I code software developer for 33 years are, do a bit with MT4, but I trade to get away from coding and don't really believe in automated systems, don't think the market is that easy. Human brain required IMHO!!
Tend to agree with you Turveyd, particularly with respect to automated rule-based signals which should become predictable and thus easier to game, however I suspect that response-time is also an important element at play here. However, I am also interested about the value of programming perhaps in generating signals/ideas from which point a *human* may then able to differentiate from.
Just to be clear all the information I gave was based on SUCCESSFUL traders who code and trade. But I don't know of any that do that and have not succeeded. So, it's possible some people can code or know how, but do not mention it to me or I didn't know because what they did wasn't effective.
Absolutely trdes - awesome that you were able to highlight your own availability and survivorship biases But certainly an encouraging start, "I don't know of any that do that and have not succeeded."
Rust sounds very exciting but from what I've read, libraries aren't there yet. It's likely to be the most important language in a decade or so though.
Two different things: 1. Code to test strategies, backtesting. 2. Code to automate trading. Which one are you talking about? I recently attended Coursera, learned VBA-excel to code BSM to backtest. But don't have a clue how to code for auto-trade. I won't call myself a successful trader, just a lucky trader so I should not be part of your statistics. None of the strategies I studied worked for me.
Truly, both wave, but as d08 and others point out, there are different levels of scope depending on what the trader's objectives are, ie. research and backtesting, automated execution, signal generation etc.