Hacker News new | ask | show | jobs
Ask HN: Automated bot trading is our future?
4 points by Kshitijmore 1492 days ago
I wants to know that automated bots trading have a good future or not.
2 comments

It is likely as they are already widely used.

> A study in 2019 showed that around 92% of trading in the Forex market was performed by trading algorithms rather than humans

https://en.m.wikipedia.org/wiki/Algorithmic_trading

I would like to know too, I have never seen an automated bot that is entirely profitable. Most of those that have been tried have over 60%+ losses.
Why would you expect to see a profitable bot?

Look at it from the other side. Suppose that I have a consistently profitable bot. What good does it do me to let you know about it?

If I sell it to you, the version that you run will cut into the profits that I make by running my own version. It's unlikely that you'll pay me enough to make up the difference.

If I give you enough details to make something yourself, the same issues arise (if you're successful), and you're going to pay me even less.

The only reason to let someone else know about a successful trading bot is to get resources from them for that bot.

Yes, most people who try to make trading bots fail.

That's because it's not enough to have a good strategy, you also must execute it well. In fact, the latter is more important than the former; a mediocre strategy executed well can be fairly profitable. A fantastic strategy implemented poorly will lose money.

It's sort of like "amateurs discuss tactics, professionals discuss logistics".

FWIW, "execute well" is much harder than you might think. There is a surprising amount of relevant detail.

But, don't let me stop you. Write or buy a bot. Trade. Heck, pay someone to trade for you. (Make sure that they actually trade and don't just steal.)

>Suppose that I have a consistently profitable bot. What good does it do me to let you know about it?

Two main reasons come to mind:

First, you cannot know whether a particular algorithm (bot) will be consistently profitable in the future. It may lose everything it has ever won if the circumstances change (e.g quickly rising interest rates, war breaking out). There's risk involved. A service provider may prefer making software to taking market risk.

Second, investing profitably without taking excessive risk requires far more capital than writing algorithms. The customers of a bot maker may be able to raise capital far more cheaply than the bot maker itself, e.g. by investing their personal savings. Whether a particular algorithm is profitable or not depends on the cost of capital.

Please don't take this as an endorsement of algorithmic trading or of investing your money that way. It's certainly not how I personally invest my savings. I'm not a finance professional either.