|
|
|
|
|
|
by throwawaywego
2501 days ago
|
|
|
If you know that 5.6% of the users (automated, or parroting the preacher) of an exchange will use fibonacci retracements, and that 15% of the amateur market will follow the market price change caused by either buying or selling activity, then you can play roulette with a decent edge. Of course, not as much as the preacher, who is allowed to bet before (s)he will speak to his or her congregation. When you gather enough of these commonly used technical analysis, it's like having to predict in which startup Ron Conway will invest, but you can calculate Conway in a Python one-liner, and keep up-to-date by going to weekly sermon. |
|
|
I wish the technical analysis flock would merely incorporate different things. TA takes a time series chart and imagines time series patterns. It primarily neglects what didn't get printed on a time series chart, and why. How big are the orders at the resistance level, do you have a record of the order sizes that appeared at the last resistance level? who is selling at the resistance level and why? The TA answer is "just because its a psychologically round number for the resistance price" or "because thats how high the last high candle was", but you can greatly improve your win rate by understanding who is in the market and why, which is possible to understand and a large portion of my trading strategies. It can be much more data intensive though so I can see why 1980s gurus did not do it.