3
donuts
3y

Anyone have experience with technical analysis, algo trading?

Was reading an article about how markets are probably going to crash next year based on some technical indicators and it reminded me of the days back in uni when I tried to create (simple) trading algos using all sorts of technical indicators...

Never got anything better than random... And I recall was basically throw some indicators together and backtest using different parameters. I think it's called p-hacking.

But anyway just wondering how is it actually done such that it works... And well how can technical analysis ever work if it's completely subjective to the parameters u use? Like just take Moving Average. You can take a average over any range 1s, 1min, 1d, 1 month, etc.

So many combinations u could use, how can you make any rules that you are confident enough to put lots of money on...

Comments
  • 1
    >Was reading an article about how markets are probably going to crash next year based on some technical indicators

    Welcome to the last 10 years?
  • 0
    @N00bPancakes but you got high frequency algo trading too... So just reminded me how do they even find what indicators, parameters to use...
  • 1
    Sorry, I can't help with your original question, but I'm interested in 2021 crash article. Could you share more about it?
  • 3
    Technical analysis doesn’t make much sense, but it totally does work. I have no idea why all markets seem to follow the same rules and patterns, nor why those rules and patterns make no bloody sense. But it’s true, and they totally do.

    I’d recommend reading up on Gann as a starting place. (As in the person who came up with the Gann Fan, among others.)

    Also, basically every article you find online on what the market will do is going to be wrong. Why? People don’t know what they’re doing, and some do but intentionally publish incorrect information because they can profit off people who follow it. This is especially true of exchanges and those claiming to teach trading strategies. Notable example: the “technical analysis” needle on TradingView, which ranges from “strong sell” to “strong buy” is almost always exactly backwards. It takes a lot of effort to be wrong more than 95% of the time. Is it any wonder why the overwhelming majority of traders lose money? The entire thing is designed for them to do just that. Almost all of it is “fake news” because someone profits off of you losing money on trades. Doesn’t matter if you’re paying them to teach you. God, fucking con artists can get bent.

    Anyway. Another great place to start is Anton Kriel. Imagine a Ramsay of trading; that’s him. Tells it like it is, doesn’t put up with bullshit, calls everyone else out on their bullshit, and explains how all the bullshit works, and what to do about it. He founded a trading institute and sells bloody expensive classes that actually have valid, useful information. One of the lessons in ... I forget which risk management class changed my life in trading, and we’ve made significantly more money since. (Box A vs Box B)

    But seriously, apart from him, the only place you are likely to find factual information is in the boring as hell books on technical analysis. The older and more boring the book, the better.
  • 1
    I was actually reading into the matter last year and I found out that a lot of the technical indicators make a lot of sense if you realize that the stock market is one big psychological game.
    The indicators probably work because almost everyone else uses them as well. The market is one big positive feedback loop: you see people are buying? You probably buy too. You see a lot of stocks being sold? You're gonna sell as well to cut losses/make quick profit before you're too late.
    And yeah, risk management is the way to prevent huge losses. Make sure to spread and only about 5% of your budget into one stock or something like that I guess
  • 1
    It’s only working in a short term investments like HFT / day trading or something.
    There are like 2 I know from forex that were popular among analysts. Those are Ichimoku, triangle / harmonic pattern.
    In a long term it’s only a helper that shouldn’t impact trading to much.
  • 1
    The biggest trick is not working with your own money, but other people's, and drawing a commission.
Add Comment