Hacker News Reader

Comment by jasfi | Hacker News Reader

I'm working on an automated crypto trader: https://tradecast.one.

However one of the motivations was to making crypto trading safer. The idea is to put in stop-losses and have some idea of when to close a trade. I was also thinking of adding other safety features, such as warnings when losses occur too quickly or too severely. I was even thinking of temporarily preventing trading if losses crossed a certain threshold. However, some traders could ditch the system if I do that.

Automating trading is also meant to reduce mistakes, especially those that come from trading fatigue. How many people have entered the wrong price, when not using market orders, and lost a ton of money?

Any feedback here on this idea? I'm not out to ruin anyone's financial well-being or mental health, the exact opposite in fact.

View on HN

> Any feedback here on this idea? I'm not out to ruin anyone's financial well-being or mental health, the exact opposite in fact.

Yes, as a person that's been in this since the 2nd Goxxing you can't really compete with the HFT bots back then as a privateer, the days of the willy bot collecting fat margins are all behind us since institutional money and corps with VC backing have entered this space.

With that said, I'd like to see returns which would have to span over 5+ years before I'd even consider using your service, the idea of just 'set it and forget it' in these markets is what created these hype bubbles and subsequent crash during the ICO bubble. Getting uninformed people into this space is not just wrong morally, it stifles progress long-term.

Granted 99% of alts are all pump and dump scams to begin with.

With that said, yeah trading is stressful, but you know what isn't 'holding your coins in cold storage' and it has been beating the traders with greater returns in all but the most recent quarter (assuming you bought at ATH).

You know what is also incredibly more stressful than that? Calculating your tax liability on short term capital gains and entering an incredibly opaque legal structure that even the IRS still hasn't completely clarified.

The service isn't meant to compete with HFT bots. There's still room to make a profit, in my opinion.

The trading system lets traders specify rules/parameters, there's no one algorithm. I want to implement copy trading, so that less sophisticated traders can simply copy, but also learn in the process.

Calculating your tax liability is kind of easy, you just download your statements and send them to your accountant/tax consultant.

You make it sound like finding an accountant/tax consultant that won't truck-drag your burn-rate is easy!

Be extremely careful with your server's security. I would never want to be responsible for someone else's exchange access keys. It's a honeypot if there ever was one. (Even with limited access rights) Focus on the "local Open Source client" even though that will probably be much harder to monetize. Anyway, building trading bots is a lot of fun and there are plenty of strategies that can make a decent return.

The security aspect is something I'm concerned about for sure. I have even considered making the Open Source client mandatory, but that could be overkill.

Reply on news.ycombinator.com