rpietila Wall Observer - the Quality TA Thread ;)

[Its About Sharing]

My Gann chart is interesting. Worked extremely well in the past for me. Note how we bounced just as we were about to hit green resistance. The black and red lines above would be the next resistance levels.

looks reasonable, but why linear?

Great question and I'm afraid I've just fallen into the habit of using linear charts on the daily.
But, I have to play with it (Log) now, especially on my weekly chart.
Thanks for the idea!

BTW - I backtested different scales and found this one (6:1) to be the most reliable. But it varies.

Edit - I just tried and I can't lock the scale with the log chart. In other words, no way to reliably save the chart. And, it doesn't look good technically, at least regarding this move.

[SlipperySlope]

And the rally is looking better hour by hour. Here is the 3-day resolution chart from Bitstamp that shows the penetration of the resistance trendline that I drew back to the November 2013 peak.

I suppose that fiat holders waiting for prices to go even lower are buying back in a panic now. This happened to me last summer and I vowed never again. I am going to the the Bitcoin ATM to buy my daily fractional bitcoin.

[zimmah]

as you can see from this chart it has followed either the bottom or the middle downtrend line (that's a log chart btw) quite consistently. The stability of the past week has moved us away from the middle negative trendline and we are heading for the first resistance line (at about $508) which we will likely hit around the same time we will break the upper downtrend (which we will probably ignore in favor of the lower uptrend).

The top uptrend is so high it's not even visible on the chart, but it's there and it currently at about $3000 or so, but it will reach $5000ish in July, like many others have predicted.

As the price rises it will most likely rush towards the upper uptrend, like it always did till now.

If the price stays above 508.45 for a day or breaks the upper downtrend for a day i personally declare the bull run officially started. Especially if the volume of these moves is high.

Also note that I have drawn most of these trendlines weeks ago, but i have never published these charts because i never charted before and i wanted to make sure the trendlines were no fluke.

As you can see they are pretty damn accurate.

[SlipperySlope]

Also note that I have drawn most of these trendlines weeks ago, but i have never published these charts because i never charted before and i wanted to make sure the trendlines were no fluke.

As you can see they are pretty damn accurate.

Great start in your role as chartist. Plus it is much more fun to chart a breakout.

Finally we had a week in which the saying "this week is critical" actually held true.

[dnaleor]

Just found out that the market seems to react to the exponential downtrend trendline. That surprises me...
I really thought that "the crowd' would use the linear charts for TA and the log chart for long term



edit: compare this chart with the linear chart:

[zimmah]

Probably because a break from the log trend confirms that the trend is truly reversing while a linear break could also be caused by a random price movement.

Maybe they also kinda reinforce each other.

[dnaleor]

Maybe they also kinda reinforce each other.

that is indeed a possibility. Will give more attention to log charts for medium term from now on. I really thought the rally would start a few days sooner.
Now it seems we need to break a second log downtrend around 550 USD.
But i'm pretty confident it will happen. This is just the last hurdle

[bitfair]

Here's a fresh set of forecasts - hopefully self-explanatory:

[BitchicksHusband]

Here's a fresh set of forecasts - hopefully self-explanatory:

What happens when you apply this same forecast to the historical data before the previous run-ups?

Apply it to Jan 1, 2013 and see if you get $266 in April.

Apply it to September 1, 2013 and see if you get $1200 in December.

My guess is that this is too conservative, but I would like to see what you get.

[bitfair]

What happens when you apply this same forecast to the historical data before the previous run-ups?

Apply it to Jan 1, 2013 and see if you get $266 in April.

Apply it to September 1, 2013 and see if you get $1200 in December.

My guess is that this is too conservative, but I would like to see what you get.

I don't have time to do the exercise, but I can muse about the most likely results: the multiples predicted by my method should presumably be relatively stable, or at least they should change slowly. From Jan to Apr 2013 the price went up 20x, whereas this model would say there was a 50% chance of a 70% increase in three months. It would say that the 20x increase in price was a 1% probability event. But then again, the increase in price didn't really last either, so even if it happened it quickly unhappened...

But then again, you are asking about predicting the top of a bubble from the bottom of the valley, which is not the intention of the model - this model is intended to be more in the trend-department rather than the bubble prediction department. And in that respect, I think it is as good as any and better than most.

[Biodom]

What happens when you apply this same forecast to the historical data before the previous run-ups?

Apply it to Jan 1, 2013 and see if you get $266 in April.

Apply it to September 1, 2013 and see if you get $1200 in December.

My guess is that this is too conservative, but I would like to see what you get.

I don't have time to do the exercise, but I can muse about the most likely results: the multiples predicted by my method should presumably be relatively stable, or at least they should change slowly. From Jan to Apr 2013 the price went up 20x, whereas this model would say there was a 50% chance of a 70% increase in three months. It would say that the 20x increase in price was a 1% probability event. But then again, the increase in price didn't really last either, so even if it happened it quickly unhappened...

But then again, you are asking about predicting the top of a bubble from the bottom of the valley, which is not the intention of the model - this model is intended to be more in the trend-department rather than the bubble prediction department. And in that respect, I think it is as good as any and better than most.

Great model. Who can predict bubbles? Impossible.

[zimmah]

Here's a fresh set of forecasts - hopefully self-explanatory:

What happens when you apply this same forecast to the historical data before the previous run-ups?

Apply it to Jan 1, 2013 and see if you get $266 in April.

Apply it to September 1, 2013 and see if you get $1200 in December.

My guess is that this is too conservative, but I would like to see what you get.

seems much too conservative to me as well, and it seems to be a lot of work to create an accurate probability chart like that. If i have a lot of spare time this summer and don't know what else to do i might give it a try but i doubt it. Regular charting seems to be much faster.

[bitfair]

seems much too conservative to me as well, and it seems to be a lot of work to create an accurate probability chart like that. If i have a lot of spare time this summer and don't know what else to do i might give it a try but i doubt it. Regular charting seems to be much faster.

Only in Bitcoinland is a 10x price increase in one year labelled as "conservative"...

[zimmah]

seems much too conservative to me as well, and it seems to be a lot of work to create an accurate probability chart like that. If i have a lot of spare time this summer and don't know what else to do i might give it a try but i doubt it. Regular charting seems to be much faster.

Only in Bitcoinland is a 10x price increase in one year labelled as "conservative"...

yes, i know, makes you wonder why so many people are still skeptical.

Yesterday the price jumped $30 in like two hours or less and people were still saying "it's not a real price jump"

If that would happen on forex people would go insane.

[BitChick]

seems much too conservative to me as well, and it seems to be a lot of work to create an accurate probability chart like that. If i have a lot of spare time this summer and don't know what else to do i might give it a try but i doubt it. Regular charting seems to be much faster.

Only in Bitcoinland is a 10x price increase in one year labelled as "conservative"...

yes, i know, makes you wonder why so many people are still skeptical.

Yesterday the price jumped $30 in like two hours or less and people were still saying "it's not a real price jump"

If that would happen on forex people would go insane.

$30 in two hours.  If we had just $10 per two hours for next week the price would be over $1300 in 7 days. 

When Bitcoin decides to move up it can happen insanely fast.  I was just thinking that with a greater number of people involved now we could even see larger jumps in price than we did in November. 

[phatsphere]

Here's a fresh set of forecasts - hopefully self-explanatory:

Is the code somewhere online? github/bitbucket?

[rpietila]

My 2H-volume based technical trading algorithm made a call this morning. Switched from BEAR to BULL.

It had been a BEAR since 2014-4-16, at the price of $515.

I think it is about time for the rest of us to switch to bulls also. The risk/reward ratio from being a bear in these circumstances is declining rapidly.

It is possible that the deck was arranged for a dip last weekend (I also noticed it in the charts and warned you). But since nobody panicked, some of the players took action on the greener side.

I tried to send a longer treatise on the possible developments in price yesterday, but since I was on a cruise ship, the internet didn't work properly. I think I'll scrap that one, because of the indicator going long, and good news just cascading on the face of price being -60% below trend. So now I advice to buy for all timeframes from 1 month to long term.

Maybe, but I still think there will be another dip below 400.

^^-- I think we have run out of suckers to accomplish that.

The last moves that totalled 10.5% in 2 days were carried out with a total volume of 32kBTC in Bitstamp. If we assume that the global net demand volume was twice that (very conservative indeed), it means that 64kBTC, which is 0.52% of the bitcoin market cap, was bought with the effect of 10.5% increase in market cap. This is 20:1.

To raise bitcoin price to the level that many dream of, $7000, only $4 billion is needed if the market-cap-increase-leverage is 20:1.

We are talking about 1.7 million bitcoins changing hands during the uptrend. That is 14% of all bitcoins.

According to my research, the average sum a person would invest in bitcoins is $1000. Therefore, 4 million new users are needed in addition to the 1 million current ones to reach $7000.

[wachtwoord]

My 2H-volume based technical trading algorithm made a call this morning. Switched from BEAR to BULL.

It had been a BEAR since 2014-4-16, at the price of $515.

I think it is about time for the rest of us to switch to bulls also. The risk/reward ratio from being a bear in these circumstances is declining rapidly.

It is possible that the deck was arranged for a dip last weekend (I also noticed it in the charts and warned you). But since nobody panicked, some of the players took action on the greener side.

I tried to send a longer treatise on the possible developments in price yesterday, but since I was on a cruise ship, the internet didn't work properly. I think I'll scrap that one, because of the indicator going long, and good news just cascading on the face of price being -60% below trend. So now I advice to buy for all timeframes from 1 month to long term.

Maybe, but I still think there will be another dip below 400.

^^-- I think we have run out of suckers to accomplish that.

The last moves that totalled 10.5% in 2 days were carried out with a total volume of 32kBTC in Bitstamp. If we assume that the global net demand volume was twice that (very conservative indeed), it means that 64kBTC, which is 0.52% of the bitcoin market cap, was bought with the effect of 10.5% increase in market cap. This is 20:1.

To raise bitcoin price to the level that many dream of, $7000, only $4 billion is needed if the market-cap-increase-leverage is 20:1.

We are talking about 1.7 million bitcoins changing hands during the uptrend. That is 14% of all bitcoins.

According to my research, the average sum a person would invest in bitcoins is $1000. Therefore, 4 million new users are needed in addition to the 1 million current ones to reach $7000.

That seems like a lot to ask for the next 7 months. I think chances are some bigger investment funds are going to have a big impact though.

[Biodom]

To raise bitcoin price to the level that many dream of, $7000, only $4 billion is needed if the market-cap-increase-leverage is 20:1.

We are talking about 1.7 million bitcoins changing hands during the uptrend. That is 14% of all bitcoins.

According to my research, the average sum a person would invest in bitcoins is $1000. Therefore, 4 million new users are needed in addition to the 1 million current ones to reach $7000.

Who is going to be bitcoins Hunt brothers and will try to "corner" the market. As moderator quote shows-it is relatively easy to accomplish right now. Many hedge funds can swing $4 bil. All it takes is 4-5 medium-large hedge funds and a "strategy".

[Skele]

Here's a fresh set of forecasts - hopefully self-explanatory:

What happens when you apply this same forecast to the historical data before the previous run-ups?

Apply it to Jan 1, 2013 and see if you get $266 in April.

Apply it to September 1, 2013 and see if you get $1200 in December.

My guess is that this is too conservative, but I would like to see what you get.

Im sure that by the end of the month those graphics can become reality, we can stand up to as much as 5,000