**TL;DR: As a result of analyzing the "Mt.Gox leak", the number of bitcoin holders is likely 1.0 millions. Of these,**

* the 400 bitcoin millionaires with at least BTC2,000 hold 40% of bitcoins.

* 100,000 owners with BTC10-BTC2,000 ($5,000-$1,000,000) hold 50% of the bitcoins.

* 900,000 owners have a holding of BTC0.002-BTC10, valued at $1-$5,000, and hold 10% of the bitcoins**This is NOT a ***largest addresses/wallets -list*, we are talking about largest owners here. An owner can have many wallets, and a wallet can have many addresses.

**22. Apr 2014**

#People | #Bitcoins | #TotalBitcoins |

70 | BTC10k+ | 3.6M |

930 | BTC1k-10k | 2.2M |

13k | BTC100-1k | 3.0M |

85k | BTC10-100 | 2.3M |

250k | BTC1-10 | 0.8M |

340k | BTC0.1-1 | 0.1M |

230k | BTC0.01-0.1 | 0.0M |

90k | BTC0.002-0.01 | 0.0M |

Total:

**12.2M bitcoins** (0.5M bitcoins assumed lost)

The changes from previous estimate:

- Total number of holders revised down from 2.0 million to 1.0 million. This did not change much in the distribution of coins. If new users are added, the large holders do sell according to an experimentally derived rate of 17% per doubling in price, but some newish holders also add to their positions. All-new entrants can come to any bracket, because no amount of coins is yet too expensive to buy. The perceived static totals of the analysis hide the fact that Bitcoin trade is quite brisk and the people who consist of a bracket do change all the time.

- Number of 10k+ holders recalculated because we now know better the average number of coins in each bracket (thanks to Mt.Gox data). So the large holders still have the same number of coins in total, but there appears to be more of the holders.

- Same for 1k+ and 100+. More people but approx the same number of total coins.

- 10+ category turned out to be larger than previously estimated. The Mt.Gox data gave us the confidence to use a steeper

*j* parameter, which means a more "equitable" distribution with more coins in the middle brackets.

- 1+ category remained intact, and the fractional categories were significantly sized down. The lower threshold of "owner" was also raised to

BTC0.002 because of the decline in price.

- Number of Bitcoin millionaires at this price ($500) is about 400. This means that only 400 people worldwide can raise a $million by selling bitcoins, but about 30,000,000 people can invest a $million to buy bitcoins (without going into debt).

- The Bitcoin millionaires command about 4.9 million bitcoins, which is 40% of all bitcoins.

*The control of world's gold and fiat currency is more centralized with a smaller number of individuals controlling higher percentage.***Earlier estimate of 27. Jan 2014**

#People | #Bitcoins | #TotalBitcoins |

46 | BTC10k+ | 3.6M |

820 | BTC1k-10k | 2.4M |

10k | BTC100-1k | 2.8M |

68k | BTC10-100 | 2.0M |

250k | BTC1-10 | 0.8M |

550k | BTC0.1-1 | 0.2M |

690k | BTC0.01-0.1 | 0.0M |

430k | BTC0.001-0.01 | 0.0M |

Total:

**11.8M bitcoins** (0.5M bitcoins assumed lost)

There is less holders than before in the category of

BTC70+. The number of holdings smaller than this increased.

The number of bitcoin millionaires, previously estimated 930, decreased to 790 (-15%). Their combined wealth is 5.9M bitcoins, which is 50% of all bitcoins.

**Earlier estimate of 3. Dec 2013**

#People | #Bitcoins | #TotalBitcoins |

47 | BTC10k+ | 3.5M |

880 | BTC1k-10k | 2.6M |

10k | BTC100-1k | 3.0M |

63k | BTC10-100 | 1.8M |

210k | BTC1-10 | 0.6M |

350k | BTC0.1-1 | 0.1M |

350k | BTC0.01-0.1 | 0.0M |

210k | BTC0.001-0.01 | 0.0M |

Total:

**11.6M bitcoins** (0.5M bitcoins assumed lost)

**1,200,000** people own bitcoins that are valued at more than

**$1** (

BTC0.001 or more)

**280,000** people own bitcoins that are valued at more than

**$1,000** (

BTC1 or more), up from 120,000 last month

**930** people own bitcoin that are valued at more than

**$1 million** (

BTC1,000 or more)

**1** person owns bitcoins that are valued at more than

**$1 billion** (Satoshi,

estimated to be BTC980,000)

Changes from previous month:

- Bitcoin total number adjusted to account for increase in money supply

- Thoroughly re-engineered the number of bitcoin users from original data of the

userbases of several services +

blockchain usage. This shows that the last month estimate was probably too small, and perhaps 50% of the growth in number of users can be attributed to correction of a previous underestimate.

- Corrected the highest bracket a little bit (

BTC0.6M

) upwards because it was over-corrected downwards last time. Number of holders in this category went down a little but we have better estimate of the holdings. As this is a base case model, we do not assume large, completely anonymous entities without at least circumstantial evidence of their holdings.

- Number of people in each bracket behaves according to binomial distribution as found in another data with the parameters: total=1.2 million users, highest bracket=fixed, #of_coins=fixed.

As a result:

- Bracket 1,000-10,000 saw the greatest loss -20%, bracket 100-1,000 lost 10%. Sub-

BTC10 holdings gained a lot.

Earlier estimate of 26. Oct 2013

#People | #Bitcoins | #TotalBitcoins |

50 | BTC10k+ | 2.9M |

1100 | BTC1k-10k | 3.1M |

12k | BTC100-1k | 3.3M |

58k | BTC10-100 | 1.8M |

110k | BTC1-10 | 0.4M |

110k | BTC0.1-1 | 0.1M |

57k | BTC0-0.1 | 0.0M |

Total: 11.5M bitcoins, 350k owners

120,000 people own bitcoins that are valued at more than $1,000 (

BTC5 or more).

Changes from previous estimate:

- Bitcoin total number adjusted to 11.5M to account for lost coins

- Significantly reduced the number of bitcoin users to account for the assumption that few people would play with stashes that are <$10 in value, and other evidence pointing to a lower number

- Significantly reduced the number of bitcoins owned by the highest bracket due to the lack of evidence that there are many large holders. As this is a base case model, nefarious premine schemes are not assumed without proof. Sorry for falling into that initially.

- Number of people in each bracket behaves according to binomial distribution as found in another data.

Thread is self-moderated to retain readability; user content will be amalgamated to the table and possibly deleted. If this happens, don't take it personally.