Weeell, another day, another totally failed prediction by the Chinese Slumber Method. But, I had a new epiphany!

Checking the prediciction for Huobi

Prediction posted: 2014-02-17 02:27 UTC
Prediction valid for:  2014-02-17 19:00--19:59 UTC
Huobi's predicted price: 3600 -- 3660 CNY (3630 plus or minus 30)
Huobi's actual price (L -- H): 3890 -- 3919 CNY (3905 plus or minus 15)
Error center-to-center: 275 CNY

The prediction is the rightmost blue rectangle on the following plot, just below the price bars.  It brings back memories of being a small child and unsuccessfully trying to catch a chicken.   And of being a grad student and unsuccessfully trying to solve this math problem.  But read on!



The epiphany: True Slumber Points

As before, the large dots over the price graph in the above image are the prices at the "Chinese Slumber Times", every day at 19:00 UTC (3:00 am in China).  This time, however, the plot distinguishes the True Slumber Points (yellow), when there really was no trade at Huobi, from the False Slumber Points (olive-brown), where some of the clients (or robots) continued to trade nonstop through the night.  

As you can see,  all five True Slumber Points after the First Karpeles Catastrophe (on Monday Feb/10)  lie very close to  straight trend line; whereas the three False points seem to be far more random.  That observation gives new hope for the Method.

Checking the prediction for Bitstamp

Recall that the prediction for Bitstamp is just the Huobi prediction divided by the currency factor R = 6.12.  

Bitstamp's predicted price: 580 -- 600 USD (590 plus or minus 10)
Bitstamp's actual price (L--H): 630 -- 638 USD (634 plus or minus 4)
Error center-to-center: 44 USD

In the image below, the yellow and brown dots are Huobi's prices at the Slumber Times, divided by R (6.40 for feb/07--feb09, 6.12 for all other days)

What is wrong with having an epiphany at lunch, grammatically or philosophically?  I had quite a few myself.

On the other hand, I think that having an epiphany for lunch would be unwise, nutritionally.  One would feel very hungry by 3pm. 

I wonder how many bitcoin buyers have been lured by MtGOX's extremely low price, and discovered only too late that they cannot get their coins out nor their money back.

Mark Karpeles is a human being, sure.  But there are several more specific terms for a human being that receives the property of other human beings for safekeeping, and refuses to give it back. 

It was claimed that they need to limit the amount of BTC withdrawals because their coins are in "cold storage" and their hot wallet is small.

But even if their "cold storage" is a laptop hidden in the catacombs of Paris, they should be able to get it in a couple of days, move it to a closer (but still safe) location, and then keep transferring enough coins to the hot wallet, every day, to execute all the withdrawal requests on the queue.  What is the risk of that?

Thanks for the reply.  Indeed that is what they would have to do in the "best case" assumption that the only problems were the hack and their incompetence at handling it.

My guess is that (because of the hack or some other reason) they know that they do not have enough bitcoins or cash to cover all the legitimate client account balances, so they may have to do a "haircut" on them.   That indeed would require a full auditing before any further withdrawals.

Indeed it is true, when I started posting on forums (internal mailing lists then) we had no emtoticons; but that did not prevent me from doing that mistake. 

And you can imagine how immensely needed was that invention.  I hope that bitcoin can be at least half as succesful as "". 

But that should not be the case of people who are publicly demanding their coins and/or dollars, right?  Like Rick Falkvinge, or those people who went to protest in front of MtGOX's offices.

When Rick requested a BTC withdrawal, the transaction issued by MtGOX must have pointed to his external wallet, and the hack would not change that; all that the  hacker could do is trick MtGOX into crediting back the BTC to Rick's account, in spite of the transfer haviing occurred.   But Rick and all those other clients apparently did not get his BTC.  So what happened in those cases?

(And note that, AFAIK, Rick is still a fan of bitcoin, in spite of his problem with MtGOX.)
 
EDIT: typos

Yes. I have been posting to forums for some 35 years, and I still keep making the same mistake, over and over again: thinking that readers will recognize blatant sarcasm even without the " "

You already know that I am totally ignorant on how bitcoin works, and too old to learn; so forgive me for asking:

* How many weeks does it take for the network to process a transfer of a few thousand bitcoins from MtGOX's cold wallet to MtGOX's hot wallet? 

* How many such transfers from the same wallet can the network handle per month?

* If bitcoins were transferred into a wallet using certain custom software, can they be transferred out using a different software?

* If so, how much would MtGOX have to pay for a software that they could use to process the pending withdrawals, while they fix theirs?

From what I have read, Mark is holding at least 40,000 BTC that do not belong to him, whose owners have been trying to withdraw for months. 

Does that make any sense?

I suppose that the market was afraid of another blooper from MtGOX, like last Monday's, and relaxed when they saw that MtGOX just spun their clients around for a few more days.

The press is loving it, e.g.:
"BITCOIN TOUCHES $US220 ON MTGOX AFTER HORRIFIC 60% PRICE CRASH"
http://www.businessinsider.com.au/bitcoin-crashes-to-220-on-mtgox-2014-2

It is not just MtGOX's fault.  Blame should fall also on those bitcoiners that still refer to it as the main exchange, and on the chart sites that still show it by default.

Why do bitcoiners still refer to MtGOX as "the main bitcoin exchange", or even as a bitcoin exchange at all? For sentimentalism? For personal friendship? Because Mark is "one of us"?

Neptune.  Imaged by Voyager 2 in 1989.

http://voyager.jpl.nasa.gov/gallery/neptune.html

    Deifenbach: Mr. Lundegaard, this is Reilly Deifenbach calling from GMAC. How are you this morning?
    Jerry: Yah, real good. How you doin'?
    Deifenbach: Pretty good, Mr. Lundegaard. You're damned hard to get on the phone.
    Jerry: Yah, it's pretty darned busy here, but that's the way we like it.
    Deifenbach: That's for sure. Now, I just need, on these last, these financing documents you sent us, I can't read the serial numbers of the vehicles on here, so I--
    Jerry: But I already got the, it's okay, the loans are in place, I already got the, the what, the -
    Deifenbach: Yeah, the three hundred and twenty thousand, you got the money last month.
    Jerry: Yah, so we're all set.
    Deifenbach: Yeah, but the vehicles you were borrowing on, I just can't read the serial numbers on your application. Maybe if you could just read them to me--
    Jerry: But the deal's already done, I already got the money.
    Deifenbach: Yeah, but we have an audit here, I just have to know that these vehicles you're financing with this money, that they really exist.
    Jerry: Yah, well, they exist all right.
    Deifenbach: I'm sure they do, but I can't read their serial numbers here. So if you could read me--
    Jerry: Well, but see... I don't have them in front of me. Why don't I just fax you over a copy?
    Deifenbach: No. Fax is no good, that's what I have and I can't read the darn thing.
    Jerry: Yah, okay, I'll have my girl send you over a copy, then.
    Deifenbach: Okay, because if I can't correlate this note with the specific vehicles, then I gotta call back that money.
    Jerry: Yah, how much money was that?
    Deifenbach: Three hundred and twenty thousand. See, I gotta correlate that money with the cars it's being lent on.
    Jerry: Yah, no problem, I'll just fax that over to ya, then.
    Deifenbach: No, no, fax is--
    Jerry: I mean send it over. I'll shoot it right over to ya.
    Deifenbach: Okay.
    Jerry: Okay, real good, then.

I hope someone is saving the transaction logs of all the exchanges.  They would be a gold mine for historians, reporters, market analysts, etc.

That data may one day may be worth many bitcoins. 

MtGOX today was 61% of all non-China,  Last Tuesday it was only 14%, and falling.

MtGOX always had suspicious robot activity, perhaps fake trade.  No one knows what is going on there now.

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


             !    Sat !    Sun !    Mon !    Tue !    Wed !    Thu !    Fri !    Sat !    Sun !
  EXCHANGE   !  02/08 !  02/09 !  02/10 !  02/11 !  02/12 !  02/13 !  02/14 !  02/15 !  02/16 ! Currencies considered

  MtGOX      |  26.82 |  23.48 |  50.07 |  18.70 |  25.21 |  33.85 |  79.56 |  60.15 | 104.46 | USD,EUR,GBP,AUD,JPY
  BTC-e      |  11.67 |  12.51 |  47.04 |  28.64 |   9.97 |  18.53 |  52.18 |   9.16 |  21.75 | USD,EUR,RUR
  Bitstamp   |  13.50 |  15.06 |  71.86 |  40.06 |  15.51 |  28.15 |  63.38 |  20.76 |  26.40 | USD
  BitFinEx   |  11.12 |  12.07 |  41.98 |  39.54 |  12.85 |  17.17 |  53.68 |  10.83 |  16.75 | USD
  Kraken     |   0.37 |   0.45 |   1.45 |   1.11 |   0.58 |   0.93 |   1.79 |   0.41 |   0.91 | EUR
  Bitcoin.DE |   0.37 |   0.47 |   1.54 |   0.67 |   0.51 |   0.78 |   1.63 |   0.22 |   0.57 | EUR
  CaVirtEx   |   0.57 |   0.23 |   0.77 |   0.53 |   0.29 |   0.41 |   1.16 |   0.15 |   0.15 | CAD
  CampBX     |   0.27 |   0.24 |   0.37 |   0.11 |   0.10 |   0.21 |   0.41 |   0.05 |   0.07 | USD

  SUBTOTAL   |  64.69 |  64.51 | 215.08 | 129.36 |  65.02 | 100.03 | 253.79 | 101.73 | 171.06 |

  Huobi      |  77.00 |  64.08 | 134.09 | 120.08 | 131.86 |  82.39 | 236.27 | 122.80 | 110.57 | CNY
  OKCoin     |  37.63 |  34.63 |  70.49 |  64.22 |  60.79 |  62.49 | 147.26 |  51.45 |  63.63 | CNY
  BTC-China  |   6.86 |   5.75 |  18.23 |  11.17 |  11.01 |   8.02 |  24.87 |   7.55 |   9.08 | CNY
  Bter       |   0.26 |   0.33 |   0.85 |   0.76 |   0.69 |   0.65 |   1.54 |   0.74 |   0.56 | CNY

  SUBTOTAL   | 121.75 | 104.79 | 223.66 | 196.23 | 204.35 | 153.55 | 409.94 | 182.54 | 183.84 |

  TOTAL      | 186.44 | 169.30 | 438.74 | 325.59 | 269.37 | 253.58 | 663.73 | 284.27 | 354.90 |



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.

The exchange Bter was added to Bitcoinwisdom's menu on 2014-02-15; its volumes for 02/07 to 02/14 have been added
retroactively to the table.

Coinbase is said to use Bitstamp for currency conversion.

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: A few days ago, 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.

How does one read this "Bitcoin mining profitability" chart?  Is it something to worry about?

https://coinplorer.com/Charts/PriceDifficulty?fromCurrency=BTC&toCurrency=USD

If anyone still cares (after three grossly failed attempts in a row) here is the Chinese Slumber Method prediction for Monday feb/17 UTC:

The Slumber Points are the Huobi prices at 19:00 UTC of each day.  They are the yellow dots in the image below (ignore the rectangles and the balloons):


Unfortunately, there is still no clear trend yet for those Slumber Points.  For the lack of a better idea, I will use the straight line that best fits (in the least squares sense) the last three Slumber Points, namely 4004, 3961, and 3733 CNY on feb/14--16.  That line, extrapolated one more day, gives ~3628 CNY; which is ~593 USD using the usual currency conversion R = 6.12. Therefore:

Prediction valid for: Monday 2014-02-17 19:00--19:59 UTC (not before, not after):
Huobi: 3600 -- 3660 CNY (3630 plus or minus 30)
Bitstamp: 580 -- 600 USD (590 plus or minus 10)
MtGOX: 0 -- ∞ goxUSD (something plus or minus anything)

May I ask again that UTC be used (otpionally or always) for all times and dates?

I am constantly referring to your charts on other threads of this forum. Since the readers are inernational, I must convert all hours and dates in my head from Brazilian local time to UTC.

Moreover, the data on the 1-day chart is actually binned in UTC days, not local days; which puts the dates almost 1 day off here in Brazil.

For example, to get the total volume for feb/07 on the 1-day chart, I must put the cursor on the bar labeled feb/06, since that bar actually covers feb/06 22:00 to feb/07 21:59 Brazilian time, which is feb/07 00:00 to 23:59 UTC. 

Moreover Brazil just switched to savings time, so when trying to locate or get the time of an event on the more detailed charts I must add/subtract either 3 hours or 2 hours from the time shown, depending on which part of the plot I am looking at.

Several other sites use UTC times, I do not think that people will object to the switch, if it is not optional.  In fact I do not see the advantage of displaiyng local time; bitcoin events are not connected to local events.

Thanks again for your wonderful work.