Whale Alert: Satoshi Nakamoto holds 1.12 million Bitcoins worth $10.9 billion

Whale Alert released a report on Satoshi Nakamoto on July 20, which made the most accurate estimate of the number of blocks mined and bitcoins owned by Satoshi Nakamoto. Satoshi Nakamoto mined 54,316 Bitcoin blocks and held 1,125,150 Bitcoins, worth $10.9 billion.

In addition to transactions, other details related to the mining process, such as nonce and extranonce values ​​(used by miners during the mining process), are also added to the blocks. In 2013, Sergio Demian Lerner discovered that there was a clear pattern visible in this extranonce value in the coinbase transaction (the transaction that creates the reward bitcoins for mining blocks) of each block.

Patoshi Pattern

The graph above shows the outlink values ​​for blocks below 20,000, with each block represented by a dot and forming a distinct saw-shaped pattern. Lerner believes that these diagonal lines are created by miners using the standard, publicly available Bitcoin software: the straight lines they leave behind are the result of miners incrementing the outlink during their own mining process or each time they mine a new block on the network. The distinct saw-shaped pattern is created by resetting the outlink value to zero, which happens every time a miner restarts. The nearly vertical line in the graph is attributed to a miner that Lerner named Patoshi.

We know for sure that Patoshi was operated by Satoshi because its pattern was present at the dawn of the network, and it mined blocks that created the bitcoins sent to Hal Finney. Lerner found additional evidence for his claim in the nonces (another value used in the mining process, also stored in every block, but not on the blockchain) of blocks mined by Patoshi miners: the last byte of the nonce was always in the range 0 to 9 or 19 to 58, while all other miners used the full range of 0 to 255. This finding made it easier to attribute blocks to Patoshi because it dropped non-final digits outside the range. However, attribution of blocks remained difficult at the intersection of Patoshi and standard miner patterns, especially after 20,000 blocks, when the number of active miners increased significantly.

When adjusting for outlink ranges, the Patoshi pattern becomes clearer

We suspect that the limited range of the last byte of nonces is due to work being divided among a number of computers or CPU cores, each using a different number to prevent duplicate work. With this in mind, we checked whether the range itself changes as mining power is added or removed, and we found that it does change over time. For example, the range [0-9][19-58] was only used in the first 18,015 blocks, then reduced to [0-9][19-48] until block 21,308, and the same happened for subsequent block periods (the table below lists the ranges for each block period) (similar findings have been described in this anonymous blog). Our research also revealed some interesting new details: between blocks 21,467 and 25,777, the range [0-9] was only used on each Patoshi chain (each chain is a single run of the miner, hence the single line in the graph above), and the number 39 was only used sporadically. Between blocks 25,811 and 54,316, some chains lack the number 29 entirely. These anomalies could indicate that certain computers or cores were defective or shut down during these periods. These findings allow us to rule out more blocks that were not mined by Satoshi and provide us with a clearer picture (see figure above), which is especially valuable at higher block heights where mining activity on the network increases dramatically.

Patoshi nonce ending range

Improved Patoshi block attribution allows us to accurately determine the average mining speed of Satoshi's miners, and surprisingly, we find that changes in the nonce end range do not correlate with changes in mining speed: the average mining speed remains incredibly constant for a long period of time. For example, over the entire range of blocks between heights 2,000 and 16,000, the average number of blocks mined per hour by Patoshi is almost exactly 0.6 per 10 minutes, despite the large variation in the number of miners during this period. From the data, it can be seen that Patoshi miners adjust their speed between blocks to maintain the average level, and when one Patoshi chain creates more than 0.6 blocks per 10 minutes, the average block time of the next Patoshi block decreases, and vice versa. Satoshi wanted to maintain this average for two reasons: first, he believed that 51% attacks were the biggest threat to the growing network, and by maintaining 60% of the processing power, he could prevent these attacks from happening while leaving enough blocks for others to mine. As more "honest" miners joined the network, a 51% attack became less likely, and Satoshi was able to gradually scale back his mining activities. Second, Satoshi stated that the ideal block time was around 10 minutes, and that by controlling enough processing power, the block time could be artificially kept around this time when there was insufficient or excessive activity on the network. We speculate that Patoshi consisted of at least 48 computers, with one machine for coordination and more machines on standby in case of attack, which would explain the missing range of [10-18]. Once Satoshi believed the network was strong enough, he reduced Patoshi's block target of 10 minutes to give others a better chance of mining blocks.

Blocks mined every 10 minutes per Patoshi chain (over 20 blocks)

Knowing the changing nonce range allows us to attribute more blocks to Satoshi with greater certainty: an estimated 22,503 of the first 54,316 blocks mined. Of these, 50 had their coinbase transactions (or mining rewards) spent, including one that was definitely spent by Satoshi in a transaction of 10 BTC to Hal Finney. 31 of the spent blocks are likely fake, meaning they matched Patoshi by chance and belonged to a different miner. We are confident that 18 of the spent blocks belong to Satoshi, so the total amount spent by Satoshi is 907 BTC (the price per BTC was less than $0.01 at the time these transactions occurred), leaving 1,122,693 BTC unspent.

The goal of this research is not only to find out how much money was spent, but also why Satoshi mined in this particular way. Did Satoshi stop mining with the Patoshi miner after block 54,316? Whether the mining software was changed to become undetectable as a result, or whether Patoshi continued mining with the publicly available mining software, we have no way of knowing. Some anomalies have been found in higher blocks, such as a disproportionately high proportion of non-CES ending in [0-9] after block 70,000, and a strange pattern of using the same range between blocks 109,500 and 112,500, but it is currently certain that the Patoshi miner was shut down in May 2010. The timing of the shutdown, the mining behavior, the systematic reduction in mining speed, and the lack of spending strongly suggest that Satoshi only wanted to grow and protect the young network. The bitcoins mined by Patoshi may have been a byproduct of these efforts, and it is unlikely that the remaining bitcoins were spent, although the question is why Satoshi did not simply burn them in this case. Our findings do not rule out the possibility that Satoshi was also running a miner using publicly released software, even if only for testing purposes, and we believe that at least one of the non-Patoshi modes is likely to belong to Satoshi.

Note: According to our research, the following blocks have been mined and used by Satoshi: 9, 286, 688, 877, 1,760, 2,459, 2,485, 3,479, 5,326, 9,443, 9,925, 10,645, 14,450, 15,625, 15,817, 19,093, 23,014, 28,593, and 29,097.