They're all based on historical results and often leave out crucial fees and spread that you would incur in high volatility environment. Nor they account for slipping, missing the bus, buying late etc.
A- if it's to good to be true, it is
B- in a gold rush the people getting consistently rich are the ones selling shovels and buckets.
I'm sure technically wise this app is an achievement and I don't want to take away from that, but when it comes to trading algos: ideas are cheap, execution is everything. And you won't beat the big boys at this game.
The front-and-center portfolio is the "Boglehead" three-fund portfolio, which is stock/bond index funds. It's not exactly the too-good-to-be-true gold rush model.
Certainly is! Which is why live-trading isn't enabled just yet. This library is my personal one, but the moment you provide investment advice or manage other's money you run into regulation.
This seems like a “problem” that shouldn’t be solved. If the thing keeping you from algorithmic trading is difficulty in deploying, you probably have no business doing it.
Cynically, this seems like a way to draw more naive novices into the market. As the old saying goes, if you don’t know who the mark is, it’s you.
This is a collection of my investment modules - strategies codified from my experience in quantitative investing and software development. All categorized into a short compendium.
These models (algo-investors as I call them) serve different purposes, some just remove the stress and emotion of maintaining a portfolio (automated rebalancing, etc), some seek to capitalize on binary events (like elections or regulation changes), some seek to minimize volatility, some seek to "beat the market" (leveraged modules, hedge fund trackers), and some are just fun (wallstreetsbets sentiment tracker).
As some have pointed out: these results DO account for slippage, and many are showing actual live results (not just backtests).
This is a highly regulated environment (I'm US-based), and while not yet registered (in the process), hence why live trading isn't allowed yet, the intention is to have live trading available as soon as possible.
I recommend the SPY algorithm. Go to whatever brokerage you like and dollar cost average by putting in 1/365 your budget daily in SPY.
Zero fees, and you can set it and forget it!
In all seriousness the Boglehead strategy looks neat, I'll take a look. Is it difficult to start an ETF? Seems like all of these should just be ETFs you can purchase at any brokerage.
Glad you recognize this fact. Love explaining the SPY as "rank public equities by revenue, take the top 500, then weight by market capitalization".
Starting an ETF is highly regulated, this differs in that these algorithms just manage your money interfacing with the public market rather than moving in ETF. Less regulation, but encounters tax implications (hence the less frequent trading of these models) among other things.
If we're talking about long term investment than dollar cost averaging is about the same as lump sum on January 1st.
As for starting an ETF... Well minimal capital requirement is 100k, to launch it's probably 300k total investment. Then there is a lot of legal requirements and obligations. Whole process is at least 6 months
I guess it depends on what's your goal. Some brokerages let you create a bundle of stocks and share them publicly or privately. Which is sound more like what you want.
Ticker data is expensive! Many of these models have their results from live data, but some are just backtests. Relatively inexpensive datasets for backtesting at the ticker level start at 2008 (unfortunately, obviously, given that considering a flashpoint like that is important for rigorously confirming a model works).
As soon as feasible, I want to purchase the larger datasets that often go as far back as 1985 for a more comprehensive picture.
It would be nice. My first impression was that the selected models seems very risky and would realy depend on the last decade which was probably the best in 70 years. I would like to see how they would have performed during a crisis (and bonus point for using a log scale on long duration to avoid minimizing the past)
My 2cents: to lower the onboarding, would be great to set it to get notification about what /would/ be bought or sold; eventually users will trust the system and move some of their money to it.
This is the thing I always think about with these kinds of systems, whether someone else's advice or do-it-yourself algos or "go here to find this information".
If you follow the advice and record the trades you would have made, you can get an idea of if the advice is solid, and you can also get a hang for not panic-selling your first dip due to new-investor jitters. It also means that you won't follow crazy trends (like impulse-dumping everything into GME) because it's so obviously an outlier that it won't give you any useful data anyway.
When I was in high school there was a faux-stock-trading website, they gave you ten thousand dollars in fake money and you could fake-buy stocks with it. A great way to play around with, and learn about, the stock market with zero actual risk or investment required.
I ended up putting everything in Diamond Multimedia at $5 and pulling it out the next day at $6, then dumping the rest into eBay (within a month after their IPO IIRC) and forgetting about the website until one day I logged in and my investment was 10x. I patted myself on the back and never invested anything in anything again.
Yea, I make a tool that does this. So, my friends who use it just see the historical Actions (buy/sell) and price points.
And one "trick" for backtesting is overlay SPX or SPY or DIA (or other favorite) with the results. Maybe it's obvious but folks new to the game are really helped by a comp to the market visual.
If they implement paper trading (as their survey suggests they're thinking about) then that should build the same level of trust. If your fake money is growing then you'll want to put real money in.
Most of these type things (quantopian is one I've tried) offer "paper trading" that is just that. You simulate it on "paper" until you gain confidence and use it to invest. I think the docs for this says they do as well
I'm a little confused. I created an account just for kicks and it immediately said there were 58,912 people ahead of me and 2,877 behind me. This means that 2,877 people signed up for the service in the time between my account being created and the page rendering?
But the number hasn't gone up at all, so me and the other 2,877 people were I guess part of a huge surge of signups all within the same 100ms or so, and no one else has signed up since?
This is quite confusing, and does not encourage trust.
Sorry, could certainly clarify this better. Everyone who hasn't completed any actions falls into the same batch. Providing feedback or referring moves you to the next batch.
In other words, those 2,877 people are others who signed up and didn't complete any further actions, and are all in the same batch for access. Hope this makes more sense.
I would be careful when playing around with a viral queue mechanism. You need to be very clear (even if only for yourself) with your own incentives why you're using the strategy.
It can work well if you're filtering for high-quality signal in your own feedback loop. But it can also backfire badly if the intention is to generate hype amplification.
Don’t know how to edit comments, but ahead/behind numbers I’ve made before can be difficult to implement based on the actions you want to incentivize so they may just be manually updated after reviewing
I gotta be really honest with you, the title is very catchy but when I open the website, I don't find any detail regarding how this would work, what exactly is this. All I know is it's got some algo trading modules but what next?
Like, what is this thing? Is this legal? Is this just more like a compilation of strategies that tell the user what to invest in or is this like a bot like it says it is in which case, the users would need to trust this thing to invest using an algo-trader bot deployed by a venture no one's really heard of.
If this is a project, then great, the concept sounds pretty cool but if this is more like an actual startup, then the path is gnarly, since there are well-trusted platforms already out there including:
I'm not trying to say your platform is not good per se or anything but if it's meant to be a startup, then it's not doing a particularly great job at it as of now.
It'd be great if you could answer this for me ( if this is a startup and not a project for fun ) :
Are you aware that in order for your platform to be trusted, it doesn't need to be slightly better than the competition but better on a large magnitude? If you know so, then what makes your platform way better than other platforms out there?
I appreciate the feedback (rereading the site I can see the confusion), but I think concluding this is “doing poorly as a startup” is a little premature and something you can’t accurately judge, especially since from your submission history it looks like you have little in the way of a successful startup background when this field is very much a learn by doing one.
This is closer to a roboadvisor than those niche hobby brokerages. The massive difference is those all have incredible barriers to entry - knowing how to code, or at least knowing how to build a profitable automated investment strategy. These are all turnkey, transparent, and no code customizable. Regular investors can use these, only a few can properly take advantage of those you listed. Most of my users came from Fidelity, Schwab, Betterment, and so on. Not Quantconnect (though plenty have come from there and others).
Compared to roboadvisors: these models are more far more innovative and diverse. Use crypto or quantitative strategies made by professionals or curated from other users instead of standard stock bond allocations. Even innovative roboadvisors like m1finance only go as far as hedge fund followers. The difference is huge - the goal here is to be your platform for higher risk and higher return investing, for the 15-20% of your portfolio you want to go further at the risk of larger drawdowns.
I was where you were on my entrepreneurial journey recently, so my best advice is just to do. “10x improvements” and other academics they teach you in SUS is super useful (and often true), but until actually applying the learning, finding what works for you in the current market, you’re just repeating catchphrases without understanding them. You likely won’t even understand the magnitude of an improvement until you do your research and try things out, so best to focus on that when your idea has been validated first.
For your reference: this scrappy site with no optimizations, no contact, no how it works, no fluff, has already garnered several millions of $ in pledged user investment, has profitable and repeatable acquisition channels with an on page organic conversion rate of a large fraction of visitors, has collected thousands of data points, surveys, and interviews towards product market fit and all aspects of the model from target customers to revenue streams, and all this only a few weeks (late December) into prelaunch. I added only what was important to my preexisting library and put it out there. All my research and customer feedback so far suggests that this is a magnitude of improvement, and many are desperate for a platform like this.
Hope that clears it up and thanks again for the feedback!
- Based on my submission history, it can be deduced that I don't have a successful startup history and that's absolutely true but the questions I asked you were from my understanding of YC SUS and how successful startups usually work, I'm not saying my understanding is 100% correct but it is, to some extent and it has little to do with anyone's startup background. I don't mean this as an excuse in any way but I'm 17 right now and I believe that starting up defensively by throwing that I don't have a successful startup isn't the most respectful thing when you could've just politely answered my questions ( or not ).
- I never claimed that your startup is doing poorly, I merely pointed it's not doing a "great job" *as of now* from what I could tell by evaluating the points from your competition and your offering. But, I wasn't totally sure of that either, hence wanted you to answer those questions for me ( if this was actually an startup, which as it turns out, it is ). And, thanks for answering those questions.
Again, I mean no disrespect to you or your startup.
I, as a user of your website, felt that there was stuff left unclear or missing. If you're still gaining massive user growth, honestly, good for you!
Although, if I'm your potential user, I think you should work on the feedback I gave, be more lucid and transparent with your product.
Otherwise, you should happily discard what I have to say, since that'd legitimately not be the best use of your time.
Also, good luck with your startup, hope it gets even more traction and becomes usable, transparent, and clear enough for a user like me :))
P.S As for the questions about your startup, I think I got my answers. If you were a little less defensive with the initial part, it would've been nice too :)
I began my comment by pointing out how your comment was disrespectful, and was blunt it making that point, but I digress.
My main point is that you misapplied those otherwise true principles from YC SUS because you haven't had enough practice with them. Quantconnect and Streak have totally different customers and value proposition - they aren't comparable to Quantbase, and so your entire analysis is incorrect at the start from faulty assumptions (in part perhaps because I need to further clarify the value prop - a marginal improvement).
A 10x improvement is a great rule of thumb because the costs of switching and trust is high - which means that magnitude improvement is relative to the alternatives, and is in the direction of an actual value proposition. People don't use Quantconnect because they have a great How It Works explanation, they use it for the community and trading API. Likewise, missing a How It Works section on my site - while a great suggestion and one I'll eventually add - is likely just a marginal improvement (because it isn't a core value prop) and certainly would be a waste of time for validation (unless my site was too confusing to understand in the first place for most people so my value prop isn't being communicated). Because Quantconnect and Quantbase aren’t alternatives, and How It Works sections aren’t in the direction of a value proposition for either, mentioning the need for a 10x magnitude improvement from a non competitor suggesting a non value prop is just non sensical.
Again, I appreciate your feedback, you just came across as disrespectful by jumping to conclusions on incorrect applications of principles and that's what I took issue with. YC SUS is an awesome resource, but make sure you understand their teachings (by testing them yourself and building your own startups), otherwise you may misapply them.
This is just wrong, none of those have anything to do with this website from the looks of it. Also, how can you judge a startups health with a cursory look at the site?
Well I read whatever was on the outlet of that startup and asked my questions and presented my observations on that, I didn't claim them to be right, the questions were to clarify those claims, that's the whole point of discussions.
49 comments
[ 3.0 ms ] story [ 106 ms ] threadBut be very careful of such apps.
They're all based on historical results and often leave out crucial fees and spread that you would incur in high volatility environment. Nor they account for slipping, missing the bus, buying late etc.
A- if it's to good to be true, it is
B- in a gold rush the people getting consistently rich are the ones selling shovels and buckets.
I'm sure technically wise this app is an achievement and I don't want to take away from that, but when it comes to trading algos: ideas are cheap, execution is everything. And you won't beat the big boys at this game.
Cynically, this seems like a way to draw more naive novices into the market. As the old saying goes, if you don’t know who the mark is, it’s you.
This is a collection of my investment modules - strategies codified from my experience in quantitative investing and software development. All categorized into a short compendium.
These models (algo-investors as I call them) serve different purposes, some just remove the stress and emotion of maintaining a portfolio (automated rebalancing, etc), some seek to capitalize on binary events (like elections or regulation changes), some seek to minimize volatility, some seek to "beat the market" (leveraged modules, hedge fund trackers), and some are just fun (wallstreetsbets sentiment tracker).
As some have pointed out: these results DO account for slippage, and many are showing actual live results (not just backtests).
This is a highly regulated environment (I'm US-based), and while not yet registered (in the process), hence why live trading isn't allowed yet, the intention is to have live trading available as soon as possible.
Zero fees, and you can set it and forget it!
In all seriousness the Boglehead strategy looks neat, I'll take a look. Is it difficult to start an ETF? Seems like all of these should just be ETFs you can purchase at any brokerage.
Starting an ETF is highly regulated, this differs in that these algorithms just manage your money interfacing with the public market rather than moving in ETF. Less regulation, but encounters tax implications (hence the less frequent trading of these models) among other things.
As for starting an ETF... Well minimal capital requirement is 100k, to launch it's probably 300k total investment. Then there is a lot of legal requirements and obligations. Whole process is at least 6 months
I guess it depends on what's your goal. Some brokerages let you create a bundle of stocks and share them publicly or privately. Which is sound more like what you want.
As soon as feasible, I want to purchase the larger datasets that often go as far back as 1985 for a more comprehensive picture.
I don’t know what your intent is but this doesn’t fill me with any confidence whatsoever.
If you took this seriously, you would wait until after you’ve met regulatory requirements to launch.
This isn't so much a "launch" as a "show". This is my personal library. I've been adding to, maintaining, and using this for a while.
My intent is primarily to gauge interest and share something I've worked hard on.
If you follow the advice and record the trades you would have made, you can get an idea of if the advice is solid, and you can also get a hang for not panic-selling your first dip due to new-investor jitters. It also means that you won't follow crazy trends (like impulse-dumping everything into GME) because it's so obviously an outlier that it won't give you any useful data anyway.
When I was in high school there was a faux-stock-trading website, they gave you ten thousand dollars in fake money and you could fake-buy stocks with it. A great way to play around with, and learn about, the stock market with zero actual risk or investment required.
I ended up putting everything in Diamond Multimedia at $5 and pulling it out the next day at $6, then dumping the rest into eBay (within a month after their IPO IIRC) and forgetting about the website until one day I logged in and my investment was 10x. I patted myself on the back and never invested anything in anything again.
And one "trick" for backtesting is overlay SPX or SPY or DIA (or other favorite) with the results. Maybe it's obvious but folks new to the game are really helped by a comp to the market visual.
But the number hasn't gone up at all, so me and the other 2,877 people were I guess part of a huge surge of signups all within the same 100ms or so, and no one else has signed up since?
This is quite confusing, and does not encourage trust.
In other words, those 2,877 people are others who signed up and didn't complete any further actions, and are all in the same batch for access. Hope this makes more sense.
It can work well if you're filtering for high-quality signal in your own feedback loop. But it can also backfire badly if the intention is to generate hype amplification.
I would guess these numbers are meaningless and only exist to motivate you to click the 'Bump me up in the queue' button and complete a survey.
Like, what is this thing? Is this legal? Is this just more like a compilation of strategies that tell the user what to invest in or is this like a bot like it says it is in which case, the users would need to trust this thing to invest using an algo-trader bot deployed by a venture no one's really heard of.
If this is a project, then great, the concept sounds pretty cool but if this is more like an actual startup, then the path is gnarly, since there are well-trusted platforms already out there including:
- https://www.algotrader.com - https://www.quantconnect.com, etc.
And if you're referring to "no-code" platforms :
- https://www.algogen.io
- https://www.quantreex.com
- https://streak.world
I'm not trying to say your platform is not good per se or anything but if it's meant to be a startup, then it's not doing a particularly great job at it as of now.
It'd be great if you could answer this for me ( if this is a startup and not a project for fun ) :
Are you aware that in order for your platform to be trusted, it doesn't need to be slightly better than the competition but better on a large magnitude? If you know so, then what makes your platform way better than other platforms out there?
Thanks!
I appreciate the feedback (rereading the site I can see the confusion), but I think concluding this is “doing poorly as a startup” is a little premature and something you can’t accurately judge, especially since from your submission history it looks like you have little in the way of a successful startup background when this field is very much a learn by doing one.
This is closer to a roboadvisor than those niche hobby brokerages. The massive difference is those all have incredible barriers to entry - knowing how to code, or at least knowing how to build a profitable automated investment strategy. These are all turnkey, transparent, and no code customizable. Regular investors can use these, only a few can properly take advantage of those you listed. Most of my users came from Fidelity, Schwab, Betterment, and so on. Not Quantconnect (though plenty have come from there and others).
Compared to roboadvisors: these models are more far more innovative and diverse. Use crypto or quantitative strategies made by professionals or curated from other users instead of standard stock bond allocations. Even innovative roboadvisors like m1finance only go as far as hedge fund followers. The difference is huge - the goal here is to be your platform for higher risk and higher return investing, for the 15-20% of your portfolio you want to go further at the risk of larger drawdowns.
I was where you were on my entrepreneurial journey recently, so my best advice is just to do. “10x improvements” and other academics they teach you in SUS is super useful (and often true), but until actually applying the learning, finding what works for you in the current market, you’re just repeating catchphrases without understanding them. You likely won’t even understand the magnitude of an improvement until you do your research and try things out, so best to focus on that when your idea has been validated first.
For your reference: this scrappy site with no optimizations, no contact, no how it works, no fluff, has already garnered several millions of $ in pledged user investment, has profitable and repeatable acquisition channels with an on page organic conversion rate of a large fraction of visitors, has collected thousands of data points, surveys, and interviews towards product market fit and all aspects of the model from target customers to revenue streams, and all this only a few weeks (late December) into prelaunch. I added only what was important to my preexisting library and put it out there. All my research and customer feedback so far suggests that this is a magnitude of improvement, and many are desperate for a platform like this.
Hope that clears it up and thanks again for the feedback!
However, I would like to point a few things out :
- Based on my submission history, it can be deduced that I don't have a successful startup history and that's absolutely true but the questions I asked you were from my understanding of YC SUS and how successful startups usually work, I'm not saying my understanding is 100% correct but it is, to some extent and it has little to do with anyone's startup background. I don't mean this as an excuse in any way but I'm 17 right now and I believe that starting up defensively by throwing that I don't have a successful startup isn't the most respectful thing when you could've just politely answered my questions ( or not ).
- I never claimed that your startup is doing poorly, I merely pointed it's not doing a "great job" *as of now* from what I could tell by evaluating the points from your competition and your offering. But, I wasn't totally sure of that either, hence wanted you to answer those questions for me ( if this was actually an startup, which as it turns out, it is ). And, thanks for answering those questions.
Again, I mean no disrespect to you or your startup. I, as a user of your website, felt that there was stuff left unclear or missing. If you're still gaining massive user growth, honestly, good for you! Although, if I'm your potential user, I think you should work on the feedback I gave, be more lucid and transparent with your product. Otherwise, you should happily discard what I have to say, since that'd legitimately not be the best use of your time.
Also, good luck with your startup, hope it gets even more traction and becomes usable, transparent, and clear enough for a user like me :))
P.S As for the questions about your startup, I think I got my answers. If you were a little less defensive with the initial part, it would've been nice too :)
My main point is that you misapplied those otherwise true principles from YC SUS because you haven't had enough practice with them. Quantconnect and Streak have totally different customers and value proposition - they aren't comparable to Quantbase, and so your entire analysis is incorrect at the start from faulty assumptions (in part perhaps because I need to further clarify the value prop - a marginal improvement).
A 10x improvement is a great rule of thumb because the costs of switching and trust is high - which means that magnitude improvement is relative to the alternatives, and is in the direction of an actual value proposition. People don't use Quantconnect because they have a great How It Works explanation, they use it for the community and trading API. Likewise, missing a How It Works section on my site - while a great suggestion and one I'll eventually add - is likely just a marginal improvement (because it isn't a core value prop) and certainly would be a waste of time for validation (unless my site was too confusing to understand in the first place for most people so my value prop isn't being communicated). Because Quantconnect and Quantbase aren’t alternatives, and How It Works sections aren’t in the direction of a value proposition for either, mentioning the need for a 10x magnitude improvement from a non competitor suggesting a non value prop is just non sensical.
Again, I appreciate your feedback, you just came across as disrespectful by jumping to conclusions on incorrect applications of principles and that's what I took issue with. YC SUS is an awesome resource, but make sure you understand their teachings (by testing them yourself and building your own startups), otherwise you may misapply them.