back from the dead edition>quant newshttps://alphapulse.org (beta release)https://www.ft.com/quant-investing>quantopian lectureshttps://www.youtube.com/playlist?list=PLRFLF1OxMm_UL7WUWM31iynp0jMVf_vLW>event driven backtester designhttps://www.youtube.com/watch?v=OhGdVdolqkU [Embed]>defi/mevhttps://info.arkm.com/research/beginners-guide-to-mevhttps://www.monoceros.com/insights/maximal-extractable-value-bookhttps://arxiv.org/html/2406.16573v1>recommended brokersstocks and options - https://alpaca.markets (Robinhood doesn't have an official API)futes - https://www.tradovate.com (modern API)crypto - https://www.kraken.com (data: https://support.kraken.com/articles/360047543791-downloadable-historical-market-data-time-and-sales-)Discord: (Code is temporarily being hidden)prev: >>61197029
my thanks to the prev baker >>61197131hyperlquid is a dex, I didn't list any DEXes as I felt a user algotrading DEXes probably knows what they're doing already. on the next bake, I'll include some for crypto derivatives since us amerifats dont' have access to those instruments yet except through DEXes.
>updates I think I almost have my multi-hop arbitrage system working (just two more weeks!), just ironing out bugs in the swap simulator right now. currently only works with Uniswap V2/V3 protocols and derivations like Quickswap V3's Algebra, but easy to add more once the core works. I want to get back to working on a 0dte SPY bot, it would be fun to try running a very fast ML system that determines optimal options spread (or single leg if optimal). Or maybe some SPY-VIX statistical arb thing. still don't expect the defi arbitrage thing to turn any profit as a vanilla inter-dex arb machine (especially without mempool access) but the intent is to use it as base infra and explore niches within defi or use the same infra code for tradfi.
I remember being very excited about this 10 months ago before finding a bug that invalidated the results. I only use event driven architectures now due to this.
i wanted to get into this years ago. even made an alpaca account etc but never really started. idk would love to get something simple done to back test it and maybe deploy. crypto seems easier though for testing. still cant believe how well all these mev bots work to snipe retards all day on ETH.
Some spreads across DEXes I'm manually verifying now. The highlighted one was accurate, but there's just dust in the pool so no arbitrage possible (only 1 WPOL - ~$0.19 of luquidity available). Having Codex write a separate script that takes a list of pools and addresses to confirm my system is generating accurate data now. Probably add a button for it that calls the script with the metadata it needs. The script will just be a simple Python script that makes the required RPC calls on those pools/tokens to confirm, using separate code from my systems code so there's no contamination. I only use RPC calls sparingly and aside from when discovering a new pool, the system accumulates state lazily so I don't to pay RPC cost. I just switched to Codex from Claude after discovering this repo: https://github.com/just-every/code -- I found Codex rather unusable without, but with this Code wrapper for Codex I feel I'm getting much more intelligence out of the tool). Should probably include some intro comp sci / programming / coding workflow stuff in the OP at some point. Gemini 3 is about to drop too.
When my system has basic functionality I'm going to rewrite everything by hand though. Over time I have come to regret not investing in and maintaining my developer skills and loosing the intimate deep codebase knowledge that's forced on you when writing by hand. The AI's offer a hypothetical increase in development speed at the cost of >not using your brain as often - I'm shitposting right now while Codex is working>not maintaining your skills - I would honestly struggle getting a hello world program to compile in any language right this second without looking up syntax >as a result, becoming dependent on costly third party sources for your development / brain>when they inevitably rate limit you, jack up the rates, drop the quality etc, you're screwed>when they inevitably produce bug-ridden code you don't understand and can't trust, you're screwed>when it inevitably takes even longer to use LLM's over just doing everything by hand due to the compounding benefits of it (using your brain, intimately knowing the codebase, not writing hallucinated buggy code designed to look correct, etc), the entire purpose has been undermined and you find yourself poorer and less skilledIf I ask myself which developer I'd rather be it's definitely the skilled one over the unskilled. IDK, when I started this project back in March I didn't think it'd take this long or be this complex (it started as a simple one-off Python script to implement a given paper, based off this code: https://github.com/zzzac/rule-based-forex-trading-system), so AI assistance for simple data scrapers and data parsing made sense..
>>61201762>i wanted to get into this years ago. even made an alpaca account etc but never really started.Same, I'm trying to break out of the mindset that there's no point because everythings already been done by some army of Chinese math olympiads and warehouse of GPUs. I was lucky to have a pretty simple algo be profitable initially and that emboldened me to keep pushing through it.>idk would love to get something simple done to back test it and maybe deploy.Same, I'm trying to make a site that makes backtesting really easy, feel free to check it out it's the beta release in the OP. Homepage is a blank news feed right now (hit the ngrok limit on requests), then there's a develop tab (essentially an IDE w/ terminal with BYOK AI model), a research page for running backtests with a Jupyter notebook UI, and then a markets page which I intend to be a sort of customizable market dashboard for people. Working on the site hasn't been a priority lately though, so everything is mocked up and not polished, just "concepts of a plan".
>>61201740Congrats you made a super natural algorithm not even fucking intelligently
Wow thanks for making another thread ignoring mine guys.>math olympiadsJust indicator slop and ping trumps twitter account if youre feeling fancy. Crypto doesnt have revenues or price earning ratios and shit.
>>61202064where is yours? I cntrl + f'd "algot" and didn't see anything. that image is too small to see. are you just scraping twitter? their API costs are bonkers. $5k+ a month for the minimum level of usability.
>>61202080Since the market pumps or dumps based on the tweets of the orange king you can find sn alt source for his tweet like cnn or use puppeteer.
>>61202102Yeah, or just scrape it. But being able to plug up to twitter for larger scale analysis would be cool. Blue Sky has no API cost, but isn't as active. I wish it were. Do you run the tweets through some sort of LLM or something for parsing and deciding how to react? I've heard Trump's tweets are notoriously difficult to analyze because of how phrases things. Grand insults and compliments all rolled together and such, he weaves it bigly.
>>61202136Theres sentiment analysis but id just set it to sell and pause after a tweet just in case.
>>61202254Saw your other thread, what are you trading with this >>61202064? It seemed to be calling lows and highs with extreme accuracy. Is that image from a backtest or results from live trading?
>>61202285Well id give the pinecode but thought i should ask if i can have like $5 worth of eth.Its not begging cause im selling alfa0xB39F7ffb9205e2089E339DF27cA45129fb0ECb52
>>61202304if it's trading live like that, you'd be able to make much more than $5.
>>61201936>I just switched to Codex from Claudeupdate: it's doing really well, but I'm 5hr limits too frequently. About 2-3 hours in. Super annoying, gonna spring for a local LLM setup once the hardware hits a reasonable price point for good performance. I'd happily pay ~$10k if I meant I could run Codex-GPT-5 level LLM locally 24/7.
>>61203318>I'd happily pay ~$10k if I meant I could run Codex-GPT-5 level LLM locally 24/7.just switch to API usage and pay for the usage when you arent on the plan. then switch back when the time is over.
Provided I can make a single cent from arbitrage, I'm probably going to get pretty into MEVhttps://docs.flashbots.netbut I really do want to get back to working on my 0dteS SPY bot. need to start hitting 5% daily Can someone here tell me why I've never heard of ALEO crypto until now? Is it a scam? I'm tempted to get a bag if everything checks out
>>61203540>just switch to API usage and pay for the usage when you arent on the plan. then switch back when the time is over.what? oh you mean like when I hit the 5hr limit, switch over to API? I'd blow through my net worth if I did that though. $200/mo is already too much.
Tried freqtrade with binance and was quite underwhelming desuanyone here has been successful?
>>61204164What did you try exactly?
>>61204179thishttps://github.com/iterativv/NostalgiaForInfinity
>>61201651>my thanks to the prev baker You are quite welcome >>61201629Regarding the discord code, I think you can go ahead and share the code, unless you think something more should be done to itSpeaking of which, I'd like to ask your opinion on the forum post I made regarding a generalized model framework, if you'd care to take a look
>>61203318I am telling you the $4K Nvidia box looks pretty good
I was beginning to think this gen was gone for good, ive been searching the catalog regularly for it with no results :(Just trying to get started, I got an alpaca API key and started putting some stuff together in python. Besides the links in the OP, does anyone have suggestions for reading material/lectures that would help me at this very early stage? thanks
>>61204196what's the TL;DR on the strategy is used? anyway, you can't really expect to just download a bot with a preconfigured strategy and make any money.
>>61205411>>61205396>DGX Spark I guess the Mac M4 Max Studios are pretty good for running inference too. I did check out the Spark and it's got me GASing for local llm setup. Re the model framework, NautilusTrader has this posted on their website and I think it's a good model, similar to what you posted. I tried using Nautilus for awhile but it has a lot of boilerplate just to get a simple thing going and the author was kinda weird when I shared some work I had done for front-end for it (since the boilerplate was annoying) so I decided to write my own. My version is architectured slightly different to be a little more flexible and performant.Just blew through my weekly limit for Codex, only lasted a single day! Terrible.
>>61201629Do you have any youtube videos you recommend to watch to get started?
>>61207310The cheapest option overall is to get an RTX5060 and running that, however the DGX has quite the pow pow vroom>NautilusTrader has this posted on their website and I think it's a good model, similar to what you postedthanks, thats useful, yeah I will redo the model to involve less grouping, but first I will write a massive one which has everything I want my system arhitectureto hold with threading, ssh, SQL DB etc
>>61207349Yeah I think for me mostly I just want inference, not training, to run a local LLM and stop paying subscriptions every month. Then later I’ll grab a gpu for training. Another thing I didn’t like about Nautilus is that there wasn’t a way to run parallel backtest. If I wanted to grid search a bunch of indicators, it’d do seperate complete backtest for each (iterating over the entire dataset in each instance). I set my mine up so that I can run any number of strategy services in parallel that consume the data so I only iterate over it once, and there’s no state leakage since they’re separate services. >>61207349The quantopian lecture series is a good place to start for the theory. If you just want to get your hands dirty asap I don’t have a good place i can recommend yet, I’m hoping the first link in the OP will be good for that when it’s finished.
>>61207349Also, TradingView is an easy way to poke around, if it's super limited and inaccurate. You can have an LLM generate whatever scripts you want to create indicators and strategies and get an idea of the concept of automated trading.Personally I tend to work backwards from what the statistics say and refine a strategy from there. Or you can do something that's essentially strategy-less like arbitrage or statistical arbitrage. I've come to think that a strategy-less approach is a good way to start, since it simplifies the infrastructure required and gets you trading and doing analysis sooner (as opposed to endlessly dicking around in a Jupyter Notebook).
>>61207456Thank you I appreciate it. Yea I've utilized LLMs to generate scripts but just fed to cursor or even omen. Haven't gone past that yet but I need to stop dragging my feet.
>>61207688I've been trying to work on this site when I get time, the idea being you could just pop open a strategy and get to work inside a Jupyter notebook almost immediately, tweak things etc. Hoping to lower the barrier to entry essentially and the backend will be open sourced
>>61205411It will be slow for inference and training because apparently it has unified memory with low memory bandwidth. Terrible.According to some people it can be used to fine tune, train low ranks or to prototype your workflow to see if your setup works for real chips with HBM3.
>>61207805I thought unified memory was a good thing? IDK anything about hardware spec stuff though desu, I've been a 2GB RAM stinkpad X60 user for most of my digital life. Haven't really felt a need for better hardware until recently.
>>61207773The UI is beautiful. looks great nice work. Excited to see more of it.
A bit off topic but why is no one talking about Bitcoin implementing ZKPs? This is huge if it's capturing the technical benefits of ZKPs in the same way ZEC does. Imagine if daddy Bitcoin could be used as a privacy coin, that'd be insane.https://bitcoinmagazine.com/technical/a-zero-knowledge-proof-is-verified-on-bitcoin-for-the-first-time-in-historyhttps://www.youtube.com/watch?v=1gMORYs10Xkking ohttps://arxiv.org/html/2507.21085v1o>>61208443Thanks for the feedback! I'm going to resume working on it once the backend trading framework is done. Hoping for a functional beta by Christmas. One of the design goals is to enable functional use on a mobile device.
>>61208546looks like this all concerns zk-rollups, not privacy. too bad
>>61208546>>61208763section on privacy at the end of the arXiv paper
Bump for keeping this alive
This is the best explanation of ZKPs on the webhttps://blog.cryptographyengineering.com/2014/11/27/zero-knowledge-proofs-illustrated-primer/
>>61209574Bitcoin will never have privacy. If it takes over they will know you bought a rolex snd youre going to have to deal with it.
>>61209591Maybe, maybe not. I don't think the core team is as opposed as people think, they're just conservative.
I think I'm finally possibly detecting arbitrages with accurate simulation amounts. Whether or not I can execute on these remains to be seen.
>>61207310>Just blew through my weekly limit for Codex, only lasted a single day! Terrible.Got my subscription refunded, tried out GLM 4.5 for $3/mo and it's doing better than Codex was.>China numba wan!
>>61202304>gimme 5 USD sar, im giving alfa!
>>61210009Long story but i cant buy or sell crypto rn and its stuck. Last thing i bought was tether gold. :/
>>61210038dear saars, are you saying you cannot redeem?
it's overbut honestly Im really impressed with GLM 4.5, and from I understand this can run on a Mac (which happen to be better for inference), so I think I'm gonna go with that. I'm loving GLM 4.5 routed through Claude Code's CLI right now.
Going to par-bake the next OP so I don't forget to add some links:>quant newshttps://www.ft.com/quant-investinghttps://alphapulse.org (beta release)>videos https://www.youtube.com/watch?v=OhGdVdolqkU [Embed] https://www.youtube.com/playlist?list=PLRFLF1OxMm_UL7WUWM31iynp0jMVf_vLW>introshttps://www.quantstart.com/articles/Successful-Backtesting-of-Algorithmic-Trading-Strategies-Part-I/>defi/mevhttps://docs.flashbots.nethttps://medium.com/@solidquant/how-i-built-my-first-mev-arbitrage-bot-introducing-whack-a-mole-66d91657152ehttps://ethereum.github.io/yellowpaper/paper.pdfhttps://news.polymarket.com/p/automated-market-making-on-polymarket>recommended AI/LLM (plugs into Claude Code CLI), $3/mo:https://z.ai/model-api (top performance, plugs into Claude Code CLI, only $3/mo!!)>bookshttps://github.com/zslucky/algorithmic_trading_bookhttps://www.amazon.com/Quantitative-Momentum-Practitioners-Momentum-Based-Selection/dp/111923719X>recommended brokersstocks and options - https://alpaca.markets (Robinhood doesn't have an official API)futes - https://www.tradovate.com (modern API)crypto - https://www.kraken.com (get free data: https://support.kraken.com/articles/360047543791-downloadable-historical-market-data-time-and-sales-)>githubs & stuffhttps://skfolio.orghttps://github.com/joelowj/awesome-algorithmic-trading?tab=readme-ov-file>unconventional yet succulent cornbread recipe https://www.shugarysweets.com/instant-pot-cornbread/ (I add jalapeños and corn as well personally)
>>61211018He forgor the discord code
You guys are in luck because if just so turns out that trump actually uses truth social not x. So you can use that for your (my)bot.
>>61211895Is your bot just a Trump detector?
>>61212091What else is needed. Just have it sell when he tweets.
>>61210864Most important to me is the 120GB of VRAM, so you can run a bunch of models and it only runs 200W, I already knew the performance was subpar from videos covering it and comparing it to rigs with two RTX5090 running together
Astrology shit MovingAverage: Price=$114540.61, ShortMA(50)=$114347.24, LongMA(200)=$109002.99 Golden Cross detected Price near MA, reducing confidence MovingAverage: Signal=BUY, Confidence=56.0% VWAP: Price=$114540.61, VWAP(20)=$111651.49 Price >2% above VWAP - sell VWAP: Signal=SELL, Confidence=75.0% RSI: Current RSI(14)=58.28 RSI in neutral range RSI: Signal=NEUTRAL, Confidence=50.0% BollingerBands: Price=$114540.61, Middle=$112074.28, Upper=$119299.05, Lower=$104849.50 Price in middle range of bands BollingerBands: Signal=NEUTRAL, Confidence=60.0% MACD: MACD=-596.01, Signal=-1136.91, Histogram=540.90 Bullish MACD crossover MACD: Signal=BUY, Confidence=70.0% SupportResistance: Price=$114540.61, Support=$102000.00, Resistance=$126199.63 In upper half of range SupportResistance: Signal=NEUTRAL, Confidence=50.0% ATR: Current=2306.70 (2.01% of price), Avg=3674.49 (3.21% of price) NORMAL VOLATILITY regime High volatility market - reducing signal confidence ATR: Signal=NEUTRAL, Confidence=45.0% MeanReversion: Z-Score=0.68, RSI=58.3, Price=$114540.61 SLIGHTLY OVERBOUGHT: Price > 2% above SMA MeanReversion: Signal=SELL, Confidence=65.0%
>>61212097Well I wish you luck with that>>61212176Where are the charts? I cant read, please help
>>61212628Sorry, I'm retarded and I can't draw lines.Astrology shit MovingAverage: Price=$115457.35, ShortMA(50)=$114365.57, LongMA(200)=$109007.58 Golden Cross detected Price near MA, reducing confidence MovingAverage: Signal=STRONG_BUY, Confidence=68.0% VWAP: Price=$115457.35, VWAP(20)=$111668.08 Price >2% above VWAP - sell VWAP: Signal=SELL, Confidence=75.0% RSI: Current RSI(14)=60.28 RSI near overbought (>60) RSI: Signal=SELL, Confidence=65.0% BollingerBands: Price=$115457.35, Middle=$112120.11, Upper=$119418.16, Lower=$104822.07 Price in upper 30% of bands BollingerBands: Signal=SELL, Confidence=70.0% MACD: MACD=-522.88, Signal=-1436.05, Histogram=913.17 Bullish MACD crossover MACD: Signal=BUY, Confidence=70.0% SupportResistance: Price=$115457.35, Support=$102000.00, Resistance=$126199.63 In upper half of range SupportResistance: Signal=NEUTRAL, Confidence=50.0% ATR: Current=2306.70 (2.00% of price), Avg=3674.49 (3.18% of price) NORMAL VOLATILITY regime High volatility market - reducing signal confidence ATR: Signal=NEUTRAL, Confidence=45.0% MeanReversion: Z-Score=0.91, RSI=60.3, Price=$115457.35 SLIGHTLY OVERBOUGHT: Price > 2% above SMA MeanReversion: Signal=SELL, Confidence=65.0% VolumeProfile: Found 6 high-volume nodes Insufficient volume data VolumeProfile: Signal=NEUTRAL, Confidence=50.0% ForceIndex: Current=5552944.34, Smoothed=-6871175.86, Price=$115457.35 Mixed signals ForceIndex: Signal=NEUTRAL, Confidence=50.0% IWantToSuckCocks: FeminineCocksOnly=CONSOLIDATING, Noisy=true, NoiseLevel=1.1x High noise environment (1.1x avg) - reducing trading confidence Market: Context=CONSOLIDATING, Confidence=30.0%
>>61213152>>61212176stop
>>61204196This 'strategy' is curve fitted to hell and back don't use it or you'll get fucked. If you wanna use freqtrade you need to make your own strat. You could also check out one of their strategy repositories, but pretty much all public stuff is trash.Personally I use freqtrade + custom hyperopting with some success. You need to be able to at least understand python basics + vibe code to get anything out of freqtrade imo
>>61214308>python basics + vibe code>profitable strategyNot really, you have to learn a lot of stuff. You cannot really brute force a profitable strategy. And if you lack mathematical or statistical understanding, no amount of LLM slop will make up for that.
>>61214663No bros just throw all those fancy strategies to the trash and stoch rsi macd and truth social for selling whenever trump tweets and that's it.i want to say sentiment analysis but the ai wont get trumps tweets.
>>61214828I wish 4chan would let OPs filter out posters thread-wide. You're just posting garbage anon
>>61214663>you have to learn a lot of stuffOf course, but LLMs make this so much faster and easier (unfortunately not easy) than a few years ago if you don't prompt like 'sir pls giv 100% redeem strategy pls sar village cow very sick'For explaining finance lingo, establishing rules for a robust back/forward testing regimen, quick notebooks to visualize you backtests, other odd tasks etc. ai is god tier imo
>>61215163Well yeah i mean if you want to make a god tier bot youd pay for the x api. Indicator slop just tells you when to enter and leave but it wont give you alfa.
>>61216173>For explaining finance lingo, establishing rules for a robust back/forward testing regimen, quick notebooks to visualize you backtests, other odd tasks etc. ai is god tier imoWhat you described still requires a lot of background knowledge. You need to know a lot of stuff to get anything of value out of LLMs, since you basically have to define exactly what the features of the product or strategy you intend to make. If you just tell it to make a profitable strategy using MACD and RSI, it will just copy many and merge many of the jeet tier strategies it has been trained from. Trust me on this one, it is not possible to get a profitable bot without a solid Math background.
Well today has sucked balls for me so far. My honeymoon phase with GLM 4.5 has worn off thoroughly and turned into more of a hangover. I have come to value detailed mermaid diagram documentation and wasted time running a local Mermaid Editor server before looking into VS Code which can support them natively. But I can’t stand VS Code (emacs user). I read about Zed and it seemed like a great fit - written in rust, extensible, etc and upon installing it loved the look and feel. Unfortunately ChatGPT lied to me when it said it had mermaid diagram support, it only has syntax highlighting support. Not deterred by this I began writing my own extension, only to find the extension API is practically non-existent, limited to editing modes not do-whatever-you-want like Emacs. So I wasted about six hours total. I cloned the repo to see if I can get something working, but had to install XCode, but had to update my system first. So that’s where I’m at!
>>61216729How solid? Should I learn manual trading first? My Math background is mostly lineal algebra and differential calculus.
Zed editor is insanely dank. >>61217779I disagree that you need a solid math background at all. What you need is the ability to gloss over math and ignore it, only returning to reference it if needed, and a high level understanding of the processes. The math is all obviously delegated to the computers, and any implementation already done in some library.
>>61217842I think I’m going to try to have my trading framework be integrated into Zed. It uses GPU for graphics so it should be able to run Jupyter Notebooks and Lightweight Charts as plugins.. It’d be so dope
>>61201629Has anyone tried implementing an ichimoku cloud based bot?
>>61217779>My Math background is mostly lineal algebra and differential calculus.Thats enough, complement with advanced statistics and you will be alright for the most part.>I disagree that you need a solid math background at all.>>61217842If you disagree, you probably have a solid math background yourself (people with good math intuition tend to underestimate how clueless mathlets are), and if not, you probably would have a way easier time to optimize and model good AI based strategies leveraging mathematical functions such as Fourier decomposition (helps greatly to improve some noisy base features).LLMs are not nearly as good as people paint them to be, for the most part LLMs are super weapons only in hands of people that are already good at their craft and have good verbal intelligence.
>>61221417>advanced statisticsdo you mean something like monte carlo analysis or z-score correlations or something else? Can you pls give an example?>improve some noisy base featuresAre you familiar with the works of John Ehlers? tldr he was a radar dude at nasa and is now apllying those filtering calculations to charts, pretty interesting read imo
>>61217779>Should I learn manual trading first?learn prediction markets. prophex is exactly where you should begin
>>61221670>do you mean something like monte carlo analysis or z-score correlations or something else? Can you pls give an example?Bayesian Statistics, Markov Processes. And for math, stochastic processes.>>61221670>Are you familiar with the works of John Ehlers?Yes, cool stuff and really really related to some really interesting stuff i'm working regarding strategy optimization.
>>61222077Ah, also i recall some anon mentioning group theory, i have not used it directly on my strategies, but i can only imagine what is he using that for.
>>61221417>you probably have a solid math background yourselfYeah, fair enough. But what that really taught me was to glaze over the math and get comfortable ignoring it outright - viewing it as more like numerical plumbing in a higher-level process. Just like I don't need to know thermodynamics to work on my car. >LLMs are not nearly as good as people paint them to be, for the most part LLMs are super weapons only in hands of people that are already good at their craft and have good verbal intelligence.I agree with this completely.Not that this is a unique idea, but on a societal level, I think LLMs are going to end up being like leverage on IQ. Smart people will fly away while dumbos get dunked on.
TOTAL DUMBO DEATH
>>61208763uh, it's largely related to privacy also
>>61222077>Yes, cool stuff and really really related to some really interesting stuff i'm working regarding strategy optimizationWould you be willing to sharesome infos (maybe just in the broader sense)I use his cycle calculations (within upper/lower bounds) as input length for indicators. I found that this improved resilience to overfitting and also improved profits>>61222245>Smart people will fly away while dumbos get dunked onthere's already studies showing brainlets are actually getting dumber due to overreliance on ai lel
>>61218923I implemented a very basic Ichimoku strategy (just moving averages cross and not taking trades depending on clouds. My backtest shows 4.32% gain (worse than just buying and holding but it works in bear market condiitons too).I will try and improve it and maybe post the code here if anyone is interested.
>>61221959What's prophex?
>>61222404>uh, it's largely related to privacy alsoseeing houdiniswap with that massive onchain liquidity across chains makes it one privacy protocol to rule for years
1% net average each 6 h (1.5% maybe considering commissions), using high leverage and few signals, stoploss of 30%, BTCUSDT. reinvest all capital. possible?
>>61222404Not for the Bitcoin implementation portions I was looking at. I thought they were experimenting with a shielded pool but it was just rollups. That wouldn’t do anything for L1 privacy.
