Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: TrendFi – I built AI trading signals that self-optimize

trend.fi

Post reply on HN
Print view View original

11–20 of 56 posts

Re: Show HN: TrendFi – I built AI trading signals that self-optimize

#11

Post by wolfman1 »

bko wrote: post #8

I'm very skeptical of systems that claim to help with trading signals. Primarily because if you had a system, then you would use it yourself. Trading is pretty much a zero sum game especially when it comes to signals. If you had a profitable signal and you publicized it, it would disappear almost immediately. If your performance is correct and you're getting 200% per year, why sell the sauce for $20 a month? I would…

This is typically true for day trading signal services. I'm very skeptical of them too. I've tried many of them and none worked for me. There are a lot of scams.

The best model I tried was from The Technical Traders. They offer a trend based model for SPX/QQQ and they consistently outperform buy and hold for 10+ years now.

We were inspired by their service and wanted to take a modern spin on trend based models with a more affordable price point (they charge $2500yr).

Trend changes don't happen all that frequently. Therefore, they wont disappear immediately either. Timing trend changes correctly can make a huge difference.

Re: Show HN: TrendFi – I built AI trading signals that self-optimize

#12

Post by deadbabe »

Vaslo wrote: post #10
cortesoft wrote: post #5

Earlier quoted context omitted.

Usually, the reason is you lack capital to trade on your signals yourself.

Seems like you could build the capital fairly quickly if your system works. And you should be able to have people invest in you if you show results.

Money now is better than money later. If you have a secret that is guaranteed to make money, but only over the course of several years, you could just sell it now and get that profit today instead of waiting.

The person buying the secret will pay less than the long term profit as a fee for them giving you the money upfront now.

But if you have rights to resell the secret multiple times you will earn money far quicker, perhaps more than the secret’s long term value.

Re: Show HN: TrendFi – I built AI trading signals that self-optimize

#13

Post by ImPostingOnHN »

wolfman1 wrote: post #11
bko wrote: post #8

I'm very skeptical of systems that claim to help with trading signals. Primarily because if you had a system, then you would use it yourself. Trading is pretty much a zero sum game especially when it comes to signals. If you had a profitable signal and you publicized it, it would disappear almost immediately. If your performance is correct and you're getting 200% per year, why sell the sauce for $20 a month? I would…

This is typically true for day trading signal services. I'm very skeptical of them too. I've tried many of them and none worked for me. There are a lot of scams. The best model I tried was from The Technical Traders. They offer a trend based model for SPX/QQQ and they consistently outperform buy and hold for 10+ years now. We were inspired by their service and wanted to take a modern spin on trend based models with a…

> The best model I tried was from The Technical Traders. They offer a trend based model for SPX/QQQ and they consistently outperform buy and hold for 10+ years now.

- Has that been independently verified?

- Have they only been trading for 10 years?

- How do we make sure this isn't a case of chance + survivorship bias?

Re: Show HN: TrendFi – I built AI trading signals that self-optimize

#14

Post by mempko »

henning wrote: post #2

The usual response applies: if you have profitable trading signals, why would you sell it? Every time you sell a copy, that makes it less effective.

The market goes up exponentially over time. It's a mathematical certainty. The reason giving away trading signals is even possible is because the market isn't efficient. But not only that, there are forces behind prices beyond just speculation where signals can be correct over the long term. Also it takes a while to understand if particular signals are good outside of backtesting so there is no definitive 'these signals are perfect and everyone will use them'

In other words, markets have varying scales of information and are certaintly not efficient despite what academia has you believe. Trend following works precisely because there can be underlying reasons for the trend, and traders piling on just reinforces that trend. Where profitability of the signal depends finding robust trend changes as early as possible.

Re: Show HN: TrendFi – I built AI trading signals that self-optimize

#15

Post by wolfman1 »

bcyn wrote: post #9
wolfman1 wrote: post #6

Earlier quoted context omitted.

Most trading signals aim to exploit a secret edge. A temporary market inefficiency. The value is in the secret itself. Once you sell the secret, the small window of opportunity is competed away almost instantly. Trend based models are different. The "secret" isn't a secret at all. Think of a trend based trading strategy more as a disciplined methodology. Trend based models follow price and attempt to identify as earl…

This doesn't make as much sense as you think it does. If you could predictably trade a flip from bearish to bullish (for example, of course there are other trend-based signals), you would not share that signal because others would overcrowd your trade (by buying/shorting and moving the price more quickly towards the trending direction than you). A potential argument is that these signals are only applicable to a cert…

Overcrowding an entry on highly liquid assets is something that is so far from reality for our service.

Re: Show HN: TrendFi – I built AI trading signals that self-optimize

#16

Post by wolfman1 »

ImPostingOnHN wrote: post #13
wolfman1 wrote: post #11

Earlier quoted context omitted.

This is typically true for day trading signal services. I'm very skeptical of them too. I've tried many of them and none worked for me. There are a lot of scams. The best model I tried was from The Technical Traders. They offer a trend based model for SPX/QQQ and they consistently outperform buy and hold for 10+ years now. We were inspired by their service and wanted to take a modern spin on trend based models with a…

> The best model I tried was from The Technical Traders. They offer a trend based model for SPX/QQQ and they consistently outperform buy and hold for 10+ years now. - Has that been independently verified? - Have they only been trading for 10 years? - How do we make sure this isn't a case of chance + survivorship bias?

Not sure, but I personally used their service for 2y and they're the real deal. Looks like they've been around since 2008. So almost 20y now. The main guy that runs it (Chris V) is top notch. I'm not affiliate with them in anyway but I would highly recommend their service to anyone.

Re: Show HN: TrendFi – I built AI trading signals that self-optimize

#17

Post by ImPostingOnHN »

wolfman1 wrote: post #16
ImPostingOnHN wrote: post #13

Earlier quoted context omitted.

> The best model I tried was from The Technical Traders. They offer a trend based model for SPX/QQQ and they consistently outperform buy and hold for 10+ years now. - Has that been independently verified? - Have they only been trading for 10 years? - How do we make sure this isn't a case of chance + survivorship bias?

Not sure, but I personally used their service for 2y and they're the real deal. Looks like they've been around since 2008. So almost 20y now. The main guy that runs it (Chris V) is top notch. I'm not affiliate with them in anyway but I would highly recommend their service to anyone.

- From where, then, did you get the info that "they consistently outperform buy and hold for 10+ years now"?

- What does "they're the real deal" mean?

- Two years is not enough to test a strategy like this. 20 years would be better better, but it sounds like they already did that and the strategy fails half the time (hence the 10-year claim)? Does it fail randomly, or did they market a strategy that worked for the first 10 years, then didn't, and now have a new strategy with the same claims?

I'm asking these questions because people usually work hard for their money, and when someone makes (or relays) a claim that they can consistently multiply it better than expected, skepticism and scrutiny is imperative.

Re: Show HN: TrendFi – I built AI trading signals that self-optimize

#18

Post by runako »

Vaslo wrote: post #10
cortesoft wrote: post #5

Earlier quoted context omitted.

Usually, the reason is you lack capital to trade on your signals yourself.

Seems like you could build the capital fairly quickly if your system works. And you should be able to have people invest in you if you show results.

200%/yr starting from $10,000 => still needs a day job for a number of years.

This is why Wall Street is able to hire talented young traders, some of whom will develop profitable systems. Over time, some of them are able to amass enough savings and the track record necessary to get investors. But at t=0, somebody has to pay the rent.

Re: Show HN: TrendFi – I built AI trading signals that self-optimize

#19

Post by mdorazio »

Looking at the Performance page, this doesn’t seem that impressive to me. The profits are primarily driven by going long on coins during a crypto boom, plus some favorable Rivian and Nvidia trades.

If you exclude all the crypto related stuff (a lot of people here don’t want to touch that) the 2024 average individual trade return was a bit over 7%, strongly boosted by 3 outlier trades. Hard to calculate the fund performance for that subset though. Meanwhile if you bought and held MAGS over the same time period you would have gotten about a 1.67x return with long term cap gains tax benefit - probably a better total play than this model even with the crypto trades (especially if accounting for crypto trade fees).

Also there is no indication of YTD 2025 performance, which would be more indicative of model strength due to the March-April dip and slowdown in crypto + AI bull markets. I’ve been burned before by services like this and am not in a hurry to do so again.

Re: Show HN: TrendFi – I built AI trading signals that self-optimize

#20

Post by cheald »

cortesoft wrote: post #5
henning wrote: post #2

The usual response applies: if you have profitable trading signals, why would you sell it? Every time you sell a copy, that makes it less effective.

Usually, the reason is you lack capital to trade on your signals yourself.

If you have genuine reliable alpha, leverage exists. But you'd better make damn sure your edge is as good as you think it is.
Post reply on HN