AFAIK it crashed once already, from ~5000 all the way down to 1, is that right?

(I came in late, the second bubble was already underway.)

Er.... from the comments in that reddit thread, I understood that the transaction attempted by MtGOX was not a test after all, just one of many small transactions that they have been doing since the lock-in.  Did I understand wrong?

Perhaps the news "MtGOX is testing!" spread further and farther than the "Oops, soory"?

Let's see whether I got it right: someone noticed that MtGOX made a tiny transaction, which actually was like several others that they had been doing all along since the lock-in; a transaction which had nothing to do with withdrawals, and moreover was rejected by the network because it contained mined bitcoin that was too fresh -- a bug that MtGOX's software was known to have, and which apparently has not been fixed yet.  That is why the market suddenly got excited and optimistic, and the price jumped up by 10% in a few hours . Correct?

My guess, based on the amount of transactions that were stuck until recently (~40,000), is that the sum N of the BTC balances in all of MtGOX accounts (the total number of goxBTC) is at least 200,000 goxBTC.

I would guess that the sum M of of the national currency balances in all of accounts (the total number of goxUSD, goxGBP, etc.), has about the same value as N times the market price outside MtGOX (~600 USD); that is, M is equivalent to at least 120 million goxUSD.

By these guesses, MtGOX owes at least 240 million USD to their clients, in BTC or national currencies.

Are these guesses reasonable? Too low perhaps?

Today (Saturday, Feb 22 UTC) volume was lower than yesterday in all exchanges.

Outside China and MtGOX, volume fell from ~84  to ~54 kBTC, which is a little below the daily average for last week.

In China, volume fell too, but not as much: from ~344 to ~252 kBTC, which is well above last week's average.

As a result, China's fraction of the total volume (excluding MtGOX) increased a little, from ~80% to ~83%.

Daily volumes of BTC trade to/from USD and other national currencies (in kBTC):


             !    Fri !    Sat !    Sun !    Mon !    Tue !   Wed !    Thu !    Fri !    Sat !
  EXCHANGE   !  02/14 !  02/15 !  02/16 !  02/17 !  02/18 ! 02/19 !  02/20 !  02/21 !  02/22 ! Currencies considered

  BTC-e      |  52.18 |   9.16 |  21.75 |  15.07 |   8.08 |  5.84 |  23.37 |  23.64 |  18.68 | USD,EUR,RUR
  Bitstamp   |  63.38 |  20.76 |  26.40 |  19.90 |  14.83 | 20.76 |  34.97 |  29.61 |  18.08 | USD
  BitFinEx   |  53.68 |  10.83 |  16.75 |  15.69 |   8.47 |  4.71 |  25.27 |  28.46 |  14.73 | USD
  Kraken     |   1.79 |   0.41 |   0.91 |   0.89 |   0.51 |  0.38 |   1.60 |   1.15 |   1.08 | EUR
  Bitcoin.DE |   1.63 |   0.22 |   0.57 |   0.57 |   0.38 |  0.54 |   0.70 |   0.76 |   0.61 | EUR
  CaVirtEx   |   1.16 |   0.15 |   0.15 |   0.15 |   0.21 |  0.11 |   0.75 |   0.60 |   0.20 | CAD
  CampBX     |   0.41 |   0.05 |   0.07 |   0.28 |   0.10 |  0.14 |   0.73 |   0.24 |   0.26 | USD

  SUBTOTAL   | 174.23 |  41.58 |  66.60 |  52.55 |  32.58 | 32.48 |  87.39 |  84.46 |  53.64 |

  Huobi      | 236.27 | 122.80 | 110.57 | 130.41 |  74.04 | 34.12 | 131.14 | 187.12 | 154.12 | CNY
  OKCoin     | 147.26 |  51.45 |  63.63 |  63.05 |  51.46 | 26.29 |  53.49 | 140.75 |  87.71 | CNY
  BTC-China  |  24.87 |   7.55 |   9.08 |   7.86 |   3.86 |  2.04 |  10.97 |  15.23 |   9.29 | CNY
  Bter       |   1.54 |   0.74 |   0.56 |   0.48 |   0.35 |  0.37 |   0.77 |   0.63 |   0.48 | CNY

  SUBTOTAL   | 409.94 | 182.54 | 183.84 | 201.80 | 129.71 | 62.82 | 196.37 | 343.73 | 251.60 |

  TOTAL      | 584.17 | 224.12 | 250.44 | 254.35 | 162.29 | 95.30 | 283.76 | 428.19 | 305.24 |

  MtGOX      |  79.56 |  60.15 | 104.46 |  65.69 |  63.50 | 40.75 | 127.48 | 103.06 |  89.24 | USD,EUR,GBP,AUD,JPY



All numbers were collected by hand from the site http://bitcoinwisdom.com. Beware of possible errors.

For each exchange, the numbers include only the trade volume to/from the currencies listed in the rightmost column. Trade between BTC and other cryptocoins, such as LiteCoin, is NOT included.

Dates on the header line are UTC. Specifically, "01/15" means "from 01/15 00:00:00 UTC to 01/15 23:59:59 UTC". (Beware that Bitcoinwisdom uses your local time, so the date may appear to be off by 1 day.  For example, if you are 2 hours west of Greenwich, it may show "01/14 22:00" when the UTC time is "01/15 00:00".)

NOTE: MtGOX has suspended withdrawals of Bitcoins and national currencies.  Therefore it is essentially isolated from other markets, and its price is completely out of the norm.  The exceptional trade volumes above may be meaningless.

Thanks for the reply.  Well, I am surprised. 

I would never deal with someone if I know that he has scammed someone else,  I would not think that I am smart enough to outsmart him.  I guess it is a matter of temperament, like aversion to risk.

I think that it is very dishonest (to use an euphemism) to allow deposits when there is no date for the resumption of withdrawals of either BTC or USD.

New investors, who have not read these forums, are likely to be attracted by the very low price, deposit money there, and then discover that they cannot get their BTC out nor their money back, until god knows when.

When an internet store does that sort of thing, don't we call it a scam site?

So, no one dares to guess how much ordinary gox-money (M) and how many goxBTC (N) are in MtGOX's client accounts?  Not even how many figures those numbers may have?

Considering the amount of BTC widthdrawals that were pending last month (around 40,000), I would say that N is at least 200,000 goxBTC, and M is at least 100 million goxUSD.

Then MtGOX owes at least 200 million dollars (in whatever form)  to their clients.

Are these estimates reasonable? Too small?

Knowing how many BTC MtGOX actually owns (X) is only 1/4 of the data one needs.

One would also need to know how much money (USD, JPY, etc.) MtGOX has in their bank accounts (Y).

Then one would need to know how much MtGOX owes -- that is, the total of their clients' account balances, in national currencies (M) and in BTC (N). 
These are not real coins and dollars, of course, but virtual "goxBTC" and "goxUSD" that are converted to their real counterparts only at deposit or withdrawal time.

One would also need to know any subtantial debts they may have: loans, payments due for large coin purchases (or the reverse), etc.  But probably these are smallcompared to M+N.   ("Market cap" is not the right name for M+N, is it?)

MtGOX would be insolvent if M + N (both converted to a common unit, say USD) were much larger than X + Y (ditto).  (For these conversions, one should use the market price outside MtGOX, of real BTC in real USD.)

I do not see how one could estimate M and N without an independent and thorough audit of their internal ledgers.  Did MtGOX ever provide such numbers?

Transactions between MtGOX clients exchange goxUSD for goxBTC; they show up as trade volume, but not in the blockchain.  Some goxBTC and goxUSD may not have been traded at all in months, some may have been traded many times in that interval.   Therefore, the total trade volume in any period may be much smaller or much larger than M + N.

Er, that 2/3 (now more like 1/2) is not my estimate, it is what their clients apparently believe, given the price that they agreed at MtGOX.

Well, I can think of three ways in which MtGOX could have lost a large part of their coins (and theft is not one of them).  But there may be others.

"Too big to fail" has been said of many companies which failed soon after.  

The reason I suspect that MtGOX is insolvent is because Mark is doing everything that one expects from the CEO of an insolvent company: stalling and delaying payments, communicating only through vague press releases widely spaced  apart, avoiding clients, blaming "the system", moving to a high-security building...

And he is not doing any of the things that he should and could do if MtGOX were not insolvent and he cared about its business and the health of bitcoin. Like, proving that he controls a large stash of coins, giving explicit numbers (even if approximate) about their financial situation, bringing in some competent accountants and programmers, giving interviews about the crisis, checking and processing by hand some large withdrawals...

If MtGOX is not insolvent, Mark is feigning it extraordinarily well.


Good question.  Perhaps because few people heard about it, and there was no follow-up?

EDIT:markup
 

Same as banks and governments then
[/quote]

Governments are usually bought or held hostage by banks, so their "tolerance" is understandable.

Common folk who get screwed by a bank will usually curse it and move to another bank.  But that does not seem to impress their acquaintances much, they will usually continue using that bank even after being told of the incident.

So, indeed, perhaps the problem is not specific to bitcoin exchanegs and their clients...

Well, it isn't exactly a "small" Latin American country 

The lack of legal remedies is a complicated issue.  On one hand, that country has laws and a fairly competent police section specialized in inernet crimes, so methinks that the necessary instruments are there.  On the other hand, those bitcoin sites probably work without contracts and receipts, so clients may not be able to prove that they sent in the bitcoins in the first place.   (The blockchain could be used to prove the transfers, if the addresses coud be tied to the persons; but there still would be no evidence of why they were transferred.)

But my question was not about legal redress, but rather why people who know about that incident (including some of his victims) still choose to use his exchange, and leave their money and coins there.

That situation seems similar to that of MtGOX and Bitcoinica; at least, to the extent that the clients of the closed site are still demanding their lost coins from  the exchange owner, without suiccess.  (Is this description correct?)

If, when they locked the exit gates, the sum of all their client BTC accounts was X, then, in spite of all the trading that has happened there, they still owe X-6000 BTC to all their clients.  (Unless MtGOX itself has been buying BTC from their trapped clients, at their current panic price; which would be the closest think to armed robbery that they could do.)

Their trapped clients seem to think that Y, the amount of real BTC that MtGOX currently owns, is about 1/3 of X-6000.  Although that perception may change rapidly on faint echos of rumors.

Anyone dares to guess what was the value of X (basically, the total supply of goxBTC)?  (Note that the total volume traded at MtGOX over the last two weeks may not be a reliable estimate, since there may be many goxBTC that were not traded once in all that time, and many that got traded several times.)

I would nominate Mark Karpeles to the Oscar for his superb acting in the movie /Honey, I Shrunk the Coins/.

Seriously though: by allowing deposits while suspending withdrawals, MtGOX became a scam site in my book.

Any sensible person would flee from MtGOX at the first opportunity.  Yet by posts in this and other threads, many still defend Mark and expect MtGOX to continue trading there once it resumes withdrawals.

I just learned a story about a 'bitcoin investment fund' in a certain Latin American country.  The fund took several thousand coins from clients,  promising large returns in BTC.  Then the owner claimed that the fund had been hacked an all the coins were stolen, and closed the fund.  The clients have not yet received their coins back.

What makes the story interesting is that that same person is the owner of that country's main bitcoin exchange.  And his clients see nothing wrong with that.

Why is there so much tolerance to dishonesty in the bitcoin community?

Does it have something to do with the libertarian philosophy, in some broad sense of the term? Perhaps those loyal clients think, "The guy is not a scammer, because anyone who lost money to his scam is stupid, and stupid people deserve to lose their money"?

Chinese Slumber Method prediction for Sunday Feb/23

Today Feb/22, Huobi's volume for 19:00 -- 19:59 UTC (3:00--3:59 AM China Standard Time) was 1742 BTC, which, by the criteria I have been using, makes it a False Slumber Point.  Therefore, I still have only one valid data point (3425 CNY at Feb/21) for the new trend line that presumably started after the Second Karpeles Catastrople of Feb/20.

I will therefore assume the same trend line used for the last prediction, namely the straight line that goes though that point and has slope minus ~29 CNY per day.  For Bitstamp, as usual, I will use Huobi's prediction divided by R = 6.12.

Therefore:

Prediction valid for: Sunday 2014-02-23, 19:00--19:59 UTC (not before, not after)
Huobi's predicted price: 3368 CNY.
Bitstamp's predicted price: 550 USD.



NOTE: The 95% confidence interval for both predictions is [ 0, +∞ ].

Checking the Chinese Slumber Method predictions for Feb/22

Prediction posted on: Saturday 2014-02-22, 01:58 UTC
Prediction valid for: Saturday 2014-02-22, 19:00--19:59 UTC

The prediction for today (Saturday Feb/22) was again way off the mark.  However, in the measurement interval (03:00 -- 03:59 China time) there was still significant activity (1742 BTC traded).  Therefore, by the rules of the game, that is a False Slumber Point, and the prediction should be considered void.

Indeed, that interval came a couple of hours after a small rally, perhaps triggered by rumors that MtGOX was testing withdrawal software.

Anyway, here are the numbers:

Huobi's predicted price: 3396 CNY
Huobi's actual price (L+H)/2: 3621 CNY
Error: 225 CNY

Bitstamp's predicted price: 555 USD
Bitstamp's actual price (L+H)/2: 597 USD
Error: 42 USD

The Huobi prediction in question is the rightmost blue rectangle on the chart below.  The light blue-gray rectangles are the previous predictions.  The orange and grey dots are the True and False Slumber Points, the mean prices at 19:00 UTC every day.  The orange line is the trend that was assumed for the prediction.


The following chart shows the Bitstamp prices and predictions. The orange and grey dots are Huobi's prices at 19:00 UTC every day, divided by R = 6.12.  The orange line is the trend used in the prediction, which is Huobi's trend divided by R.

Warning: more Chinese Slumber Method stuff to follow.  You may want to read my next two posts with your eyes closed.

You forgot the "".