You mean, "EXPE starts accepting dollars from people who already had bitcoins but have decided to sell them". 

Perhaps that is why all those "adoption" news have failed to move the price?

Well, I myself predicted, back in April, that the price would fall to somewhere between 80$ and 340$ if the Chinese exchanges were to close (orderly, without goxing their clients, so that their coins moved back to the Western markets).  I still do not know whather that will happen, or whether this drop is the beginning of that event. I can't even figure out whether the Caixing article is a step in that direction.  Some day we will know, I hope.

I also trivially predicted in April that the price would rebound to about 580$ (the value before the March 29 PBoC scare) if the previous Caixin leak was false or the PBoC decision was reversed.  Neither of these things happened, but apparently exchanges and traders thought that they had found acceptable workarounds, and the exchanges promised to be nicer to non-professional traders; that is my explanation for the stable price from Apr/25 to May/22.

The rally after that may have been due to a general surge of optimism in the Chinese market about a stable and lasting status quo having been reached; but that explanation does not fit well with the sudden onset and the discrete spurts, so I am still leaning towards the insider info theory.

Peace, each one with his religion... 

Maybe your TA has been right; I have not kept track who posted what, among lots of TA pointing to every possible price move in every time frame...

I think we do have at least one candidate "fundamental" for the current fall, the Caixin article of June 12 (China time).  There may be others, in China or in the "West".  Just because we do not know any external cause for a price move, it does not mean that there was no external cause, much less that the move was driven by price history alone.

But in fact we do know, with great confidence, the external causes for most of the major price moves that happened since early 2013, at least.  That should justify looking for the causes of this last rally, no?

"Bitstamp troubles question mark" is those theories that others have been posting here, e.g. that pastebin "report". 

It is 4:00 am in China now.  The continuing drop in price at Huobi etc. for the last hour must be arbitragers tracking the fall at Bitstamp & c.

Could it be that there are two separate sources of gloom in the market, one in the "West" (Bitstamp troubles?) and one in China (the Caixin article?).  They take turns in causing individual traders to dump, and then each dump gets echoed on the other side by arbitrage and panic selling?

Ah.  Indeed, I have noticed a peculiar characteristic in the OKcoin transaction log: there appears to be many small transactions scattered throughout the actual bid/ask spread, rather than at the ends of the spread (as one would expect from legitimate buys or sells).

[ click on image for a full-size version ]
Here each dot is a trade; the area is roughly proportional to the amount of BTC involved.

The effective bid/ask spread for each exchange can be inferred as the upper and lower envelope of the dots.  One explanation for those OKCoin trades in the middle of the spread could be a robot that posts a buy order in that gap and a matching sell order, or vice versa, in close succession.

You are aware that the order book shown on bitcoinwisdom and other chart sites is truncated, right?  You cannot get the bid/ask ratio rfom that.

One more clueless guess won't hurt, right?

My explanation for the mini-rally of the last few weeks is that some Chinese traders got insider info of some impending positive development X, and all the other traders simply went along with them, without understanding why.  Now some of those traders may have got tired of waiting for X, or began to doubt the rumors, or got scared by that Caixin article.

So, what will happen next depends on whether X will actually happen, and whether it is indeed a positive thing.

A little hurried no? Let's see first what the guy will do at OKCoin.  He may work on a new overseas BTC/CNH exchange, for example.  Or stop bitcoin and litecoin trading at OKCoin, and turn it into a Dogecoin-only site.  

Huobi still seems to have the "cleanest" BTC/CNY chart, with a tight spread and volume that changes as expected according to the local time and holidays.  OKCoin has a steady ~10% bachround volume, 24/7, that may be fake trade.

EDIT: seriously, every exchange may be using fake trades to boost their volume or manipulate the price; and every exchange may be trading against its clients, with obvious unfair advantages, or doing other unethical things.  Methinks that bitcoinwisdom should not give much thought about those issues; its best interest should be to provide whatever its users want.

Thanks! 

Any guess for what could be the premium for, say 50 kBTC? Closer to ±5%, or to ±50%?

This post by @cryptoceelo may be of interest to miners: https://bitcointalk.org/index.php?topic=640717.msg7266240#msg7266240

A public auction is the normal way to liquidate the assets of a failed company.  Any other method (such as selling them in the open market, or to a private buyer) would expose the liquidator to charges of favoritism, incompetent trading, etc..  That also holds for the coins that the FBI seized from SilkRoad.

The FBI will use their government auction staff, but the MtGOX liquidator will probably hire some private auctioning firm.  Presumably they will divide the coins into a couple dozen lots, each auctioned separately, in order to attract more buyers.  Other than that, I don't expect them to worry much about getting the best possible price.

The people who will buy those coins at the auction will then dispose of them according to their interest. 

By the way, I suppose that the prices off-market for "bulk" trades are not much different from the open market price.  If the off-market price was much lower, there would be people buying bitcoins in bulk off-market to sell them little by little on the open market.
Conversely, if the off-market price was much higher.   Is this correct?

According to WIkipedia, Liberty Reserve was charged with crimes that (in the USG view at least) were far more serious than being an unregulated stock exchange.  The USG may have extrapolated his jurisdiction in some points, but the business was hardly innocent.  in any case the SEC does not seem to have been involved.

From a quick reading of posts in this forum and other sources, I would think that the SEC got them to stop marketing to US persons, yes; but the final collapse was not by direct action of the SEC, although the loss of US clients may have caused it.  Moreover it seems that there is some obscure aspects about Bitfunder's collapse -- Danny Brewster got involved with Ukyo at some point, and someone is suing him, it seems https://bitcointalk.org/index.php?topic=348468.0

Still on that topic, an exchange owner can profit at his clients' expense, even without arbitrage, if there is a delay of a second or two between the submission of orders by clients and the display of those orders in the public order book and trade log.

Suppose that, at some moment,  the highest bid is 400 USD and the lowest ask is 500 USD.  Client A submits an order to sell 1 BTC at 420 USD.  One second later, client B, who still has not seen A's order, submits an order to buy 1 BTC at 480 USD.  The exchange operator notices the overlapping orders, and then inserts his own orders, with fake times, between those of A and B: first a buy into A's offer at 420, then a sell order at 480, which is then taken by B.   When those transactions finally appear in the trade log, both clients think that they were lucky to get their orders immediately filled; when in fact the operator took from B the 60 USD that he would have saved without his intervention.

(Why invest in bitcoin, if one can make millions *now* by opening a bitcoin exchange -- even a zero-fee one?  )

One does not need to move funds in real time between exchanges to do arbitrage.  Or maybe not at all.

An arbitrage opportunity occurs whenever someone posts a bid at one exchange A that is higher than the lowest sell at the other exhange B (or someone posts a sell at B lower than a bid  at A).  At that moment, an arbitrager at A, if he became aware of that situation before other clients, can sell into that bid, while his collaborator at B (need not be the same person) buys into that ask.  Both traders will then profit, relative to the mean bid-ask price that will result from their actions.  These trades are internal to the exchange, and therefore instantaneous - they do not appear in the blockchain at all.

In fact, one could view arbitrage as two people trading, each within one exchange, with the usual "buy low, sell high" rule.  The difference is that, by knowing the other exchange's order book before other clients, each trader can, in a sense, peek into the immediate the future - he gets a hint as to whether the price at his exchange is currently "low" or "high", relative to the mean price that the arbitragers themselves are about to fix.

With those hints, both traders can consistently "buy low, sell high" and keep increasing the total vaue of their accounts.   It does not matter whether price is in an increasing or decreasing trend, or varying randomly within a band; as long as the bids and asks get posted independently at each exchange, these occasions are likely to happen.

The age seems to match (N&B's CEO  was said to be 27 in news articles).  However one must be careful, there are several "Daniel Brewster"s in the UK.

I wonder whether the Cyprus police has done anything (via Interpol or other) to extend the arrest warrant to the UK.

Is that confirmed? (I don't know about Havelock, but I have seen mentioned in this forum that at least one bitcoin company out there turns down investments from Americans, apparently for that reason.)

In the real stock market, are the exchange operators alowed to trade in their own exchanges? E.g., to do arbitrage?

I would think that the exchange operator would have tremendous advantages over his clients if he could do that, since he can know the trades a few seconds before them.  Two colluding exchange owners could know each other's books a few seconds before anyone else, and thus use all arbitrage opportunities in full.  And surely there are many more ways which I cannot imagine.

I can see how Huobi and OKcoin could make big profits without charging any trade fees.  Indeed, by not charging fees they get very narrow bid/ask spreads, so that the opportunities for arbitrage between them are much more frequent.

The money that an exchange owner can make this way must come, of course, from his clients' pockets.

EDITS: grammar, sorry

Well, Huobi and OKCoin are reacting.  Volume jumped (even though it is only 6:00 am in China) and price fell some 15 yuan (~2.50 USD).

Bitstamp led this time, right?  The first big dump there was at 21:45 UTC, then 21:46 at BTC-e, Bitfinex, then 21:47 in China.

That difference may not be real.  The price in USD displayed by Bitcoinwisdom is computed by the chart scripts using some value for the CNY:USD exchange rate (presumably the official one, not necessarily up-to-date), that may not be the one assumed by arbitrage traders.