Good analysis, as always, TERA. Glad to have you on board here.

One thing: I'd be careful to place too much trust on *daily* StochRSI. It moves fast by design, and on the daily time frame it can be a useful trading tool for individual buy/sell decisions, but to judge the likelihood of a longer-term reversal, I believe it is preferable to look at it on a weekly scale:



In which case, we see that your main point stands: We're nowhere near the oversold state we were in in June/July, timewise. That, among other things, makes me rather skeptical this bear market is over yet... even though I'm also skeptical we'll probe new lows soon -- more like a continued, reluctant decline, is what I could see in the coming month.

Maybe the bear mood finally got me, but I wouldn't be surprised if we end up not only at the daily EMA200, but the 300 again.

re: "this rally has legs"... that's exactly what I'm doubting, and as much as shorts can be seen as a "load down coil", it can also be the suppressor that keeps it down. I would have really liked to see us go through 620 on decent volume, but neither are we near that price, nor did volume keep up. Maybe it's a resting state after being overbought, but right now, I'm as skeptical as I was on the 23rd (i.e. when we were at 640/650, before dropping back to 400). Not saying that'll happen again, but like I said: my sentiment is short&mid term bearish, and it takes some more to convince me we're turning around this time.

Hmm, I see. Can I make a suggestion? Maybe look at bid/ask on the link I already gave above, or on http://coinorama.net/ for example, and see if it helps your own trading in the coming weeks or so.

If it does help you, maybe you can consider implementing it, but if you conclude it's a useless indicator, that's a good reason not to implement it.

That remark, as innocent as it sounds, is probably *the* most central problem we have when doing TA: we simply don't have much data yet.

Don't get me wrong, I don't disagree with you (about the daily EMA200, or if you're conservative, EMA300 which we hit 2 days ago), but a nagging voice in my head always tells me when I base some TA on this argument I make in my head (implicitly):

"Situation now resembles this and that aspect of the bubble/pop cycle 2012 and the bubble/pop cycle 2013 (but not the bubble/pop cycle of 2011, but I ignore that one because times were different then)".

I hope you can see my point. If you go by some semi-objective volatility measure (like BBW), we had 3 big cycles (2011, 2013, 2013/14), and a smaller one in 2012. Of those, at most 4 cycles, 1 was different, but me (and many others doing TA) tend to think of the 2011 cycle as an outlier, that can mostly be ignored.

I do the same, let me say that once more.

But, as I said above, a nagging voice in my head tells me to be careful doing so, "ignore 2011 at your own risk".

I'm not sure I understand the problem. There are two things you mention, right?

1) Problem: Not all exchanges return the entire order book. Solution: Work with what you get. Also, it seems that the currently most important exchange, Bitstamp, does return the full order book, right?

2) Problem: Too much data needed. Solution: Maybe I missed something, but you don't need to keep the order book data, no? Just calculate the ratio once, and keep that. Or did I misunderstand you?

I don't know what happens at the backend of your website, so maybe I underestimate the amount of code needed to make bid/ask work, but to me it seems that if you only store the bid/ask ratio value itself, not the order book data, it's not such a huge amount of data you need to handle... but maybe I'm wrong.

Good point. Short-term bets on a retracement piling up on top of the shorts betting on the overall downtrend. That'd explain the last day or two of increase in the graph I posted.

This doesn't make much sense. It depends on how long you can avoid margin calls, but the overall trend would definitely favor going long on leverage. But I agree, if you trade near your limits, you can easily be wiped out in a single flash crash.

But anyway, even if shorting is the natural inclination in BTC, it still doesn't explain the divergence I point out, that the rest of the markets are starting to look a bit more bullish again, while shorting increases. Like I said, in the end I guess it's a bet by finex traders against the recovery, but it wouldn't be the first time that their opposition is simply wiped away by the rest of the markets (think the last that happened was in October last year)

I beg to disagree. THat's exactly what everyone claimed post-260 ATH, that it'd take several attempts to break it, selling pressure all the way up, yadda yadda. In reality, the hard bit was going up from ~60+ (July) to 130 (end of September), then shaking off the SR crash. Everything afterwards was just one smooth ride, and the previous ATH wasn't even any serious battle anymore (3 days were spent around it, then: done). Sure, we had Chinese help, but I'm just saying that breaking the previous high isn't such a huge deal if the buying pressure is there. (which, admittedly, is still lacking right now)

I can't explain the big jump we've seen yesterday (or 2 days ago?) either. I read a comment by TERA (EDIT: or matthecat) who in fact linked to the shorting going on on stamp through finex (in the sense that those who sell on stamp through finex will pump USD into stamp), but that still doesn't explain the almost vertical increase, in my opinion.

I'd say the most likely explanation is one of the following two: (a) attempt by a large market participant to "paint the order book", i.e. to give the impression of buying pressure where there is none, or (b), more positively: USD sitting on stamp's account, but not the order book, finally entered the race and got put on the order book.

Like I said, it's possible it's just a lagging signal. But what really surprises me that, including today, the total BTC swaps are still going *up*. I really don't see that as anything other than finex traders collectively betting against this recovery. We'll see who comes out on top in a few days, I guess.

It's certainly starting to look a bit more bullish again, doesn't it? There's still uncertainty in the air about how the gox situation will resolve, but that rebound from 400 was beautiful: loads of volume, and remarkably controlled retracement so far. All in all, I think I'm not the only one who looks at the charts and starts wondering if the post-December-ATH bear market is maybe over...


If it weren't for the stats from Bitfinex. This is a view of the past week. 'Total sum of active USD swaps', i.e. people going long on leverage is, well, unimpressive.





No big deal, maybe. But then there's total BTC swaps:





It's a new ATH (at least as far as the website, charts-bfxdata.rhcloud.com, goes back). In other words, traders on Bitfinex are still shorting the hell out of Bitcoin.

I can see two ways this will resolve: Bitfinex reaction is lagging, price continues to climb (which will lead to a lot of headache for those with active shorts), or the shorting traders are onto something, and we're coming down again, hard.

Personally, I can't say exactly which way I'm leaning -- I'm still cautiously optimistic, but the Bitfinex stats certainly put some more doubt into my mind again.

Sure, I know you're not giving trading advice. I was more thinking out loud what my own thoughts are on the topic. I've played with loglinear models myself, can see how they seem to fit impressively, but at the same time, it's not at all obvious what to *do* with them. I'm still working on it

I'm going to make a donation then. Maybe not 1 btc, I'm not that loaded. But I really appreciate your work. That said, could you perhaps run me through how to get the results? Not the single loglinear trendline, I can do that myself (in R). But I have no idea how you got the "running" calculation, i.e. the red trendline. Some 'for' loop, that works on an increasingly bigger subset of the price data? Maybe wait with the explanation until my transaction gets through

To my knowledge, order book bid/ask ratio is calculated as: Total_Bid_Sum(USD)/Total_Ask_Sum(BTC).

For example, right now Bitstamp total order book BID is ~22 million USD, ASK is ~17000 BTC, so bid/ask is ~1294 USD/BTC.

There are then two ways to display this ratio over time:
(1) "raw", as calculated above, or
(2) "normalized" for the actual price at the same time as the ratio is calculated, example: currently, ratio is 1294 USD/BTC (as calculated above), current price is 585 USD/BTC, so normalized ratio is ~2.2 (dimensionless).

* * *

In my experience order book bid/ask ratio has to be used carefully, since the order book is obviously target for manipulation, but looking at how the ratio changes over time can be very useful to determine trend reversals.

order book!

damn, we talked about different things all the time, huh?

You completely misunderstood my point. Mtgox closing is not the "reason" for reaching (hypothetical) 6k coins. Nothing is ever the reason, besides market psychology (and fundamentals, as seen through the lens of market sentiment). The point is that, if we reach a new ATH, I'm sure people will look back at Mtgox closing and quoting it as "one of the reasons", just like they now quote "SR closed" or "the Chinese came in" as *the* reason, when in reality those are factors influencing the price, but never in such a straightforward and obvious way as it is made out to be in retrospect.

First thing I want to say first: thanks for the diligent work. Seriously.

Second thing: Care to make the (R, I guess) script available so others can play with it? Wouldn't be unusual to tie it to a btc donation though, btw.


Anyway... what I really wanted to say: I've said several times in different threads that also use log-linear model for price analysis that I'm generally critical about using this method, at least if the goal is active trading. The problem is of course how actual price relates to the price predicted by the regression -- in the sense that any actual trading signals are very imprecise, and can be off by an order of magnitude if interpreted naively.

(sorry if I'm not getting my point across concisely, big flaw of mine) What I mean is this: most of the time when I've seen people posting regression analysis, they take price going *above* the regression prediction as a sell signal ("We're above the target"), and price going below as a buy signal. If they would have traded like that, they would have ended up at a huge loss compared to buy&hold (e.g. they would have had to re-buy much higher in 2013, after selling way too early). Note please that I'm not saying you're giving that advice, but my point is: regression analysis has to be interpreted somehow, and it is not obvious how to interpret it, in my opinion.

Here is one preliminary observation, or attempt at a slightly less "naive" trading model based on regression: Except for 2011, we tend to stay below the loglinear trendline much shorter than above. So selling if we go *above* the line is probably less warranted than buying when we go *below*. Very crude first attempt, I know, but I'd like to get the discussion started.

EDIT: hahaha, just noticed, you're a member since 2012. Probably don't feel the need to ask for mBTC donations :D

Complete agreement. Bit of an unsatisfactory answer maybe, but "the next few 6h candles will decide how it'll resolve" is the best answer I think that can be given at the moment.

I'm directly comparing the end of the post-April 2013 downtrend (in early July 2013) to the situation now, on the 6h view, and there are a large number of similarities, but also a few differences -- in July 2013 CMF took a *sharp* swing upwards right around that time and didn't come back even as price retraced a bit.

So I'd say I'm cautiously optimistic this marks the end of the post-December consolidation period/downtrend, but it's too early to say with certainty. All it takes is another large enough market sell by a whale, too early into the recovery, and we're back to consolidation time.

I sometimes have a hard time getting across my point because of my convoluted writing style. Apologies for that.

Here's my point, ultra condensed: Mtgox shutdown (in whatever form it'll happen) *does* matter. No question about that. But the question is *how* it will matter on which time frame.

Short term, it certainly will depress price (and did so already, yesterday for example). But I caution anyone who believes to know with certainty that it will depress prices for a long time. There are several plausible interpretation scenarios that will make mtgox closure *positive* in the long run, just like we know think of the SR closure as (obviously) a good thing.

Note that I'm personally not *sure* if mtgox closure will be seen as positive in the long run, but I see the real possibility for it. The market will ultimately decide, and I just mention the possibility that, in a year from now, we'll talk again about mtgox and everyone will go "Well, d'uh, of course it was good news. The worst run exchange in the BTC ecoystem *finally* went down. That's what brought us to the new ATH of 6000."

You're arepo from the speculation subforum, right?. Yeah, I'd trust you in principle, to give you that loan, but I don't see exactly the conditions being that favorable. Didn't check bitfinex lately, but I should be able to get a similar interest there, with the counterparty risk being lower even -- like I said, I'd trust you, but I probably trust bitfinex a bit more even

(btw, what happened to your TA posts? I always enjoyed them quite a bit. if you ever start a newsletter, make sure to post about it in the speculation forum, I'd sign up for it if the price isn't too outrageous)

I will refrain from pointing out that you only joined a few months after the SR crash, if you promise me to go back and read a thread or two back from those days:

At the time of the SR closure, the absolutely *dominant* sentiment among the pundits in here was that it will greatly depress price. Anyone who said differently was shouted down. Took a few days, and that sentiment very clearly changed.

So, to translate it to the current day situation: yes, I compare the two events in the sense that, at face value, you might naively say "oh, it's going to keep price down for months to come".

In reality, it will probably throw price down for some days, maybe weeks. And then, gradually, it'll be seen differently, as extremely good news.

Just a few of those reasons, just from the top of my head: (1) the formerly largest exchange was run amateurishly, and was punished accordingly. (2) cheap mtgox coins that would have otherwise flooded the market are now unavailable to arbitrage down the price. (3) Mtgox investors will look to "replace" their coins (not all of them, of course. But those who *do* want to replace them will have to buy them again.