Topic: Monthly average USD/bitcoin price & trend - page 16. (Read 118205 times)

Risto, do you advice me to follow your SSS plan- buy despite we are over the long trend? I got interested again  in bitcoins when the price was around 220 at bitstamp (news about first Bitcoin ATM published in Estonia). After that i made Bitstamp and tried to get verified. Took me 4 times and finaly the price was around 770. Made my first buy at 82x and now my average is 930. At the moment i have 50 per cent of fiat waiting in bitstamp of what i have already invested in bitcoins. I am not sure if and how much i am able to invest in the future. I just do not want to loose potential ROI as i have done so far.

I use a linear model lm() on the transformed prices.


As stated before, I'm a R n00b so this might be wrong.

That is not why people are buying it. That purpose is not what is driving the emotional need to buy before it is too late.

The nature of currency is precisely that it doesn't matter when you buy it.


It is either overextended and we come back to some rational adoption S-curve for use as a currency, or I am correct that the driving force is the notion of hyperinflation has begun and there is a mad spiraling race to see who can dump fiat the fastest and get into BTC before it is too late.

We should be able to watch the price and observe which mode we are in estimate the exponential function best approximating the price rise, but this can do nothing to tell us about the shape of the subsequent (looming) waterfall crash except we can estimate when the expected portion of the world's net worth will be in already.

It is impossible for the purchases ("adoption") to be rational w.r.t. to performance of the currency platform aspect. It can tell us nothing about the potential as a currency. We must first go through an exponential speculative rise and crash. The remaining question is what is the exponential curve of the price action. It can't be an S-curve, because of the nature of the price action is the antithesis of currency adoption. The exponential rate of price appreciation dwarfs any one buying it ONLY to use it as a currency. I am extremely confident that you can do any survey you want and it will confirm my assertion.

I don't think all people will succumb to that mania, so I think the limiting headroom is far below the net worth of the world.

If we want to measure the currency adoption, we need to use different metrics other than the price. And even transactions growth is overstated, because people are using a multitude of transactions to make themselves anonymous via coin mixers and trading back and forth between altcoins. Not to mention the speculative trading transactions, the SatoshiDice dust (useless/abuse) transactions, etc.

There may be significant currency adoption but it can't be significant compared the capital pouring in for the price appreciation.


Where is the math? What fitting curve are you employing?

I think the graph in OP is made with Excel. I posted an updated graph created with R including its source code.

https://bitcointalksearch.org/topic/m.3821474

S-curve is very applicable to Bitcoin's adoption as a payment technology "use as a currency".

Don't everyone think that the "store-of-value" aspect was a little overextended when the average bitcoin stash per owner was $12,000 a week ago?

(If not, then just scale it to the adult population of the world and we get $54 trillion bitcoin market cap when all are in (9 times that of gold). Sounds like what we've been dreaming, ergo must be true.) 

Is it true that the end stage of mania can go geometrically super-linear, meaning an acceleration of the percentage rate growth in price?

I expect we've moved into bubble blowup phase transition already, because this is psychologically to the internet as the invention of the combustion engine was to gasoline. Before that we were burning whale oil for lamp lighting.

As stated upthread, I think the logistic S curve is inapplicable to a mania. Rather it is applicable to adoption of technology for efficiency reasons, e.g. modern appliances and computers. Bitcoin adoption is driven not for primarily for efficiency although one can make arguments for and against Bitcoin being more efficient, rather for innate survival instincts that resolve around saving for the lean times, i.e. the notion that one's savings could be devalued to nothing by not buying it.

What program did you use to make that graph? I'm interested in performing some bitcoin and litecoin trend analysis and I'm trying to find more tools to look at the data with. Send me a PM or respond in the thread please.

Bitcoin may not necessarily have a traditional single S-curve hockey stick, because there are a number of completely different markets it competes for that could achieve critical mass over completely independent timeframes.

I wouldn't be so sure about that.

Almost by definition, it will be fast.

correct, and so far we still seem to be in the first half, and it will likely take a while before we reach the second half.

The accelerating part of the S-curve is only ever exponential growth. It just starts to feel "vertical" to us humans at some point. It doesn't ever go faster than exponential growth (in the ideal curve). The first half is exponential growth and the second half is logarithmic leveling off.

Not exactly a new paradigm, but if China had never exploded (in its own much faster exponential move) we would still expect the exponential trendline to hold. But China did explode and is now a major player. Unless we expect China to go back to where it was 8 months ago (x4 for slower exponential gain on the main trendline), shouldn't we count China's jump-in as a one-time bump to the trendline? Perhaps 30% up? That would mean we're still less than 2x the trend and could go much higher in this bubble.

An orderly trend is just way too boring! 

I guess boring is predictable though and there is some comfort to be had in that for sure.  Strange as it might sound, most people would probably prefer a more orderly progression.  Unpredictability is hard for most to stomach.

Where were we 30 days ago? Oh yes, $290, and just had crossed the ATH and everybody felt so good.

Now, $912 seems "cheap" to them who would like it go to the moon before Christmas.

There is still so much air in the bubble. Hard to know, which outcome I would like most, ch00 ch00 or orderly reversion to the trend.

Using all data from MtGox from bitcoincharts, the trend is 1 doubling every 120 days, 8x in a year, or 0.571% per day.  Also, the current expected value from the trend is $250.  In an equation this gives:

$ = 250*1.00571^{days from today}  
$ = 250*1.186^{months from now}
$ = 250*8^{years from now}

Only 8 x 250 = $2000 next December.  But I think the complete change in trust the past few month is a 1-time bonus that shifts the trend up without affecting the long-term growth rate.  If $900 is fair right now, then the equation is

$ = 900*1.00571^{days from today}  
$ = 900*1.186^{months from now}
$ = 900*8^{years from now}

or 8x900 = $5,600 next December.

The theoretical limit is $1 million per coin based on the world money supply.

Imagine the influx if some banks want to hold it as an asset.  All that Fed money buying up toxic assets has got to go somewhere.  It has not been going to M2 which has been stable.  They're just shoring up and dumping their toxic waste.  The bank crime against taxpayers is on going, and people act confused as to why inflation is not higher.  Banks are keeping it, not distributing it.

If we truly are going to the "more vertical" part of the S-curve, there's a possibility that the growth is so fast that it goes toward parabolic rather than logarithmic.  So we may be seeing that uptick OR we may be WAY ahead of where we should be still (although $500 seems VERY firm at this point).

You just need a new and improved chart next to your desk showing growth to $1,000,000 next year.    Hey, it is probably more accurate then the $1300 potential value for BTC that Bank of America somehow came up with last week!

New paradigm?  Sounds interesting.    If we have to adjust the trendline does that just mean that we are going to have an even higher or faster growth then we even expected based on the older charts?

I enjoy looking at the charts but I am in no way capable of making one.  If anyone wants to play with the numbers and throw up a new one with an adjusted growth rate based on this month's data it would be cool to look at.  But of course, it just might get my hopes up too much!

I'm personally leaning toward the "revert to the trend" scenario, but suspect that it will get there mostly through time - rather than much greater price drops. Of course we could be seeing an acceleration of the trend with all the breakthrough acceptance BTC has achieved in the last month. If so I will be sad to let go of the current trend chart (even in exchange for a more aggressive line), as it has been very comforting to have next to my desk. But BTC keeps blowing my mind - and that is the name of the game - so I guess no reason to stop now.

The problem is that the "crash" so far has not come even close the trendline. So either it was not a crash at all, or we will be visiting the trendline soon (down), or there is a new paradigm in bitcoin price (such as the one of growth spurts without deep crashes in 7/2010-6/2011).

It quacks like a crash. So my bets are hedged whether to visit the trend or continue higher in a sense that invalidates the trend.