Floors are thin
In a market this illiquid, a floor can be set by a single seller. The annual average dampens this but does not remove it.
Read the direction, not the decimal.
Standing work · Method
How the index is compiled
What it measures, what it deliberately excludes, where the numbers come from, and the limitations we know about. Published so the index can be checked rather than believed.
Primary data source: CoinGecko Research, metaverse land prices. Every figure in the index is reproducible from that source using the rules below, without our cooperation.
As at 2026-08-28In one paragraph
We report the annual average floor price for the primary land token of each tracked world, in ETH, sourced from published marketplace data. Peak is the highest annual average since the world launched; latest is the most recent complete year. We do not model, smooth, weight by district, or convert to fiat.
§ 01
Inclusion
| Rule | Why |
|---|---|
| The world must have a transferable land token whose scarcity the platform enforces. | Licensed land inside a publisher-run world is not comparable — it cannot be resold, so it has no floor to measure. |
| At least twelve months of continuous marketplace data. | A world with three months of trading has a price, not a series. |
| One primary token per world. | Estates, districts and bundles are derivative. Mixing them into one average produces a number that describes nothing. |
| ETH denomination. | Fiat conversion imports a second volatile series and obscures what the land did relative to the asset most holders actually measure against. |
§ 02
Known limitations
Stated here rather than discovered by a reader, because an index that hides its weaknesses is a marketing asset.
01
02
District is invisible
A world-level average blends a prime parcel with unbid outlying land. Within Decentraland the spread between them is far larger than the drawdown.
Per-district series are the next addition.03
Wash trading
Marketplace data includes transactions between related wallets. We do not currently filter these, and no public source does so reliably.
Treat volume with more suspicion than price.04
Survivorship
Worlds that failed entirely leave the series. The aggregate therefore understates the losses an index-tracking buyer would have taken.
The real number is worse than −72%.05
We hold some of this land
Metaverse Group holds positions in Decentraland — 165 parcels including the Fashion Street estate — and in Somnium Space. We have an interest in these numbers.
Stated so you can weigh it.06
Annual granularity
Annual averages are the most recent complete figures available from published research. Monthly series require primary marketplace collection.
On the roadmap, not yet claimed.
The conflict, stated plainly
We publish an index for an asset class we hold — currently Decentraland and Somnium Space. That is a real conflict and disclosure does not dissolve it. What it does do is let you check us: the source is linked, the method is above, and every figure can be reproduced from public data without our cooperation. If we ever publish a figure you cannot reproduce, that is the point at which to stop trusting the index.