7

Looking at this chart: http://blockchain.info/charts/estimated-transaction-volume

It's clear that the amount of bitcoins moved between November 14th and December 12th of 2011 was much larger than any previous month, even compared to the big boom in September.

The number of transactions is about the same as previous months. This implies a few big transactions or generally large transactions. What is the reason for this?

5

I'm not sure the volume actually did increase so much in November and December.

The actual amounts transacted each month were:

Jan 31 2011   3,679,198.19243835
Feb 28 2011   4,533,818.22264172
Mar 31 2011   9,331,953.39257620
Apr 30 2011  13,893,407.03577082
May 31 2011  11,850,890.09316489
Jun 30 2011  73,031,161.99932121 <-- biggest
Jul 31 2011  37,718,094.01889341
Aug 31 2011  17,348,635.19178118
Sep 30 2011  17,616,839.59721035
Oct 31 2011  12,008,040.32522139
Nov 30 2011  49,968,992.05725216
Dec 31 2011  46,978,824.26111485
Jan 31 2012  11,664,075.33329784

The blockchain.info graph appears to ignore coins which are moved more than once in the same block.

To see that the blockchain.info graph is ignoring some volume, consider block 133233. That's just one of the >100 blocks from 25th June, and it alone contains transactions worth 2,420,580 BTC. However, the blockchain.info graph shows that no single day in June had a volume over 1 million BTC. Scrolling to the end of block 133233 shows a whole page of small payments from a 46k BTC address, which the blockchain.info graph ignores.

According to the blockchain.info graph, the biggest day in December involved transactions worth 5.834 million BTC. There was only one day in December with that much volume, and that was December 6th, with 6.903 million BTC. On December 6th, the volume was dominated by a 440k coin being moved 15 times, accounting for 6.601 million of the day's 6.903 million volume, so the blockchain.info graph must be taking 13 of these 15 movements into consideration. The 15 movements occurred in 13 blocks - the coins move twice in each of blocks 156242 and 156346.

The origin of this huge coin was on November 16th, ten 50k coins were combined into a single 500k coin. The coin moved around a lot from address to address until 8th December 2011, when it was broken up into hundreds of small pieces.

Edit: see this graph of "bitcoin days destroyed", which ignores rapid repeat coin movements by weighting transactions by the amount of time since the inputs were created. It mirrors the data I pasted above, showing that June was a bigger month than November.

5

Could it be Christmas shopping? People who have thick bitcoin wallets, needing dollars to buy Christmas presents? Many online retailers make half their money by selling Christmas presents.

2

Let's say I have just one address in my wallet with 1,000 BTC on it. I then send to you five separate transactions, each in the amount of 1 BTC. The blockchain will then show 4,990 BTC of transfers yet only 5 BTC was truly transferred.

Additionally, if I were to begin using the encryption feature in the latest Bitcoin.org client, my coins will get moved to new addresses -- thus possibly contributing to the volume spike you saw.

  • Has it not always been the case that the "change" part of a transaction is indistinguishable from the "sent" part? The question is what's different about December. – jl6 Jan 4 '12 at 22:47
  • @jl6 indeed this has always been the case. Even if it weren't and 0.5 did introduce it as a new feature, we'd expect to see a sustained burst of volume, whereas the chart shows about a two-day spike and a return to normal volumes. I'm sure there's one gigantic transaction to blame, if someone with WAY more time than me wants to check BlockExplorer. – David Perry Jan 4 '12 at 22:51
  • 1
    @DavidPerry Mt.Gox did a bunch of big transactions like this at the time. (You can follow it back to the 424242 BTC transaction they did to prove they were still in control of the funds after the hack last summer). – jarpiain Jan 20 '12 at 22:46
  • @jarpiain you should write that into an answer since it would appear to explain this phenomena and I'd like to see you get credit for your answer. – David Perry Jan 20 '12 at 23:19
  • I think the maths is wrong in your answer. Also I think you mean "if I had just one transaction output in my wallet", since it's possible to have multiple separately spendable outputs at the same address. And then the first would move 1000, the 2nd 999, etc. 1000+999+...+996 = 4990. – Chris Moore Apr 1 '12 at 17:05
0

One reason could be that Mt.Gox cut their commissions in half from Dec 25th to Dec 31st.

  • 1
    Nope, Transactions are between November 14th and December 12th – Kristoffer Nolgren Jan 9 '12 at 15:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.