within twenty four. A source that goes down produces a gap in the series, never yesterday's number republished as today's.

Corporate actions

Two figures are published and they are not the same thing. A scheduled action means the token reports a non-zero effectiveAt and a newUIMultiplier that is not the multiplier already in force: a corporate action is announced onchain but has not applied yet. We test both, not just the timestamp, because a chain that keeps the past timestamp after an action applies would otherwise leave the action showing as scheduled forever. A moved multiplier means the stored uiMultiplier is no longer 1.0, so the action has applied and effective shares have changed. Every multiplier we show is the value read from the chain. We never assume a baseline.

Price sources, and why there are two

The Mission Ratio is built from Blockscout's reported exchange rate, and only from that. Volume, liquidity, the 24 hour move and the pool price shown on each token page come from onchain pools via DexScreener. Every stored price row records which source it came from.

The two disagree. Across the 58 tokens both sources priced on 19 August 2026 the median gap was 4.1 per cent, and the largest were 35 per cent (LLY), 28 per cent (SNDK) and 23 per cent (SMCI). That is not noise and we do not hide it.

We do not blend them, average them, or fall back from one to the other to fill a gap. A ratio built from a mixture of two sources that disagree by a third on some inputs is a figure nobody can reproduce from either source. So the Mission Ratio stays on one source and its coverage rule handles the gaps, while the market figures stay on the other and are labelled as such wherever they appear.

Migration to the per asset Chainlink feeds is planned. If it happens the readings taken under the current definition are kept rather than recomputed.

Trading pools

Volume and liquidity are summed across pools where the token is the base asset. A pool where it is the quote does involve it, so this understates rather than crediting another asset's trading to this one. Pool price is the single deepest pool by liquidity, never an average across pools, because an average is a number that exists at no venue and cannot be checked against one. The venue is recorded.

Our source returns at most thirty pools per token. Seven tokens were at that ceiling on 19 August 2026, so for those the volume and liquidity figures are floors rather than totals.

A token showing "no market" has no pool we can see at all. It is a reading, not a missing value: 36 of the 95 canonical tokens are in that state, holding real supply that cannot be traded.

Holder counts

Holder counts are reported per token. They must not be added together. A person holding three stock tokens appears three times, so any total we publish is labelled "sum of per-token holder counts" and is not a count of people. Unique holders are not yet computable without an indexer, and we do not estimate.

What we do not publish

Trade volume. A transfer proves custody moved, not that a trade happened. Proving a trade means pairing the stock leg and the payment leg by transaction hash, which is not yet built.

When a source fails

The affected figure is omitted and nothing is posted. We never substitute an estimate, carry forward a stale value, or fill a gap.