On April 16, 2019, Sergio Demian Lerner published “The Return of the Deniers and the Revenge of Patoshi,” a major update to his ongoing blockchain forensics research that consolidated six years of findings and introduced the now-canonical “Patoshi” terminology.
The “Patoshi” name:
Lerner coined the term “Patoshi” — a portmanteau of “Pattern” and “Satoshi” — to refer to the distinctive mining signature he had been studying since 2013. The name became the standard reference in all subsequent academic and community discussions of Satoshi’s mining behavior.
Updated estimates:
- ~22,000 blocks attributed to the Patoshi pattern (within a larger set of 27,680 nonce-restricted blocks, called “set M”)
- ~1.1 million BTC — revised upward from the original 2013 estimate of ~1 million BTC
- 99.9% of Patoshi blocks remain unspent, versus only ~10% of other early blocks
Timestamp inversion analysis:
The most powerful new evidence was a timestamp analysis of the first 50,000 blocks. Lerner found:
- Zero timestamp inversions between consecutive Patoshi blocks
- 224 timestamp inversions among consecutive non-Patoshi blocks
A timestamp inversion occurs when a later block has an earlier timestamp than its predecessor — this happens when different miners’ clocks are slightly out of sync. The complete absence of inversions in Patoshi blocks proved they were produced by a single PC clock running a single piece of software, making the hypothesis of multiple synchronized miners statistically impossible.
Statistical proof:
Lerner calculated the probability that all 27,680 blocks in set M could have nonces randomly falling within the restricted LSB range R: less than 2^-36,000 — a number so vanishingly small as to constitute mathematical proof that the pattern was intentional.
Conclusion:
This 2019 paper transformed the Patoshi research from a set of empirical observations into a near-certainty: one entity, using custom mining software on a single computer, accumulated approximately 5% of Bitcoin’s total supply — and never spent it.