BETI: Bitcoin Exponential Trend Index and technical analysis

[oda.krell]

And it will keep going lower, since there are more bad news than there are good ones.

What planet are you living on?  The news for bitcoin has been overwhelmingly bullish essentially ever since Gox went titsup in the river.

Price is low because of dumping on exchanges, which are easy to manipulate because they are thinly traded, because everyone is holding tight.

Why dump?  Two reasons:  Cheap coins from industrial mining, and suppressing price marks for contracted OTC prices.  These two factors enable large actors to enter the market without adverse price movement.

This dam will break.  It will break suddenly, and it will break massively.  The wave will be a tremendous wall, and when it passes, all our houses will be washed up on higher shores, if we do not make a shipwreck of them in all the excitement.

Or, as others pointed out before, the stream of good news is the main reason why we are able to sustain the current (highly speculative for current fundamentals) price level despite lackluster buying pressure. IOW, it keeps the "holders" happy and (somewhat) calm.

I agree with the "dam break" metaphor. OTC volume is probably the largest factor in the current stagnation. Where we perhaps disagree is the time frame when the "public" buying pressure has built up sufficiently to be released violently. I can see worst case scenarios where it take another year or two before that'll happen.

/OT

Thanks jl2012, and raystonn for your contributions to this thread. I really enjoy this one (even when I'm skeptical of the conclusions sometimes)

[1714501571]

1714501571

[1714501571]

1714501571

[RAJSALLIN]

Following

[Raystonn]

There is no reliable buy signal from this plot

I see this:


I do not like the repercussions of my analysis.  May GABI and COIN stop it in its tracks.

It looks like the current low matches up fairly well with the low before the June 2011 spike began.  That would mean the drop after the next bubble would be horrific, like that after the June 2011 bubble peak.  I'm hopeful COIN and GABI can change this pattern.  Also, if our bubble doesn't start soon, we may just skip it altogether and drop into the horrific valley after continued sideways trading.  I'm doubtful on that though as I we need the dumb money bought into the market before it tanks like that.  Right now the dumb money is in fiat.

[h3speros]

And it will keep going lower, since there are more bad news than there are good ones.

What planet are you living on?  The news for bitcoin has been overwhelmingly bullish essentially ever since Gox went titsup in the river.

Price is low because of dumping on exchanges, which are easy to manipulate because they are thinly traded, because everyone is holding tight.

Why dump?  Two reasons:  Cheap coins from industrial mining, and suppressing price marks for contracted OTC prices.  These two factors enable large actors to enter the market without adverse price movement.

This dam will break.  It will break suddenly, and it will break massively.  The wave will be a tremendous wall, and when it passes, all our houses will be washed up on higher shores, if we do not make a shipwreck of them in all the excitement.

Or, as others pointed out before, the stream of good news is the main reason why we are able to sustain the current (highly speculative for current fundamentals) price level despite lackluster buying pressure. IOW, it keeps the "holders" happy and (somewhat) calm.

I agree with the "dam break" metaphor. OTC volume is probably the largest factor in the current stagnation. Where we perhaps disagree is the time frame when the "public" buying pressure has built up sufficiently to be released violently. I can see worst case scenarios where it take another year or two before that'll happen.

/OT

Thanks jl2012, and raystonn for your contributions to this thread. I really enjoy this one (even when I'm skeptical of the conclusions sometimes)

Indeed, but I would like to hear more about more likely scenarios, althought it's good to prepare for the worst case too.

I'm inclined to think that It's likely that we will see new ATH before this year ends. Of course you can not know for sure, but if OTC investing rises this fast, those OTC coins have to end soon and that is going to reflect on Bitstamps price. By OTC investing I mean example these: http://www.coindesk.com/6-new-hedge-funds-seeking-bitcoin-returns/

Now I'm trying to figure out if this is just expensive hope or is there really good enough facts that support this claim.

[oda.krell]

Indeed, but I would like to hear more about more likely scenarios, althought it's good to prepare to the worst case too.

I'm inclined to think that It's likely that we will see new ATH before this year ends. Of course you can not know for sure, but if OTC investing rises this fast, those OTC coins have to end soon and that is going to reflect on Bitstamps price. By OTC investing I mean example these: http://www.coindesk.com/6-new-hedge-funds-seeking-bitcoin-returns/

Now I'm trying to figure out if this is just expensive hope or is there really good enough facts that support this claim.

The 'best case' scenario is easy, and is talked about on here often enough: new fiat gate opens and is used at or near maximum capacity. Could be an ETF, could be regular Argentinians coming on board in numbers. Effect would be the same: buying pressure that no amount of profit taking can harm. For a while.

As for what is the most likely scenario, that's up to you. I have prepared myself for a longer waiting time before the next growth spurt takes place, and more importantly: I'm not mortally afraid to "miss" it. The initial stage of those phases have been recognizable so far (say, early Nov 2013) and almost the entire swing up can be held in btc without assuming that the next big rise is "just a month away" (which seems to be the default assumption of the most bullish posters in here ever since April).

[h3speros]

Indeed, but I would like to hear more about more likely scenarios, althought it's good to prepare to the worst case too.

I'm inclined to think that It's likely that we will see new ATH before this year ends. Of course you can not know for sure, but if OTC investing rises this fast, those OTC coins have to end soon and that is going to reflect on Bitstamps price. By OTC investing I mean example these: http://www.coindesk.com/6-new-hedge-funds-seeking-bitcoin-returns/

Now I'm trying to figure out if this is just expensive hope or is there really good enough facts that support this claim.

The 'best case' scenario is easy, and is talked about on here often enough: new fiat gate opens and is used at or near maximum capacity. Could be an ETF, could be regular Argentinians coming on board in numbers. Effect would be the same: buying pressure that no amount of profit taking can harm. For a while.

As for what is the most likely scenario, that's up to you. I have prepared myself for a longer waiting time before the next growth spurt takes place, and more importantly: I'm not mortally afraid to "miss" it. The initial stage of those phases have been recognizable so far (say, early Nov 2013) and almost the entire swing up can be held in btc without assuming that the next big rise is "just a month away" (which seems to be the default assumption of the most bullish posters in here ever since April).

So you have prepared to lose some coins if this goes north fast so that there is not so big risk in losing dollars if this goes into a prolonged bear market, in other words you have big chunk of dollars waiting in sidelines? =)

[rebel24]

y-axis is log(actual price/expected price). It is the worst since Jan 2013

Firstly, thank you for providing the statistics, graphs and formulas, you rock.


However, I want to make sure the data is accurate, and I've been playing with this graph and I think I've spotted an error, correct me if I'm wrong but its quite unusual.

Simply put, I think this graph is incorrect for the part where it shows a y-axis value of around 2 in 2011, and -2 a few months later

Why?

Because a log value to the extreme of -2 or 2 would imply that the actual price vs the expected price is a ratio of 1:50 or even 1:100

Your log graph is claiming that it was up to 50x overvalued in the bubble(at $32) highs, but then more than 50x undervaled in the crash. That simply doesnt make sense (as then the price would be moving 2500x from high to lows, which it didnt, it only moved around 10x).

----
To extrapolate further and give example-
lets assume your log graph is correct through June 2011 (the peak of the bubble). The formula of log(actual price/expected price) would then be
log (32/.5)   (as the high point was about $32, and a value of $.50 for the expected price(which appears as a reasonable expected price at the time at the .6% daily growth rate), gives a log value of 1.80.
All is fine up to this point.

The problem is where your graph then reflects a y-axis value of -2.

Because if you take the low points a few months later, expected price at that point should be around $1 (again because of the .6% daily growth rate).
so log (2.50/1) (because the low point was around $2.50)  is equal to a y-axis value of 0.30. which is FAR FAR different than the data shown.


Am I missing something?

This is relevant because I'm trying to find the point of maximum pain in the past and compare it to now, as based off my own research, we are quite close to being as low below this trendline as we have ever been- but based off this log scale that isnt the case(and thus I think the log scale is incorrect for that time)

[jl2012]

Am I missing something?

That's natural log. There is nothing special with 10, except that most people have 10 fingers.

[rebel24]

Am I missing something?

That's natural log. There is nothing special with 10, except that most people have 10 fingers.

thank you, and once again thank you for your contribution in publishing this work

[ssmc2]

Update por favor?

[jl2012]

Date:    19-Aug-2014
VWAP:    479.83
x:    1494
a:    0.00588
b:    -1.74624
Rsq:    0.90281
The day's expected price:    1142.51
Actual price / expected price:   42.00%
Predicted date for today's price:    24-Mar-2014
Days ahead:    -147.50
Daily price rank:    233
Predicted date for ATH ($1126):    22-Aug-2014
   
(See OP for explanation)   
   
   
   
https://www.wolframalpha.com/input/?i=e+%5E+%28+0.00588167299457308++%28+number+of+days+since+jul+17%2C+2010+%2Fdays+%29+-1.74623767873887+%29   

[jl2012]

The actual price of 19 Aug was only 42% of the expected price. It has no been so low since early 2013

[Driv3n]

Would love to see an update when you get a chance.

[jl2012]

Date:    10-Sep-2014
VWAP:    482.21
x:    1516
a:    0.00583
b:    -1.72039
Rsq:    0.90366
The day's expected price:    1234.31
Actual price / expected price:   39.07%
Predicted date for today's price:    1-Apr-2014
Days ahead:    -161.21
Daily price rank:    244
Predicted date for ATH ($1126):    30-Aug-2014
   
(See OP for explanation)   
   
   
   
https://www.wolframalpha.com/input/?i=e+%5E+%28+0.00583024946907869++%28+number+of+days+since+jul+17%2C+2010+%2Fdays+%29+-1.72039218099432+%29   

[jl2012]

[oda.krell]

Thought this might be interesting to you, OP. Maybe you saw it already...

http://www.extremetech.com/computing/179421-analyzing-bitcoin-why-btc-is-so-valuable-whether-it-will-keep-its-value-in-the-future

Correlation analysis of price/difficulty. Looks quite rigorous to me, but didn't really read it with full attention so far. Also wondering a bit if the extremetech article is the only way they presented their results, or if they published this academically... couldn't find anything.

Reminds me a bit of gbianchi's model, who's modeling price/"network size" (where network size is estimated by no. of blockchain addresses that are/were in active use. Would be interesting to have the two look at each others models, imo.

EDIT: Peter R. had a similar model around the Metcalfe assumption, just using a different estimator for network size. Might be interesting to him as well, in case he's reading.

[ssmc2]

Thought this might be interesting to you, OP. Maybe you saw it already...

http://www.extremetech.com/computing/179421-analyzing-bitcoin-why-btc-is-so-valuable-whether-it-will-keep-its-value-in-the-future

Correlation analysis of price/difficulty. Looks quite rigorous to me, but didn't really read it with full attention so far. Also wondering a bit if the extremetech article is the only way they presented their results, or if they published this academically... couldn't find anything.

Reminds me a bit of gbianchi's model, who's modeling price/"network size" (where network size is estimated by no. of blockchain addresses that are/were in active use. Would be interesting to have the two look at each others models, imo.

EDIT: Peter R. had a similar model around the Metcalfe assumption, just using a different estimator for network size. Might be interesting to him as well, in case he's reading.

Interesting comments too. Hey look, it's Neal Palmquist  Every time I see that user name in an article's comments section it just further convinces me he's a paid shill for some major entity that is very scared of BTC. And that alone makes me happy

[minerpumpkin]

I actually gotta say that seeing the latest version of the graph is rather scary! :/ It isn't unprecedented that we went this far down, yet we're now in a relative state of mainstream adoption and already tapped some of Bitcoin's potential, so I don't hope this means decision time already, and people decided against it...

[cypherdoc]

Thought this might be interesting to you, OP. Maybe you saw it already...

http://www.extremetech.com/computing/179421-analyzing-bitcoin-why-btc-is-so-valuable-whether-it-will-keep-its-value-in-the-future

Correlation analysis of price/difficulty. Looks quite rigorous to me, but didn't really read it with full attention so far. Also wondering a bit if the extremetech article is the only way they presented their results, or if they published this academically... couldn't find anything.

Reminds me a bit of gbianchi's model, who's modeling price/"network size" (where network size is estimated by no. of blockchain addresses that are/were in active use. Would be interesting to have the two look at each others models, imo.

EDIT: Peter R. had a similar model around the Metcalfe assumption, just using a different estimator for network size. Might be interesting to him as well, in case he's reading.

interesting.  they actually claimed price doesn't correlate with tx volume as Peter argues.

"With few exceptions, the long-term hash rate drives upwards — as does the value of the currency".   i've argued this all along.

[oda.krell]

Thought this might be interesting to you, OP. Maybe you saw it already...

http://www.extremetech.com/computing/179421-analyzing-bitcoin-why-btc-is-so-valuable-whether-it-will-keep-its-value-in-the-future

Correlation analysis of price/difficulty. Looks quite rigorous to me, but didn't really read it with full attention so far. Also wondering a bit if the extremetech article is the only way they presented their results, or if they published this academically... couldn't find anything.

Reminds me a bit of gbianchi's model, who's modeling price/"network size" (where network size is estimated by no. of blockchain addresses that are/were in active use. Would be interesting to have the two look at each others models, imo.

EDIT: Peter R. had a similar model around the Metcalfe assumption, just using a different estimator for network size. Might be interesting to him as well, in case he's reading.

interesting.  they actually claimed price doesn't correlate with tx volume as Peter argues.

"With few exceptions, the long-term hash rate drives upwards — as does the value of the currency".   i've argued this all along.

That's why I said I was looking for an actual publication with the full technical details: I don't think they meant "no. of transactions" (as Peter R.) does, or "zero addresses" (as gbianchi does), but it sounded to my as if they looked at BTC transfer volume, or maybe even USD transfer volume.

I'll see if I can find the details....