Arthur

3.9K posts

Arthur banner
Arthur

Arthur

@adezresearch

Humanist, philosopher, and environmentalist. Building things (AI-powered) that actually focus on making money as the main criterion, and retiring in the woods.

Katılım Temmuz 2022
127 Takip Edilen11.3K Takipçiler
Sabitlenmiş Tweet
Arthur
Arthur@adezresearch·
@discobray Use Obsidian, auto-memory, rule files, and hooks to avoid typical LLM mistakes You discover these patterns while using them, so I can't just share mine and expect them to work for you, your harness has to grow from your own friction to be actually good
English
0
0
2
194
Arthur
Arthur@adezresearch·
@discobray I've been using Claude Code for 8 months now, more than 1,000 commits this year, 8 working projects, 2 of them launched successfully, 1 open source with 350 stars, 1 about to be launched + automated a full trading system that does around 15% annually on IB, shoot your queestions
Arthur tweet media
English
1
0
2
490
bray
bray@discobray·
Two questions for Claude Code users: 1. What do you wish you knew earlier? 2. What’s your favourite use case? Teach me something and let’s start a thread together! 🧵
bray tweet media
English
4
0
4
1.6K
Arthur
Arthur@adezresearch·
@shachikyoto Are you in the discord? I’ll give you access to some things, I’m focusing on my private investments and research this days, but send dm
English
1
0
1
18
Shachi
Shachi@shachikyoto·
@ArthurResearch Hi Arthur, I miss your genuine data backed analysis. 🥺 Come back with something similar with this thread.
English
1
0
1
12
Arthur
Arthur@adezresearch·
Bitcoin Market Cycle Analysis This Is The Data You Need To See To Avoid Getting Trapped In A Bear Market: A Critical Examination of the data on Liquidity vs. Halving Narratives A new thesis dominates crypto discourse: "Halvings are obsolete, only liquidity drives Bitcoin, the bull market extends into 2026 or longer." But when institutional market makers synchronize this exact narrative, we must examine data, not consensus. Let's analyze rigorously:
Arthur tweet media
English
10
31
117
15.3K
Arthur
Arthur@adezresearch·
I’ve been following you since 15-16k followers and used to consider you a pretty decent account It’s quite sad you ended up becoming exactly what you’re calling out here There are so many people who paid you 2-3 thousand USD for basically nothing, unreplied DMs, hidden replies, stolen content, megalomany, and well, 2-3 bias post, hedged calls constantly, and reposts of the winners always, I have a pretty solid memory, and a lot of screenshots because i used to consider your fractsls nice Anyway, unfollowed and moving forward 👋
English
3
1
25
6.5K
Killa
Killa@KillaXBT·
My goal in joining this app was to show how it’s done. It seems like people are so used to others posting multiple scenarios and then deleting the ones that don’t work out, that they automatically assume everyone is doing the same. That’s not the case with me. I’ve never deleted a single post, my trades, analysis, and thought process are all fully transparent. Instead of jumping to accusations, take the time to actually pay attention.
English
145
55
2.5K
184K
Arthur
Arthur@adezresearch·
You put together one of the best teams for futures, to be honest, a lot of transparency, real traders trading with morals and commitment from what I remember Leo, Bray, Alex, and you were bangers there If you’re getting back to full-time trading, I’m actively working on a few fun and interesting trading projects to improve my workflows and the use of AI for this things (crypto side), to make more (and more easy) money, if you’re back trading, dms always open to share insights mutually 👋 Best lucks with your projects anyway!
English
0
0
3
1.1K
kane🧪
kane🧪@traderkane·
Posted this inside of @thelab last night - farewell present as we closed it last week, no more paid Discord. Back to being traderkane, not discordkane. It was a great two years; learnt a lot about myself, helped a ton of people and am happy with what we achieved.
kane🧪 tweet media
English
35
9
466
56.2K
Raychi
Raychi@raychix·
@ArthurResearch its not always about hitting every trade. risk management makes or breaks u thats the real game
English
1
0
0
130
Arthur
Arthur@adezresearch·
Real data from 10,000+ funded accounts that got paid (FundedNext, February 2026): — Median win rate of paid CFD traders: 50% — 41% of them won LESS than half their Trades — They still got paid The difference is the median R:R of 1.49:1 Profitable trading isn’t about being right more often. It’s about losing well. You know how you actually solve trading psychology? You remove yourself from the equation Agents and automation. Let the system execute what you already know works , without you getting in the way.
Arthur tweet media
FundedNext@FundedNext

x.com/i/article/2029…

English
2
1
17
3.6K
Arthur
Arthur@adezresearch·
Financial agents are the future And now Anthropic just proved it with data Their new labor market study ranks financial & investment analysts as the 7th most AI-exposed occupation in the entire US economy , 57.2% observed exposure Not theoretical, actual , happening-right-now automation Most people read this as a threat. I read it as a roadmap The firms still paying 6 figures for someone to compile reports and forecast market conditions are operating on borrowed time The ones building agents to do it faster, cheaper, and 24/7 are already ahead Keep tune for my next post
Arthur tweet mediaArthur tweet media
Arthur@adezresearch

Hundreds of autonomous agents for trading Scanning the market from different data providers, different strategies, interacting with each other. If you want to see what real emergence in trading looks like, and what comes out of this open project, stay tuned. Conway for financial markets, coming soon.

English
0
2
25
7.6K
Arthur
Arthur@adezresearch·
Qualitative vs. quantitative analysis A mechanical strategy stripped of context will almost always look broken in a backtest, No macro filter, no higher timeframe bias, no session awareness, no news consideration Of course the results are poor @MrZincx version isn't a different strategy, it's the same one, properly contextualized. That's what makes it work I understand Zinc has a bias, he's promoting a platform, an indicator, and memberships, which isn't inherently bad if done ethically. But this is exactly what I don't like: no one can argue against the effect it has on people who believe that a single-timeframe indicator will get them the same results as someone with years of context behind every decision. That type of marketing, even when done ethically, is by default misleading. You're selling a tool as if it were the edge itself. This is why I emphasize that trading is hard, but teaching trading is harder, most traders apply layers of context intuitively without realizing it The second you strip that out and hand someone a naked mechanical system, you've removed the actual edge
Chris™@ProbableChris

Here is the 5yr backtest for this exact strategy. -IB defined as 930am ET to 1030am ET -Waits for IB range to close. -IB has not yet been broken. -If low formed first, buys the mid point, stop loss is IB low, PT is IB high. -If high formed first, shorts the mid point, stop loss is IB high, PT is IB low.

English
2
5
29
8.1K
Arthur retweetledi
Arthur
Arthur@adezresearch·
Hundreds of autonomous agents for trading Scanning the market from different data providers, different strategies, interacting with each other. If you want to see what real emergence in trading looks like, and what comes out of this open project, stay tuned. Conway for financial markets, coming soon.
English
13
18
192
17.7K
Arthur
Arthur@adezresearch·
@Menace_ICT Put notifications on, I’m updating this this weekend
English
2
0
1
237
Synapse.X
Synapse.X@Menace_ICT·
@ArthurResearch This is really something - Thank you Does it run on Local Models ?
English
1
0
1
223
Arthur
Arthur@adezresearch·
@SectorX_AI Join the discord and share with us brother!
English
1
0
0
330
SectorX
SectorX@SectorX_AI·
Good to see this Arthur!, I'll be starting on similar after I wrap current project, it will be an upgrade to this simple grid. I was going to use TouchDesigner but it seems AI is much more competent in D3.js. Mine will be tracking distance, movement and all things relative between bodies... er I mean assets hehe 🔭🪐💫
SectorX tweet media
English
1
1
0
476
Motif
Motif@TraderMotif·
It now takes me less than 1 minute to see stats going back years for any timeframe, any asset, any system. I can then tweak parameters to make the edge better. The days of sitting at the desk backtesting one system for weeks is gone. Truly incredible.
English
10
1
98
8.9K
Arthur
Arthur@adezresearch·
@bltzaXII @kurotool Hey! could you share some proof that it’s working live? (wallet history, on-chain txs, anything like that) Even if it’s not a finished product yet, I’d love to check it Sending you a Discord invite dm in case you’d like to join
English
0
0
1
156
Arthur
Arthur@adezresearch·
I've met real quants, mathematicians, engineers, who are extremely poor traders, completely outperformed by many traders I've come across here who work with hybrid heuristics Wonder why that is
Arthur@adezresearch

Qualitative vs. quantitative analysis A mechanical strategy stripped of context will almost always look broken in a backtest, No macro filter, no higher timeframe bias, no session awareness, no news consideration Of course the results are poor @MrZincx version isn't a different strategy, it's the same one, properly contextualized. That's what makes it work I understand Zinc has a bias, he's promoting a platform, an indicator, and memberships, which isn't inherently bad if done ethically. But this is exactly what I don't like: no one can argue against the effect it has on people who believe that a single-timeframe indicator will get them the same results as someone with years of context behind every decision. That type of marketing, even when done ethically, is by default misleading. You're selling a tool as if it were the edge itself. This is why I emphasize that trading is hard, but teaching trading is harder, most traders apply layers of context intuitively without realizing it The second you strip that out and hand someone a naked mechanical system, you've removed the actual edge

English
0
1
17
2.3K
Arthur
Arthur@adezresearch·
That's exactly why what I build are deterministic entry models with added layers of context.
Arthur@adezresearch

Qualitative vs. quantitative analysis A mechanical strategy stripped of context will almost always look broken in a backtest, No macro filter, no higher timeframe bias, no session awareness, no news consideration Of course the results are poor @MrZincx version isn't a different strategy, it's the same one, properly contextualized. That's what makes it work I understand Zinc has a bias, he's promoting a platform, an indicator, and memberships, which isn't inherently bad if done ethically. But this is exactly what I don't like: no one can argue against the effect it has on people who believe that a single-timeframe indicator will get them the same results as someone with years of context behind every decision. That type of marketing, even when done ethically, is by default misleading. You're selling a tool as if it were the edge itself. This is why I emphasize that trading is hard, but teaching trading is harder, most traders apply layers of context intuitively without realizing it The second you strip that out and hand someone a naked mechanical system, you've removed the actual edge

English
0
1
8
1.3K