15min MACD divergence flipping to red soon, usually means more dumps to come.

My mistake, sorry. The "little pump" I expected yesterday evening turned out to be a flat, and then downhill. It's bearisher than in September.

15min and 5min MACD still bearish on Bistamp and Bitfinex, but bullish divergence on Huobi, the extra fiat there makes a small difference.

Just realized that I made a mistake by shorting bitcoin, I should have shorted litecoin instead...

^^^
It just happened on BFX!

Bitfinex catching up! 5k dump!

Oh, come on, "total collapse"? The bubble continues to deflate, at some point there should be be a true recovery.
But it's possible to be the last week with price above 300$ for several months.

Every time the 6h PSAR flipped to bearish during the last couple of months, the downtrend lasted 2 - 3 days.
It won't take much dump to break support at 305$, and after that a large dump can take the price to 275$.

I smell an incoming little pump, in a couple of hours.

There is no official police report yet, just "According to sources in the Metropolitan Police Department".
I am looking forward to read an official police report. If they will come to similar conclusions as the "Willy report",
then I would expect a seriously bearish influence on the market. Quoting from the "Willy report":

It needs to be recognized that, whether intentional or not (though plausible ignorance only goes so far), Mt. Gox has effectively been abusing Bitcoin to operate a Ponzi scheme for at least a year. The November “bubble” well into the $1000’s – and possibly April’s as well – was driven by hundreds of millions of dollars of fake liquidity pumped into the market out of thin air (note that this is equivalent to “with depositors’ money”). It is only natural that the Bitcoin price would deflate for around 5 months since its December peak, since there was never enough fiat coming in to support these kind of prices in the first place.

I can't agree completely. Only for 2013 and 2014, when the speculative value rose (and slowly deflated) much higher than the utility value.
IMO in 2012 the speculative value was closer to reality. And since bitcoin economy has grown from SR1 to clones with a larger market activity,
there still is a lot of utility value for bitcoin. Which utility value may rise in the near future if Openbazaar will catch on.

Blitz explained well what I meant, so I'll only add this: if the market would be driven by Bitstamp (or BTC-E) it should crash and burn.
To do so, the downward movement has to gain momentum, but with a lot of fresh fiat on Chinese exchanges, this is difficult.
While I doubt that there is enough fresh fiat on Huobi and others for a true trend reversal now, it may be enough to delay, dampen, and
even truncate possible upcoming drops. To confirm a trend reversal (to bull market), IMO the market should at least break resistance at 365$.

If this (from coinsight.org) is correct, I don't know, but looks legit:

I doubt that. After reading the "Willy report" I came to a different conclusion: that a MtGox staff, possibly Karpeles himself,
pumped up the price with fake fiat, breaking through critical resistance levels and truncating corrections.
The purpose was to "build" a 2x - 4x higher price of a normal bull market, that would attract new and true speculator fiat.
If Karpeles wanted to recover stolen coins from 2011, his trading would have resulted in artificially suppressed prices.

I made a comparison of the current bid and ask sums on Bistamp and Huobi, with the ones during the second half of September 2014.
The differences are massive: for Bitstamp, they suggest a major drop starting 3 - 5 days from now, and reaching 155$ - 180$ about 20 - 25 days from now.
For Huobi, the bid sum has increased a lot, maybe the reinstated deposits are working and there has been fresh CNY entering Chinese exchanges.
This complicates the situation, because the direction of the market depends on who leads. To crash Huobi a lot of coins must be dumped.

Interesting... I thought about writing some experimental code for an exchange, but couldn't figure out an algorithm that
would allow multithreaded operation. To execute a huge number of orders per second I guessed mutithreading would be useful.
But each new order to execute depends on the results of the previously executed ones (for the price and available amounts).
Doing inter-thread communication to sort this looks too hard for me (and may kill performance), so it's in the forgotten projects drawer.

1d PSAR flipped to bearish on both BTCchina and Bitfinex.

It's interesting to see analyses more bearish than mine.
https://www.tradingview.com/v/MW7ZZW0C/
https://www.tradingview.com/v/DMMyayp9/

And why would this be nasty? I would call a 200$ bottom on the bullish side.
However, very short term a small uptrend is possible, even a fake breakout.