AJ Towns published statistics on transactions with OP_RETURN outputs between 2025-09-22 and 2026-02-11.
He finds that:
396 out of 24,362,310 txs (0.0016%) with OP_RETURN outputs had more than 83 bytes of OP_RETURN data. These “large OP_RETURNs” accounted for 2,093,184 of 473,815,552 bytes (0.44%) of all published OP_RETURN data in the period.
Interesting data. It’s wild to see that such a tiny fraction (0.0016%) of transactions is responsible for nearly 0.5% of the total OP_RETURN data volume. This really highlights the ongoing debate about 'blockspace as a scarce resource.' Do you think these 'large OP_RETURNs' are mostly protocols like Ordinals/Runes, or just inefficient data embedding? Thanks for sharing these stats
I haven’t looked at the transactions in question, but Runes are fairly compact (usually even fitting into 40 bytes or less) and Inscriptions have nothing to do with OP_RETURN outputs, so presumably it’s images, text, or other data embedding.
So, the fears of the RDTS folks are unfounded?
The most charitable word would probably be "unrealized".
I was thinking of building things like this and versionbits monitoring into
getblockstatsto not have to parse headers / blocks iteratively all the time but just use the index.man 24.3kk TXS in 4.5 months. Bitcoin is beautiful ay?