Hacker News Reader

Comment by etrautmann | Hacker News Reader

Technical analysis? Is there anything statistically robust about that or is it reading time series tea leaves?

Part of the reason why technical analysis fails so bad on stocks is because stocks are influenced by non-technical things (like profits & earnings) . Bitcoin is not like that. It's price is purely a function of supply and demand. So this makes technical analysis more useful, but this is not saying it's doable or easy. Just easier, because you have eliminated a variable. I think there are certain patterns that have repeated. The long bottoming out process is one of them.

TA is the primary form of analysis for all financial markets. The idea is that the price contains all of the information, hidden on publicly available. It's a requirement for taking the CFA.It's widely accepted by regulators, academics, and industry professionals.

The only robust thing about TA is that it's an astrology for men :)

It certainly works for many traders who consistently outperform the market by trading stuff like FX and Crypto. TA and trading price action can also be a tool to limit risk.

Btw, selling plans of executives often involve very trivial algorithms, such as "sell when price above VWAP till price reaches <some lower bound> for 30 days or 20.000 shares sold". (Some brokerages will try to create additional liquidity in the meantime, depending on what you want and how much your security is traded).

Correlation doesn't mean causation. Hamsters and octopuses have outperformed hedge funds in experiments. You can outperform the market by throwing a coin to make decisions. TA has zero scientific basis and there are no proofs for any analysis TA people make.

Reply on news.ycombinator.com