Neural Net Side Project Makes $3500/mo Trading Stocks

Let me tell you about the time I caught the "finance bug" after watching Wall Street film in 2007.

I had a near 99% accurate scalping system. I could not believe it. Every trade I was making was profitable. How could this be?

The system worked without a stop loss. Meaning 1% of the time the prices turned, it wiped out all the profits (limited reward) accumulated.

I published all of these details on my university homepage which has since been deleted as they won't let alumni's keep one. But I was very careful that while 99% system accuracy is achievable, it's the downside risk that makes the system pretty fucking useless.

Of course you can put a stop loss but the system ran on "limited rewards unlimited risk". There's just no holy grail out there. Even if you read Hull's options book, no matter how complex your option strategy gets, you always have to be right at the end of the day. It's pure speculation and we all know how casino's work.

So knowing this, if it sounds too good to be true, there's a very strong reason for it. Retail trading is a zero sum game. Even more suspicious when the downsides are conveniently left to "if I gave it away system won't work". I was fully transparent about the downsides when I published my trading system.

The last thing I wanted was to people without the financial literacy to naively believe that a holy grail trading system existed. They don't exist. History is filled with funds that blew up because they believed alpha generation would be a constant thing.

Even quant trades I question their success. In the long run, all trades are speculative and carry risks (except arbitrages which are automated to oblivion). No matter how sophisticated you get the risks are always unknown.

Reposting my comment on the first HN submission of this page

"The claims made in this interview are extremely suspect, it just not make sense. Absolutely no relevant details are included. The developer claims he was able to built an AI trading strategy that is profitable 95% of the time. No technical details about the strategy or platform for trading is provided. A few trading buzzwords thrown in a few places. The rest of the interview is platitudes and inspirational hacker talk

I think IndieHackers needs to investigate the claims and be provided proof, otherwise this appears to be a fake project for the developer's own publicity. If IndieHackers are fine with that, I will stop visiting as I cannot trust that the content is not just shallow, exaggerated claims to raise peoples profiles."

After expressing my concerns directly to IndieHackers, the interview was changed in multiple places to flesh out a number of the spurious claims which I had criticised.

Without further details or some proof from the developer, this piece stinks. Taking into account the rest of the interview which contains stereotypical inspirational hacker talk, the whole thing feels like a badly done promotional bit for SV style status.

If I am wrong, I will apologise and the developer is going to be a multimillionaire very, very quickly. Until there is more details, I am going to be totally sceptical of IndieHacker interviews going forward. Which is a shame as I think IndieHackers is a great site. How can I trust that the other submissions are not baseless, PR pieces?

I may be way to cynical for my own good, but there's no proof that this is actually real, right? Besides a couple of pretty bizarre screenshots [1]. I'm skeptical because a) these types of gains in public markets are pretty much unheard of, and b) faking a story like this would be a fairly easy way of getting lots of attention for your freelancer business (which this guy advertises right at the top of the article).

[1] Running cat on a csv file would usually just print a list of numbers, yes? Instead he gets ascii art with months on the vertical axis, making it seem like time going back and forth, and some random commentary in the right margin. I dunno but something about it screams "mockup". As does the minimalist, ultra-stylish AWS window.

Edit: also, if it were me who got incredible returns from trading at high speed on indian markets using only machine learning on historical data, I would definitely keep quiet about it. Just sayin'...

Publication bias is a gigantic problem for amateur FinTech.

How many hundreds of engineers tried the same thing and won't get an article on the front page of HN because no one wants to read about mediocre performance? I don't want to sound too skeptical, quantitative HFT is absolutely the future, but if enough people try something that's effectively gambling, some of them are bound to succeed, and those people will get dramatically disproportionate exposure.

EDIT: I do still like this post and the project, the infrastructure is great and it offers insight into something I'm sure many people are curious about.