*This post is not for you, Dear Reader. You will find the thoughts here disorganized. This post is for me to organize my thoughts. I have never read a page of Technical analysis and I am attempting some thoughts. To an experienced analyst, these will cause him to ROFL. Do yourself a favor and skip this.*

Okay, so I want to use technical analysis as a tool to speculate in BitCoins.

Belief: I believe technical analysis is a useful tool for me to know when to buy and sell and in what amounts.

Now let’s start of with some other assumptions I have

assumption#1. There is no commodity that can keep rising for ever regardless of the filter you apply on the data. At some point it has to fall

assumption#1b: A commodity can fall to zero and stay there for ever

assumption#2: If a commodity rises “too fast”, then a correction is due “soon”.

What does “too fast” mean?

a) dx/dt is at a global maxima?

b) dx/dt is at a sustained global maxima for the last few days?

c dx/dt is at a sustained global maxima as is X?

And how soon is “soon”? How much correction can we expect to see?

assumption#3: All things regress to the mean.

But what is the mean? For a commodity that has been traded for a long while, some benchmarks may be available. What benchmarks are available for something like BitCoin? What is the mean? The universal mean ofcourse is 0. But what is the short term mean? Is it even reasonable talking about something like this?

assumption#4: If it is doing something unexpected and strange, it will soon stop doing that.

How is strange defined? Something that deviates from the pattern. But how to define the pattern for something like Bitcoin which is so new and revolutionary. Use other similar commodities are templates? But which ones?

assumption#5: Technical analysis is not an exact science. Far from it.

assumption#6: Emotions are the big enemy of the speculator/trader. Define a set of trading rules and a time frame for following these rules. During this time, do not deviate from the rules ever for whatever reason. After this time period is over, you may formulate new rules from the lessons learnt. But during this time, do not deviate. This cuts out the emotions from the picture.

assumption#7: A simple algorithm is more likely to work than a complicated one. An inexperienced trader like me is better off following a simple algorithm at the start. Complexities can be added later

assumption#8: The volume of buying and selling I am doing will only have a “small” impact on the stock price. The volume I am dealing with a small fraction of the total trading volume.

Simplified trading rules

1. entry point for buying X units

2. If the price goes up, the selling point. All X units sold. Profit taking. Should this be on the upswing or the downswing?

3. If price goes down, at what point to sell. All X units sold. Stop Loss.

4. No action is taken if price stays between profit taking and stop loss levels

5. If profit taken, new entry point is the same. Should the entry be on the upswing or the downswing? And what sort of filter should we be applying on the data to define an upswing or a downswing? Stop loss and profit taking points remain the same.

6. If stop loss, then the new entry point is the old stop loss point. The new selling point is the old entry point. A new stop loss point needs to be defined.

7. Given that my long term view is bullish, should not the new entry point be higher than the old entry point for every profit taken? or at least the new selling point be higher than the old selling point? OR Should the selling and stop loss points stay the same during the defined time period?

8. Given that most chart patterns look so chaotic and noisy, is there any meaning to the words upswing and downswing? Probably not.

*********************************************************************

1. Now I decide to read something on investopedia. Apparently technical analysis has three assumptions

a. The market discounts everything: unspoken agreement

b. Prices move in trends: assumptions 2,3 and 4

c. history repeats: assumptions 2,3,4

2. I can already seem some limitations in technical analysis, i.e. Efficient market hypothesis: This is true for the long term, but not for the short term.

3. Trend: The trend for digital currency is upwards. How do I define a trend length?

Excellent observations. I feel somewhat similarly: TA is part science, part art, part voodoo and part gut feel. Observing a specific market over a period of time reveals certain patterns. These patterns remain consistent until they do not. The trick is to take chips off of the table and create “stop loss” provisions so that your downside is limited when the market turns against your presumptions.