That's a common pattern in trading strategies with negative skews or tail risks. Even large hedge funds, like LTCM, can fall into this same pitfall.
For anyone interested, I can recommend the book "Systematic Trading" by Robert Carver. You don't have to be into algorithmic trading, the sections on risk management and positive vs negative skews are already worth the read.
Had a period in my college days where I had a neural network running that could successfully trade on patterns of periodicity of non-chaotic windows of the asset. But as soon as the system would go back to being chaotic, and there was no way to identify WHEN the system was chaotic and when it wasn't, the trades would go to shit and I would lose all gains. I was up about 400-450% at an end of a successful cycle, which was 2-4 months, and then it could be a year of decline with gains being eaten up by the option issuers.
Option trading is notoriously difficult. HFTs simply don't take a position at all: they build portfolios that are as neutral as possible in every measure (not only delta, but in vol, its higher order terms, higher order delta terms, etc). This is very difficult to do unless you have substantial capital to absorb short term fluctuations, but if you do you can capture a consistent and very low risk profit.
The best is just buying the whole market via a index fund. Much lower management and overhead costs. Trying to trade without being online all the time is a battle against time delays and costs. To buy stocks, and especially options you are always against the bid/ask spread. The basic math problem for a statistics major is the Gambler's ruin problem. And basically your chances are your money over the whole table's money in total. Even worse if you don't have an exit strategy your chance of going bankrupt is near certain if the odds are even slightly below 50/50. That bid ask spread and incomplete order completion drive your odds down, and for short term trading it is basically a zero sum scenario.
This is what passive traders tell themselves to feel better about letting institutionals have free reign with their money. If there was a best strategy basic economics would tell you that it immediately becomes not the best strategy.
Because it lacks specific criticism, your comment feels a lot like shallow dismissal template that can be deployed against any position one disagrees with, e.g. "This is what $(non-DIYers) tell themselves to feel better about letting $(electricians) have free reign with their money". Substitute with PaaS, banks, healthcare providers as needed.
My comment is shallow because it is not attempting to prove that active investing is better. It's saying, "I've heard people blurt out a list of reasons not to actively manage their portfolio and it's always been from a position of laziness or fear". There is a way to advocate for SPX or VOO or Fidelity 2045 and it's not by saying that these are the optimal strategies.
> The Medallion Fund, which has been available exclusively to current and past employees and their families, surged 80% in 2008 despite hefty fees; the Renaissance Institutional Equities Fund (RIEF), owned by outsiders, lost money in both 2008 and 2009; RIEF declined 16% in 2008
A cynic in me would say that Simons used a better strategy for the Medallion Fund than for the RIEF.
The Medallion Fund does use a better strategy than REIF. It only worked up to a certain scale where they started moving the market though, which is why they ended up kicking out outsiders.
REIF just lets them use Medallion as marketing while getting those sweet AUM fees on a massive fund.
When you look at the market with a zero-sum perspective it becomes apparent that both active and passive investors earn average market returns - collectively they by definition are the market.
However, active investors have higher trading fees/management costs, so they are bound to perform at least slightly worse on average. It's just mathematics.
It is the best strategy for someone who does not do this full-time (and for many people who do!). That is the point.
A trader with experience, working in a firm with deep pockets and terabytes of historical data and FPGAs that can execute his/her strategy and a legal team and an engineering team... can outperform you yes x)
Well I certainly can't argue with something that is italicized.
But I should say that any engineer familiar with the AI tech stack could have bought NVDA at any point in the last five years knowing how big their moat is. That same engineer could have sold monthly covered calls, taking 5 minutes out of every month to do so.
And before you say it, no, they wouldn't be full port NVDA.
The ones that assign, yes. But nvidia pays like one cent per share in quarterly dividends so the real golden eggs are the call premiums. Otherwise you're just trying to buy the goose low and sell it high.
It's trivial to say in hindsight "if you had just bought XYZ in January you would have 80% gains by now!". Yet this is pure confirmation bias. Moreover, surveys routinely show that amateur day traders do not consistently beat a global stock index in a statistically significant manner.
Extrapolating from past data with N=1 to demonstrate that "any tech person can outperform the index" shows a crass lack of mathematical reasoning, which in my opinion demonstrates why you're unsuited to professional trading.
I won't elaborate too much, but consider a situation where the vast majority of of participants just bought a basket of diversified index funds and what the effect would be on the stock market... it would effectively entrench existing companies at their current levels. Regardless of whether Apple or Microsoft or whoever is profitable or not profitable, their relative market cap would basically remain static since people are not buying stocks on the basis of performance but simply buying stocks on the basis of their relative market cap as it existed at some moment in time.
Index funds work because not everyone buys them, they work because the majority of capital is allocated and constantly being rebalanced in a way that tries to reflect the performance of a corporation. It's only in this circumstance that an index fund, or any kind of passive investment, can be of any utility by leveraging the work of those who are actively trying to assess the genuine value of a corporation.
I would highly recommend that you read Mandelbrot's books and Nassim Taleb's books.
For Mandelbrot, I'd recommend starting with The (Mis)Behavior of Markets.
For Taleb, maybe start with The Black Swan.
With those books you'll be able to fully understand why even if/when AI becomes sentient, it will still not help avoid blow ups.
(One reason among many others is, for example, that there are things that cannot be properly forecasted, no matter how hard you try or how advanced your model is, doesn't matter, there's a limit to what we can know and what we can't know about the future...)
Extra: Check out also Thorp's A Man for All Markets.
Other people I've worked with on complicated trading strategies seem to be of the mindset that at best all you can ever hope for is basically breaking even... but my boss is convinced his spaghetti strategies with 500 variables can somehow win more often than anyone else has ever figured out.
>I used to trade options and had about 99.5% success on all my trades
Do you mean this literally or is that an exaggeration for effect?
I'm not sure how you'd do that unless all your trades are like selling a put with a $50 strike price expiring in a month when the stock is trading at $100.
It depends on how they measure profitability, but a 99% win rate isn't uncommon if you just write options.
Covered calls are a good example, you own the stock and sell calls with a strike that is in the money.
You make money on your option as it either expires worthless and you win getting to keep the premium you got paid to write the option or it expires in the money and you sell the shares to the option holder at the strike.
In both cases you can claim you didn't lose money on the option. But you did lose out on potential profit from your share as you were forced to sell them for less than they are worth on the open market and its very easy to get into a case where you would have been better off if you hadn't wrote the option in the first place.
If you manage your own money you can say you didn't lose on the option but you probably did on the underlying stock.
If the stock closes lower you lost money on the stock but make a tiny bit from selling the option.
If the stock closes higher than the strike you wrote you "make" money on the option from the premium but lose out on profit as you have to sell your stock for lower than what you could have in the market.
So again you can say the option didn't lose money but you are worse off than if you hadn't wrote the option, assuming it went up more than the premium you got from writing the option.
So amateur traders can fool themselves into thinking they are geniuses as their option leg doesn't lose money but the overall trade still makes them worse off than if they hadn't of sold the option at all.
I agree with your suggestion of just buying a stock and holding it, but if AI ever becomes sentient, we may well end up being the commodities in their trading exchange
Held stocks provide minimal gains. Perhaps a bot to sell calls. Then your main risk is wiping out potential upside as you cry into your pile of realized gains.
High reward, high risk. I have a day job and don't really want to sweat the details. I have over 140% gain on my long term holdings this year alone--with continued investment and on top of compounding from two decades. I'm quite happy as a Boglehead.
I understand where you’re coming from, but there isn’t a incongruity. Individual stock investments are a relatively small part of my overall portfolio.
> The discussion here assumes that you are not trying to beat the market, but instead passively managing individual stocks to create your own "DIY index fund."
this sounds like you had poor risk management...or didn't have any. I have a strike rate around 55% but all my trades are 3:1 reward:risk and i don't risk more than 1% on any one trade
This doesn't make sense. If you had a 99.5% success rate, you could simply use a fixed $ stop and clean up. I agree you need an information edge, but the difficulty of finding one inversely correlates with your scale. Buying and holding is the best strategy for almost everybody - but that is orthogonal to whether trading is a "coin flip" vs mining an edge.
It does make sense. A small fixed stop loss is just asking to be triggered. A large fixed stop loss will result in a zero-gain zero-loss scenario over many trades.
More generally, there is no optimal amount of stop loss. It ultimately gets auto-stopped out at 3:30 pm, although by then it could have gone nearly to zero anyway. Either the strategy works or it doesn't, and with Trump manipulating the market on random days with significant news, it increasingly doesn't.
OutOfHere gets it. Think about it in terms of probability. The closer to the current price your stop loss is the higher the odds of it being triggered by random market fluctuations.
The same applies to your profit target. This means that if your profit mark is $10, and your stop loss is $5 you will lose roughly twice as often as you win, all other things being equal.
What you actually CAN do is use smart money management, (something like the Kelly Criterion) to ensure that you properly capitalize on any slight edge you do manage to find without going broke in the long term.
That of course requires you to find a bet you can win 51% of the time, and that you be made of iron when it comes to sticking to the plan. Most folks can't.
All of these points are irrelevant. If you have the data that shows 99.5% win rate, you would have the data on your expected win size, and could compute a fixed loss stop that would keep you on-side. In the example, which I'm assuming must be hyperbolic, your average win would only need to stay >1/199th of your average loss. I would agree that if a person cannot manage their positions to this, that person should stick to investing.
The market is adaptive, particularly with options, and will adapt to bust your approach in a heartbeat. It will in general adapt to bust a lot more sophisticated approaches. This is not physics. The underlying assumptions change.
In general, yes - no edge is forever. But that doesn't mean trading is a coin flip where no edge can exist. The market is adaptive because humans are adaptive - but, for example, their machine offspring are often slightly less adaptive and create opportunities for persistent returns. Even more so for most retail where your position size isn't creating slippage.
I can try a dynamic intermediate stop-loss point, with the intended goal that is not too early where it triggers often, and not too late where the loss mounts geometrically. This is intended to greatly limit the max loss per trade, which is good, but it still has significant risk of being triggered unnecessarily. I am highly skeptical that this will work, because I even see in the chart that it doesn't, but it's worth an actual test.
Yes, optimizations of this form can work to manage tail risk on an edge and can reproducibly lift expectancy for a strategy. GP is essentially claiming no positive edge can exist, which is false - though these edges may be short-lived and dependent on particular market regimes.
This is pretty comprehensive. I work at a quant firm, and we don't even have some of this implemented in code. The tricky part is always going to be the integration. Nautilus has its own OMS system, but so does IBKR, and there is no guarantee that they are going to match.
For very small funds, running entirely on IBKR platform (or Alpaca if you can live with their constraints) makes sense. For very large funds, you invariably will have a home-grown system that integrates with all of your expensive vendors. But if you are starting from scratch and want to scale up, using this to bootstrap quickly is most efficient.
I wrote homegrown systems for HFT firms (citadel, RGM, etc) 23 years ago and it’s incredible what they just open sourced but agree with you that it’s the backend that might be the limiting factor. We got to a point where we had to license dark fiber at an alarming rate to stay competitive for our customers too.
Since you work in a quant firm, I am curious if there are any good public resources to learn about the techniques used by such firms. Aimed at a novice I mean. I’ve always had a curiosity about it, but I feel like whatever I can find is more basic than what is truly being done in production.
There are good public resources but they are expensive. One of the best resources for learning in depth quantitative techniques that is actually legit as opposed to Youtubers telling you how to squint to see patterns in a bar chart is the ARPM, which provides both a week long course and a year long course:
It is a very math heavy course, highly technical and requires a great deal of existing math background. But the truth is that you will not be able to do much quantitative finance without that background anyways, so if what you want is a way to gain that math background then just crack open a textbook and get working.
I work at a fund, and you need millions in capital and infra. It really doesn't make sense to trade as a retail investor. Look up Dimitri Bianco on YouTube; he's quite reputable and explains why it's not really feasible to trade with your own money.
You need a good background in mathematics and some knowledge of why you should be paid a premium by the market (e.g. carry, trend, various premia and so on)
Having wasted 6 years of my life intensively working to create an algorithmic trading system (and failing to make it consistently profitable), what they have here is the easy part. You need a system for discovering strategies. That's were almost all your effort will go. The simulator for backtesting, integrating with a broker, etc. is such a small part of it, if you're serious about it, you're probably better off writing your own.
A long time ago I figured out the difference between a Trader and Investor and realized I am very much the latter. After a few years this realization lead to my wealth increasing.
100%, other than selling RSUs to diversify, every single other investment I have is now buy and hold.
Even did a ~7 year career detour through quant finance "if I like Software Engineering and Mathematics so much, why don't I combine the two?"
Finally realized that the best use of my time was to just to work hard at a career I deeply enjoy (Software Engineering), working on products I actually care about (not valuing arcane derivatives products), and just invest the excess in diversified index funds (with some single stock selections here and there, thanks TSLA and NVDA).
Just as engineers can get "nerd sniped", I feel like "trading" is a somewhat malevolent strong attractor for a lot of folks. Folks do need general financial literacy, but an extra hour spent per day working harder to progress day job / long term career likely has a higher net present value than trading options or crypto.
An Investor is essentially a Trader operating on a very long timeframe, benefiting from a slow decision-making process, slow results, and slow emotional swings.
And I know people will argue against this. But if you are a trader, identify a sweet spot for speed/frequency, and then your trading results will suddenly improve.
I just have to say your post starts out just like those bots do on youtube finance video comments, esp with the caps. I sorta felt like replying in their style to you really
Yea incredibly hard to do so consistently. Those that do (the hedge funds) have the security of the ~2% fixed fee they earn to weather the periods when the returns are not there aka "others peoples money".
You can if you have an enormous pile of cash. Then you can pat yourself on the back for making a few $k per month (which would only be a couple % apy) for acting like a quant when you could make the same amount in bonds.
Language evolves over time but "Trader" (and to a large extent "Investor") doesn't mean what they imply it means in ads for crypto / leveraged brokerage accounts and even pension accounts. As for grifters on social media / youtube / paid for newsletters - most of the words they say or write dont mean the things they think they mean.
Algorithmic trading is a deep rabbit hole that will drive you mad the more you try to understand it. There are just too many variables to account for and I genuinely don't understand how you could make a stable trading system that reliably makes money as a retail trader.
Excluding HFT (which is reserved for people with hundreds of millions to invest in infra, fresh Ivy league quant analysts, and a fiber optic cable hooked up directly to the exchange; they likely already have an in-house tool that does what this project does), you're really just left with intraday trading or long term investing. Investing doesn't require algorithmic trading or back testing, so it seems that this projects demographic is aimed toward intraday retail traders.
With intraday trading, your chances of making a successful trading algo are near 0%. I mean, think about it: you have to account for every single variable in the stock market. How are you supposed to account for a truth social post imposing or lifting tariffs? Or a ransomware attack crippling a company? Or if a whale decides to sell all their $BIGCORP shares on the flip of a coin? It's impossible. Your only odds of success with intraday trading is manually doing it. You yourself are an "algo" trader that is capable of changing their strategy on the fly and accounting for unknown variables. A pre-programmed algo can not, no matter how much context you give it.
Furthermore, with back testing, it's impossible to accurately capture the context of the market during that time. Let's say you back test on 180 days of data. Well, do you know exactly what happened on the 71st day of that data? Did you account for that fed meeting, that tariff hike, etc? What about all the other days? Testing on OHLCV alone is not enough; you need the entire context of the market.
While the project itself it neat, I just don't see how algorithmic trading could lead to any long term success.
Interesting choice to be using CPython.
Very impressed with the completeness, and in particular the Risk Engine. Not having delved in too deeply yet, it does have the basics, but the markets are very highly regulated, and automated trading is always in the microscope.
At Goldman, my team's whole job was to make sure the automated trading desks have all the checks implemented. It's a little spoilsport, but anyone seriously looking to use this, it's not gonna be very plug and play, at least while comlying with the market regulations.
I know FINRA has regulations but the CFTC proposed regulations were shelved. Are little guys who are not FINRA members encumbered by regulations here too?
Particularly for India, for example, all algo's have to be pre-approved, and the core principle is that you cannot operate in isolation; your broker is your gateway and your primary point of compliance.
For the US, the SEC and FINRA place the onus on the broker-dealer to supervise all trading activity, including automated trading by retail clients. This includes having risk controls to prevent erroneous or manipulative orders. An individual is not typically required to register their specific algorithm with the SEC, but they are subject to all overarching rules against market manipulation and fraud.
So, end of day, you'll still have to do a ton of back and forth with your broker, to make sure their controls cover your implementation.
Not really feasible for the retail trader from the looks of it, regardless of location. But some smaller firms can perhaps utilise it.
Interesting stuff. I'm a bit confused on several points though.
If your strategies care so much about performance, can you really achieve the same or comparable performance with Cython? At that point, wouldn't those strategies be better suited for a custom "re-implementation"?
I suppose I'm not exactly sure how "high performance" and "low latency" this project is. Are there any latency stats on this project?
I am guessing it is "low latency" as in not taking 2 minutes to calculate an order size and submit it rather than "low latency" for HFT and 1,000 trades a second.
I read it does exists trading platforms were if you earn money they lose it because they launched the order internally between their affiliates, and not to the external trading networks (or whatever they are called). This if they even really launch the orders...
We can not see the traders' internal code so anything can happen. I think they look like money-collecting machines. Take care guys.
Edit: Wow, that downvote sounds very interesting, as the first that I pointed is widely known, the so called "Market Makers".
Trading platforms would make more money if you made just enough money to continue to trade long term. Especially the bucket shops crypto and forex exchanges that make most of their money front-running customer orders.
Unregulated exchanges are just a way to bring back the bad old days of "boiler room" penny stock scammers of the 80's. Most of the people making money are running 40 year old scams on 20 somethings who don't know better yet.
I spent almost 6 years trading crypto. Our best month's volume was $6B.
1. Nothing we tried with usual strategies worked consistently. Backtesting parameters, ML with smart feature selection, boosting, neural networks - everything failed out of sample. Maybe we were dumb, I don't know.
2. What worked was having a clear edge:
- Private exchange programs with rebates for high-volume teams
- Pure latency arbitrage
- Weird arbitrage trading obscure instruments (e.g. on chain AMMs vs crypto exchange futures).
Both market maker and arbitrage strategies were very sensitive to latency. We built a low-latency trading engine in Java (on top of https://github.com/OpenHFT/Chronicle-Queue). We got 130mqs from market event to order send in a hot loop on 99.99 percentile. It was fun to optimize and benchmark.
- Tail latency matters. You can have 100ms at the 90th percentile and 10-50ms at the 99.9th percentile. For low latency strategies, this is pure loss.
- Tail latency matters even more when markets go crazy. Event rates can jump from 10 per second to 1-2k per second. If your trading engine uses trades or bid/ask events, be ready. For OHLC bars it doesn't matter, but nothing based on OHLC worked for us.
---
p.s.
I wouldn't recommend trading to anyone. It's very stressful and exhausting. More importantly, all your hard work disappears like trying to hold sand in your hands. There's very little compounding of your work. Strategies constantly churn. You're always negotiating with managers for fees and API limits. You're always negotiating with managers for fees and API limits. They force you to buy and hold exchange tokens like Binance's BNB just to get slightly better fees, VIP status, or direct API access that bypasses firewalls.
The industry is extremely secretive - it's a zero-sum game with no incentive to share anything. When you meet someone who trades, it's usually hard to have a meaningful conversation. At least that was true for me. All our strategies were small-scale and we couldn't invest all our capital in them. So discussing what we did was basically saying "yes, we print money, here's how you can take it from us".
Overall, I was super lucky. We built a money-printing machine that worked for a few years. But in the end, my co-founder and I decided not to pursue it long-term. One day when yet another strategy stopped working, we just shut down all operations.
I understand that most, if not all, are in the business of making money. A platform that is sorely missing is an interactive, teaching platform for trading which has suggestions on what to trade, how to trade it, why to trade it and when. A platform that teaches the art of trading without risking any capital, by providing trading simulations, using real-time pricing of as many asset classes as possible. I think people would pay a reasonable price to learn, and gain confidence. I tried to trade on my own and ended up losing, in the end.
I've tried to write automated trading systems over the years, but they always perform worse than the stocks I pick to buy and hold and trade at most a few times a year. People do make money at this, so it's not impossible, but it's a bit of a white whale.
Wondering how average users can benefit from this platform with Claude Code and the relation to Vending Bench that tracks how much money LLM's can make.
I've been wading into algo-trading for a little while now. I've read a few books, set up a few strategies on paper trading platforms.
Right now I'm trying to figure out how to consistently make $1/day as a POC exercise. That's it. I did ask the various advanced LLMs the path to earning/clearing $200K/year as an algo-trader.
Fun & sobering responses. I'll give the LLMs this -- no uplift in these conversations.
Consistently making $1 a day may be difficult for retail trading. Rather, you could attempt to be right 51% of the time gathering a 1% edge. This is not easy for us because, it randomizes profits and losses, and we like to not lose money (make a $1 a day), and this could expose you to streaks of losses.
My operating presumption is that there's runoff underneath market makers for small (on this scale) scalping. I also hope that those who need to return to investors or generate millions/year aren't trading the way I'd be trading to generate $200K/year.
I wouldn't count on it, especially for crypto. The market is brutal. I was playing around the other day with liquidity pools and some bot executed an arbitrage for a total gain of $0.1. That's how efficient the market is.
> Right now I'm trying to figure out how to consistently make $1/day as a POC exercise.
You can do this with the following strategy:
1. Buy $10,000 of Treasury bonds.
It might be more interesting to target a particular Sharpe ratio rather than an absolute dollar amount since that will be proportional to the amount you invest.
Surely you can't? You don't have the latency advantages of big trading firms, nor the gazillion PhDs they employ. There's absolutely no way you are going to consistently make money, let alone $200k/year. At best you'll get lucky and think it's skill.
It's possible: note that each one of those quants is going to be making $400k at minimum, plausibly significantly more. So any high-maintenance strategy (eg. in some niche market) that makes less than that is just not going to be worth the time it takes to find, implement, maintain, etc.
>> $200k is totally doable with $2M investment capital and a bit of disciplined options trading.
10% probably. Though with the higher-than-market rate, there would also be dispersion in the consistency of the return. I could see some lean years where no money is made. I could also see a decade where you are buying at ever-higher valuations rather than dollar-cost averaging -- thus you could see a lean decade following.
I used to trade options and had about 99.5% success on all my trades
The problem is the 0.5% of the time, it erases all the gains made on the successful ones.
I'm convinced without information edge or some capital sunk cost edge (for HFT) you are literally just flipping coins when it comes to trading.
What's dangerous is fixation on strategies that form after a period of success.
All in all, I think just buying stock and holding is the best and most successful approach to making money.
Maybe when AI becomes sentient it will know on which days it make sense to buy and sell iron condors with huge ass wings....
That's a common pattern in trading strategies with negative skews or tail risks. Even large hedge funds, like LTCM, can fall into this same pitfall.
For anyone interested, I can recommend the book "Systematic Trading" by Robert Carver. You don't have to be into algorithmic trading, the sections on risk management and positive vs negative skews are already worth the read.
Have exactly the same experience as you.
Had a period in my college days where I had a neural network running that could successfully trade on patterns of periodicity of non-chaotic windows of the asset. But as soon as the system would go back to being chaotic, and there was no way to identify WHEN the system was chaotic and when it wasn't, the trades would go to shit and I would lose all gains. I was up about 400-450% at an end of a successful cycle, which was 2-4 months, and then it could be a year of decline with gains being eaten up by the option issuers.
Now I only do long-term funds/stocks and have:
a) much less anxiety about losses b) more money.
Is this the financial version of the 3-body problem?
A deep cut
Option trading is notoriously difficult. HFTs simply don't take a position at all: they build portfolios that are as neutral as possible in every measure (not only delta, but in vol, its higher order terms, higher order delta terms, etc). This is very difficult to do unless you have substantial capital to absorb short term fluctuations, but if you do you can capture a consistent and very low risk profit.
HFTs are typically market makers so they don't really build a portfolio as per se anyway (in practice you can target a certain position obviously)
The best is just buying the whole market via a index fund. Much lower management and overhead costs. Trying to trade without being online all the time is a battle against time delays and costs. To buy stocks, and especially options you are always against the bid/ask spread. The basic math problem for a statistics major is the Gambler's ruin problem. And basically your chances are your money over the whole table's money in total. Even worse if you don't have an exit strategy your chance of going bankrupt is near certain if the odds are even slightly below 50/50. That bid ask spread and incomplete order completion drive your odds down, and for short term trading it is basically a zero sum scenario.
This is what passive traders tell themselves to feel better about letting institutionals have free reign with their money. If there was a best strategy basic economics would tell you that it immediately becomes not the best strategy.
Because it lacks specific criticism, your comment feels a lot like shallow dismissal template that can be deployed against any position one disagrees with, e.g. "This is what $(non-DIYers) tell themselves to feel better about letting $(electricians) have free reign with their money". Substitute with PaaS, banks, healthcare providers as needed.
My comment is shallow because it is not attempting to prove that active investing is better. It's saying, "I've heard people blurt out a list of reasons not to actively manage their portfolio and it's always been from a position of laziness or fear". There is a way to advocate for SPX or VOO or Fidelity 2045 and it's not by saying that these are the optimal strategies.
As a counter-argument, from [0]:
> The Medallion Fund, which has been available exclusively to current and past employees and their families, surged 80% in 2008 despite hefty fees; the Renaissance Institutional Equities Fund (RIEF), owned by outsiders, lost money in both 2008 and 2009; RIEF declined 16% in 2008
A cynic in me would say that Simons used a better strategy for the Medallion Fund than for the RIEF.
[0] https://en.wikipedia.org/wiki/Jim_Simons#Controversies
The Medallion Fund does use a better strategy than REIF. It only worked up to a certain scale where they started moving the market though, which is why they ended up kicking out outsiders.
REIF just lets them use Medallion as marketing while getting those sweet AUM fees on a massive fund.
When you look at the market with a zero-sum perspective it becomes apparent that both active and passive investors earn average market returns - collectively they by definition are the market.
However, active investors have higher trading fees/management costs, so they are bound to perform at least slightly worse on average. It's just mathematics.
I am completely at a loss for words here.
It is the best strategy for someone who does not do this full-time (and for many people who do!). That is the point.
A trader with experience, working in a firm with deep pockets and terabytes of historical data and FPGAs that can execute his/her strategy and a legal team and an engineering team... can outperform you yes x)
Well I certainly can't argue with something that is italicized.
But I should say that any engineer familiar with the AI tech stack could have bought NVDA at any point in the last five years knowing how big their moat is. That same engineer could have sold monthly covered calls, taking 5 minutes out of every month to do so.
And before you say it, no, they wouldn't be full port NVDA.
Wouldn't selling covered calls on Nvidia have been like selling a goose that lays golden eggs?
The ones that assign, yes. But nvidia pays like one cent per share in quarterly dividends so the real golden eggs are the call premiums. Otherwise you're just trying to buy the goose low and sell it high.
Jesus, don't quit your day job to become a trader is all I'm gonna say.
Can you at least provide specificity to match your derision?
It's trivial to say in hindsight "if you had just bought XYZ in January you would have 80% gains by now!". Yet this is pure confirmation bias. Moreover, surveys routinely show that amateur day traders do not consistently beat a global stock index in a statistically significant manner.
Extrapolating from past data with N=1 to demonstrate that "any tech person can outperform the index" shows a crass lack of mathematical reasoning, which in my opinion demonstrates why you're unsuited to professional trading.
Okay since you're just tossing insults I'll just say that you don't need hindsight to have known that Nvidia is selling shovels in an AI gold rush.
Can you elaborate?
I won't elaborate too much, but consider a situation where the vast majority of of participants just bought a basket of diversified index funds and what the effect would be on the stock market... it would effectively entrench existing companies at their current levels. Regardless of whether Apple or Microsoft or whoever is profitable or not profitable, their relative market cap would basically remain static since people are not buying stocks on the basis of performance but simply buying stocks on the basis of their relative market cap as it existed at some moment in time.
Index funds work because not everyone buys them, they work because the majority of capital is allocated and constantly being rebalanced in a way that tries to reflect the performance of a corporation. It's only in this circumstance that an index fund, or any kind of passive investment, can be of any utility by leveraging the work of those who are actively trying to assess the genuine value of a corporation.
Thank you! I think it makes sense
I would highly recommend that you read Mandelbrot's books and Nassim Taleb's books.
For Mandelbrot, I'd recommend starting with The (Mis)Behavior of Markets. For Taleb, maybe start with The Black Swan.
With those books you'll be able to fully understand why even if/when AI becomes sentient, it will still not help avoid blow ups.
(One reason among many others is, for example, that there are things that cannot be properly forecasted, no matter how hard you try or how advanced your model is, doesn't matter, there's a limit to what we can know and what we can't know about the future...)
Extra: Check out also Thorp's A Man for All Markets.
Fundamentally yes you are. Making money comes down to managing your risk. Cutting your losses quickly and riding your winners.
Other people I've worked with on complicated trading strategies seem to be of the mindset that at best all you can ever hope for is basically breaking even... but my boss is convinced his spaghetti strategies with 500 variables can somehow win more often than anyone else has ever figured out.
>I used to trade options and had about 99.5% success on all my trades
Do you mean this literally or is that an exaggeration for effect?
I'm not sure how you'd do that unless all your trades are like selling a put with a $50 strike price expiring in a month when the stock is trading at $100.
It depends on how they measure profitability, but a 99% win rate isn't uncommon if you just write options.
Covered calls are a good example, you own the stock and sell calls with a strike that is in the money.
You make money on your option as it either expires worthless and you win getting to keep the premium you got paid to write the option or it expires in the money and you sell the shares to the option holder at the strike.
In both cases you can claim you didn't lose money on the option. But you did lose out on potential profit from your share as you were forced to sell them for less than they are worth on the open market and its very easy to get into a case where you would have been better off if you hadn't wrote the option in the first place.
If you manage your own money you can say you didn't lose on the option but you probably did on the underlying stock.
If the stock closes lower you lost money on the stock but make a tiny bit from selling the option.
If the stock closes higher than the strike you wrote you "make" money on the option from the premium but lose out on profit as you have to sell your stock for lower than what you could have in the market.
So again you can say the option didn't lose money but you are worse off than if you hadn't wrote the option, assuming it went up more than the premium you got from writing the option.
So amateur traders can fool themselves into thinking they are geniuses as their option leg doesn't lose money but the overall trade still makes them worse off than if they hadn't of sold the option at all.
> Do you mean this literally or is that an exaggeration for effect?
As a fellow options trader, I can confirm that it is spot on what is seen with SPY; it is not an exaggeration at all.
assuming this is something like covered calls?
That isn't it for me, but that's not to say it couldn't be for others.
> 99.5% success on all my trades
Winning trades made $0.01, losing trades lost $10k.
Pennies in front of a steamroller
> I used to trade options and had about 99.5% success on all my trades.
> The problem is the 0.5% of the time, it erases all the gains made on the successful ones.
Nassim Taleb calls this "eating like chicken and shitting like elephants"
I agree with your suggestion of just buying a stock and holding it, but if AI ever becomes sentient, we may well end up being the commodities in their trading exchange
Held stocks provide minimal gains. Perhaps a bot to sell calls. Then your main risk is wiping out potential upside as you cry into your pile of realized gains.
High reward, high risk. I have a day job and don't really want to sweat the details. I have over 140% gain on my long term holdings this year alone--with continued investment and on top of compounding from two decades. I'm quite happy as a Boglehead.
> Boglehead
> 140% gain on your holdings this year
Choose one.
Generally true but nvda and pltr are normie stocks and can account for these returns from this year.
Boggle head is basically pick 2-3 vanguard etfs and check back in 25 years.
But then it's not a Boglehead lol
https://www.bogleheads.org/wiki/Passively_managing_individua...
I understand where you’re coming from, but there isn’t a incongruity. Individual stock investments are a relatively small part of my overall portfolio.
> The discussion here assumes that you are not trying to beat the market, but instead passively managing individual stocks to create your own "DIY index fund."
> minimal gains
Vanguard says I've had a 12% rate of return. I guess I could have done active trading if I'd gone on fewer dates
Realized return or portfolio growth?
This is 100% accurate.
The only way to make money for 99.999% of traders is to trade with a firm. You should not trade on your own. You will lose in the long term.
this sounds like you had poor risk management...or didn't have any. I have a strike rate around 55% but all my trades are 3:1 reward:risk and i don't risk more than 1% on any one trade
This doesn't make sense. If you had a 99.5% success rate, you could simply use a fixed $ stop and clean up. I agree you need an information edge, but the difficulty of finding one inversely correlates with your scale. Buying and holding is the best strategy for almost everybody - but that is orthogonal to whether trading is a "coin flip" vs mining an edge.
It does make sense. A small fixed stop loss is just asking to be triggered. A large fixed stop loss will result in a zero-gain zero-loss scenario over many trades.
More generally, there is no optimal amount of stop loss. It ultimately gets auto-stopped out at 3:30 pm, although by then it could have gone nearly to zero anyway. Either the strategy works or it doesn't, and with Trump manipulating the market on random days with significant news, it increasingly doesn't.
OutOfHere gets it. Think about it in terms of probability. The closer to the current price your stop loss is the higher the odds of it being triggered by random market fluctuations.
The same applies to your profit target. This means that if your profit mark is $10, and your stop loss is $5 you will lose roughly twice as often as you win, all other things being equal.
What you actually CAN do is use smart money management, (something like the Kelly Criterion) to ensure that you properly capitalize on any slight edge you do manage to find without going broke in the long term.
That of course requires you to find a bet you can win 51% of the time, and that you be made of iron when it comes to sticking to the plan. Most folks can't.
All of these points are irrelevant. If you have the data that shows 99.5% win rate, you would have the data on your expected win size, and could compute a fixed loss stop that would keep you on-side. In the example, which I'm assuming must be hyperbolic, your average win would only need to stay >1/199th of your average loss. I would agree that if a person cannot manage their positions to this, that person should stick to investing.
The market is adaptive, particularly with options, and will adapt to bust your approach in a heartbeat. It will in general adapt to bust a lot more sophisticated approaches. This is not physics. The underlying assumptions change.
In general, yes - no edge is forever. But that doesn't mean trading is a coin flip where no edge can exist. The market is adaptive because humans are adaptive - but, for example, their machine offspring are often slightly less adaptive and create opportunities for persistent returns. Even more so for most retail where your position size isn't creating slippage.
I can try a dynamic intermediate stop-loss point, with the intended goal that is not too early where it triggers often, and not too late where the loss mounts geometrically. This is intended to greatly limit the max loss per trade, which is good, but it still has significant risk of being triggered unnecessarily. I am highly skeptical that this will work, because I even see in the chart that it doesn't, but it's worth an actual test.
Yes, optimizations of this form can work to manage tail risk on an edge and can reproducibly lift expectancy for a strategy. GP is essentially claiming no positive edge can exist, which is false - though these edges may be short-lived and dependent on particular market regimes.
With regard to anyone who is skeptical of the assertion of Trump manipulating the market, and this has cost me thousands of dollars, I refer you to https://www.zerohedge.com/markets/every-fkin-day-bro-glimpse...
Try IWM brother.
> I'm convinced without information edge or some capital sunk cost edge (for HFT) you are literally just flipping coins when it comes to trading.
This applies to everything in life.
This is pretty comprehensive. I work at a quant firm, and we don't even have some of this implemented in code. The tricky part is always going to be the integration. Nautilus has its own OMS system, but so does IBKR, and there is no guarantee that they are going to match.
For very small funds, running entirely on IBKR platform (or Alpaca if you can live with their constraints) makes sense. For very large funds, you invariably will have a home-grown system that integrates with all of your expensive vendors. But if you are starting from scratch and want to scale up, using this to bootstrap quickly is most efficient.
I wrote homegrown systems for HFT firms (citadel, RGM, etc) 23 years ago and it’s incredible what they just open sourced but agree with you that it’s the backend that might be the limiting factor. We got to a point where we had to license dark fiber at an alarming rate to stay competitive for our customers too.
Since you work in a quant firm, I am curious if there are any good public resources to learn about the techniques used by such firms. Aimed at a novice I mean. I’ve always had a curiosity about it, but I feel like whatever I can find is more basic than what is truly being done in production.
There are good public resources but they are expensive. One of the best resources for learning in depth quantitative techniques that is actually legit as opposed to Youtubers telling you how to squint to see patterns in a bar chart is the ARPM, which provides both a week long course and a year long course:
https://www.arpm.co/
It is a very math heavy course, highly technical and requires a great deal of existing math background. But the truth is that you will not be able to do much quantitative finance without that background anyways, so if what you want is a way to gain that math background then just crack open a textbook and get working.
I work at a fund, and you need millions in capital and infra. It really doesn't make sense to trade as a retail investor. Look up Dimitri Bianco on YouTube; he's quite reputable and explains why it's not really feasible to trade with your own money.
In my opinion: If you trade, trade with a firm.
You need a good background in mathematics and some knowledge of why you should be paid a premium by the market (e.g. carry, trend, various premia and so on)
Anything else requires millions of infra spend.
I've heard ibkr will ban you if they can detect algorithmic trading.
The fact that they have an api says otherwise:)
Lots of small hedge fund use them as a stop gap until they get bigger, or fail.
You heard wrong.
Having wasted 6 years of my life intensively working to create an algorithmic trading system (and failing to make it consistently profitable), what they have here is the easy part. You need a system for discovering strategies. That's were almost all your effort will go. The simulator for backtesting, integrating with a broker, etc. is such a small part of it, if you're serious about it, you're probably better off writing your own.
This is the right comment.
This is also why people in algo trading that are able to discover or already know working strategies, they are paid big bucks.
A long time ago I figured out the difference between a Trader and Investor and realized I am very much the latter. After a few years this realization lead to my wealth increasing.
100%, other than selling RSUs to diversify, every single other investment I have is now buy and hold.
Even did a ~7 year career detour through quant finance "if I like Software Engineering and Mathematics so much, why don't I combine the two?"
Finally realized that the best use of my time was to just to work hard at a career I deeply enjoy (Software Engineering), working on products I actually care about (not valuing arcane derivatives products), and just invest the excess in diversified index funds (with some single stock selections here and there, thanks TSLA and NVDA).
Just as engineers can get "nerd sniped", I feel like "trading" is a somewhat malevolent strong attractor for a lot of folks. Folks do need general financial literacy, but an extra hour spent per day working harder to progress day job / long term career likely has a higher net present value than trading options or crypto.
An Investor is essentially a Trader operating on a very long timeframe, benefiting from a slow decision-making process, slow results, and slow emotional swings. And I know people will argue against this. But if you are a trader, identify a sweet spot for speed/frequency, and then your trading results will suddenly improve.
I just have to say your post starts out just like those bots do on youtube finance video comments, esp with the caps. I sorta felt like replying in their style to you really
Yep, it’s amazing how many people seem to get sucked in by the idea you can reliably make money as a solo/small trading operation.
Yea incredibly hard to do so consistently. Those that do (the hedge funds) have the security of the ~2% fixed fee they earn to weather the periods when the returns are not there aka "others peoples money".
>> Yep, it’s amazing how many people seem to get sucked in by the idea you can reliably make money as a solo/small trading operation.
You can right!? You can make beta. Sometimes the beta is negative. Hard to make alpha.
Everyone that bought and held on the up movement this week is a legendary "Trader".
You can if you have an enormous pile of cash. Then you can pat yourself on the back for making a few $k per month (which would only be a couple % apy) for acting like a quant when you could make the same amount in bonds.
An investor is a trader. I think what you meant is to stay away from "day trading". Your whole life, careers, etc. are a trade.
Language evolves over time but "Trader" (and to a large extent "Investor") doesn't mean what they imply it means in ads for crypto / leveraged brokerage accounts and even pension accounts. As for grifters on social media / youtube / paid for newsletters - most of the words they say or write dont mean the things they think they mean.
[flagged]
Algorithmic trading is a deep rabbit hole that will drive you mad the more you try to understand it. There are just too many variables to account for and I genuinely don't understand how you could make a stable trading system that reliably makes money as a retail trader.
Excluding HFT (which is reserved for people with hundreds of millions to invest in infra, fresh Ivy league quant analysts, and a fiber optic cable hooked up directly to the exchange; they likely already have an in-house tool that does what this project does), you're really just left with intraday trading or long term investing. Investing doesn't require algorithmic trading or back testing, so it seems that this projects demographic is aimed toward intraday retail traders.
With intraday trading, your chances of making a successful trading algo are near 0%. I mean, think about it: you have to account for every single variable in the stock market. How are you supposed to account for a truth social post imposing or lifting tariffs? Or a ransomware attack crippling a company? Or if a whale decides to sell all their $BIGCORP shares on the flip of a coin? It's impossible. Your only odds of success with intraday trading is manually doing it. You yourself are an "algo" trader that is capable of changing their strategy on the fly and accounting for unknown variables. A pre-programmed algo can not, no matter how much context you give it.
Furthermore, with back testing, it's impossible to accurately capture the context of the market during that time. Let's say you back test on 180 days of data. Well, do you know exactly what happened on the 71st day of that data? Did you account for that fed meeting, that tariff hike, etc? What about all the other days? Testing on OHLCV alone is not enough; you need the entire context of the market.
While the project itself it neat, I just don't see how algorithmic trading could lead to any long term success.
Interesting choice to be using CPython. Very impressed with the completeness, and in particular the Risk Engine. Not having delved in too deeply yet, it does have the basics, but the markets are very highly regulated, and automated trading is always in the microscope. At Goldman, my team's whole job was to make sure the automated trading desks have all the checks implemented. It's a little spoilsport, but anyone seriously looking to use this, it's not gonna be very plug and play, at least while comlying with the market regulations.
I know FINRA has regulations but the CFTC proposed regulations were shelved. Are little guys who are not FINRA members encumbered by regulations here too?
Particularly for India, for example, all algo's have to be pre-approved, and the core principle is that you cannot operate in isolation; your broker is your gateway and your primary point of compliance.
For the US, the SEC and FINRA place the onus on the broker-dealer to supervise all trading activity, including automated trading by retail clients. This includes having risk controls to prevent erroneous or manipulative orders. An individual is not typically required to register their specific algorithm with the SEC, but they are subject to all overarching rules against market manipulation and fraud.
So, end of day, you'll still have to do a ton of back and forth with your broker, to make sure their controls cover your implementation. Not really feasible for the retail trader from the looks of it, regardless of location. But some smaller firms can perhaps utilise it.
Finally, a way for me to lose all of my money automatically. What an exciting time to be alive!
More seriously, for anyone else who was curious below is a list of the existing integrations.
https://nautilustrader.io/docs/latest/integrations/
Interesting not a single one of them is in the realm of traditional trading.
Can you actually buy real stocks / etfs / mutual funds with this platform?
Interactive brokers is real
Generally you can connect your OMS to a FIX connection or similar directly to a broker.
How does this compare with the LEAN engine[1] and QuantConnect[2]?
[1]: https://www.lean.io/
[2]: https://www.quantconnect.com/
There are a few dozen of these out there. That’s the first question anyone should answer.
Interesting stuff. I'm a bit confused on several points though.
If your strategies care so much about performance, can you really achieve the same or comparable performance with Cython? At that point, wouldn't those strategies be better suited for a custom "re-implementation"?
I suppose I'm not exactly sure how "high performance" and "low latency" this project is. Are there any latency stats on this project?
I am guessing it is "low latency" as in not taking 2 minutes to calculate an order size and submit it rather than "low latency" for HFT and 1,000 trades a second.
I wouldn't be surprised if this is sponsored by trading platform in order for customers to lose all money in record time...
I think the trading platforms earn more money if you make more trades. So it's in their interest that you trade for as long as possible.
I read it does exists trading platforms were if you earn money they lose it because they launched the order internally between their affiliates, and not to the external trading networks (or whatever they are called). This if they even really launch the orders...
We can not see the traders' internal code so anything can happen. I think they look like money-collecting machines. Take care guys.
Edit: Wow, that downvote sounds very interesting, as the first that I pointed is widely known, the so called "Market Makers".
The website does say they're partners with OKX, nothing better than bringing more retail customers (that will promptly lose their money).
I've seen stats from other crypto exchanges that are something like "90% of the retail customers lose 90% of their funds in 90 days or less".
Good news for the professional market makers, not for the ~marks~ retail customers.
Wait till you realize Okx, Binance, ByBit, etc. are all the same entity.
AFAIK they're distinct entities.
Trading platforms would make more money if you made just enough money to continue to trade long term. Especially the bucket shops crypto and forex exchanges that make most of their money front-running customer orders.
Unregulated exchanges are just a way to bring back the bad old days of "boiler room" penny stock scammers of the 80's. Most of the people making money are running 40 year old scams on 20 somethings who don't know better yet.
I spent almost 6 years trading crypto. Our best month's volume was $6B.
1. Nothing we tried with usual strategies worked consistently. Backtesting parameters, ML with smart feature selection, boosting, neural networks - everything failed out of sample. Maybe we were dumb, I don't know.
2. What worked was having a clear edge: - Private exchange programs with rebates for high-volume teams - Pure latency arbitrage - Weird arbitrage trading obscure instruments (e.g. on chain AMMs vs crypto exchange futures).
Both market maker and arbitrage strategies were very sensitive to latency. We built a low-latency trading engine in Java (on top of https://github.com/OpenHFT/Chronicle-Queue). We got 130mqs from market event to order send in a hot loop on 99.99 percentile. It was fun to optimize and benchmark.
- Tail latency matters. You can have 100ms at the 90th percentile and 10-50ms at the 99.9th percentile. For low latency strategies, this is pure loss. - Tail latency matters even more when markets go crazy. Event rates can jump from 10 per second to 1-2k per second. If your trading engine uses trades or bid/ask events, be ready. For OHLC bars it doesn't matter, but nothing based on OHLC worked for us.
---
p.s.
I wouldn't recommend trading to anyone. It's very stressful and exhausting. More importantly, all your hard work disappears like trying to hold sand in your hands. There's very little compounding of your work. Strategies constantly churn. You're always negotiating with managers for fees and API limits. You're always negotiating with managers for fees and API limits. They force you to buy and hold exchange tokens like Binance's BNB just to get slightly better fees, VIP status, or direct API access that bypasses firewalls.
The industry is extremely secretive - it's a zero-sum game with no incentive to share anything. When you meet someone who trades, it's usually hard to have a meaningful conversation. At least that was true for me. All our strategies were small-scale and we couldn't invest all our capital in them. So discussing what we did was basically saying "yes, we print money, here's how you can take it from us".
Overall, I was super lucky. We built a money-printing machine that worked for a few years. But in the end, my co-founder and I decided not to pursue it long-term. One day when yet another strategy stopped working, we just shut down all operations.
I understand that most, if not all, are in the business of making money. A platform that is sorely missing is an interactive, teaching platform for trading which has suggestions on what to trade, how to trade it, why to trade it and when. A platform that teaches the art of trading without risking any capital, by providing trading simulations, using real-time pricing of as many asset classes as possible. I think people would pay a reasonable price to learn, and gain confidence. I tried to trade on my own and ended up losing, in the end.
I've tried to write automated trading systems over the years, but they always perform worse than the stocks I pick to buy and hold and trade at most a few times a year. People do make money at this, so it's not impossible, but it's a bit of a white whale.
Wondering how average users can benefit from this platform with Claude Code and the relation to Vending Bench that tracks how much money LLM's can make.
https://andonlabs.com/evals/vending-bench
I've been wading into algo-trading for a little while now. I've read a few books, set up a few strategies on paper trading platforms.
Right now I'm trying to figure out how to consistently make $1/day as a POC exercise. That's it. I did ask the various advanced LLMs the path to earning/clearing $200K/year as an algo-trader.
Fun & sobering responses. I'll give the LLMs this -- no uplift in these conversations.
Consistently making $1 a day may be difficult for retail trading. Rather, you could attempt to be right 51% of the time gathering a 1% edge. This is not easy for us because, it randomizes profits and losses, and we like to not lose money (make a $1 a day), and this could expose you to streaks of losses.
if you could make that much money day trading wouldn't it be scale able and someone else would already make those trades?
My operating presumption is that there's runoff underneath market makers for small (on this scale) scalping. I also hope that those who need to return to investors or generate millions/year aren't trading the way I'd be trading to generate $200K/year.
I wouldn't count on it, especially for crypto. The market is brutal. I was playing around the other day with liquidity pools and some bot executed an arbitrage for a total gain of $0.1. That's how efficient the market is.
> Right now I'm trying to figure out how to consistently make $1/day as a POC exercise.
You can do this with the following strategy:
1. Buy $10,000 of Treasury bonds.
It might be more interesting to target a particular Sharpe ratio rather than an absolute dollar amount since that will be proportional to the amount you invest.
~$1.28/day if you just buy SGOV with that 10k.
Surely you can't? You don't have the latency advantages of big trading firms, nor the gazillion PhDs they employ. There's absolutely no way you are going to consistently make money, let alone $200k/year. At best you'll get lucky and think it's skill.
It's possible: note that each one of those quants is going to be making $400k at minimum, plausibly significantly more. So any high-maintenance strategy (eg. in some niche market) that makes less than that is just not going to be worth the time it takes to find, implement, maintain, etc.
Still not easy but it's not impossible.
Of course you can. If you have $17M invested in $VOO you'll make about 200k every year in dividends.
What's difficult is getting a higher rate of return.
$200k is totally doable with $2M investment capital and a bit of disciplined options trading.
>> $200k is totally doable with $2M investment capital and a bit of disciplined options trading.
10% probably. Though with the higher-than-market rate, there would also be dispersion in the consistency of the return. I could see some lean years where no money is made. I could also see a decade where you are buying at ever-higher valuations rather than dollar-cost averaging -- thus you could see a lean decade following.
It all depends on how much consistency you want.
Finally some advice for the common man.
Obviously we're talking about returns above the market. You don't need an algorithmic trading platform to invest in index funds.
Can you talk about maximum upside without discussing capital?
You definitely can, I know single people or small teams running strategies with 2+ sharpe making millions a year.
If you can make $1/day you're already 95% of the way there.
Obligatory xkcd: https://xkcd.com/1570/
Gold.