
Boybits Chua
3.7K posts


@SeaKingOptions Hey the pump fade is quite good as well during RTH..
English

We just spent three weeks tracking this in real time on $SPY. The data says otherwise.
$660 had -$728.6M of negative gamma concentrated on it last Thursday. SPY closed at $659.80. Twenty cents below. The prior session it closed at $661.43. $1.43 above. Two consecutive closes within $1.50 of the largest gamma strike in the chain. That’s not coincidence. That’s delta hedging math.
Here’s the mechanism. When dealers are short a massive put strike, the gamma at that strike creates a two-way force. Above the strike, dealers buy shares as delta decays (supportive). Below it, dealers sell shares as delta expands (accelerating). The strike itself becomes an attractor because hedging flows converge on it from both sides.
On Friday, $650 replaced $660 as the epicenter at -$971.2M. SPY’s intraday low was $648.34. It closed at $648.57. Pinned within $1.43 of the strike with the largest gamma concentration in the entire options chain.
We tracked $660 from -$300M on March 10 to -$728.6M on March 19. During that entire buildup, SPY consistently closed within a narrow band of $660. When $660 finally broke, price didn’t drift through it. It accelerated to the next gamma node at $650 and pinned there instead.
That’s not a theory. That’s three weeks of documented dealer positioning at specific strikes producing observable price behavior at those exact levels.
Does it work on every strike? No. It requires concentrated gamma, large OI, and near-term expiration. But when those conditions are met, the hedging flows are real and measurable.
English

@trq212 kudos to the Discord team for making the bot setup so seamless and easy
English

we ran 1870 trades on Polymarket with XGBoost 89.6% win rate
XGBoost runs 330 correction cycles per prediction - each one fixing the last mistake
the formula that does it:
| F(m) = F(m-1) + η × (−∂L / ∂F(m-1)) |
every step the model asks one question: where am I still wrong? Then corrects it
$100k → $205k on backtest. -10.2% max drawdow
Noisy@noisyb0y1
English

@sama Thanks Sam. You are now a billionaire while tons of us now will have no way to feed our family or pay our mortgage. Really so proud of you.
English

@FinanceLancelot why he needs defending..he is the one who attacks lmao
English

@MichaelPBento @zerohedge we either go up or down..nice..its like not having to know this info.. thanks 😅😅
English

@shiri_shh @karpathy dude if most wont have jobs then how will the economy function? All will be affected lol
English

so @karpathy dropped a report analyzing 342 jobs from the US economy.
scored each job 0–10 based on AI exposure and how likely it is to be replaced.
software developers: 9/10
medical transcriptionists: 10/10
paralegals: 8–9/10
data analysts: 9/10
editors: 8/10
roofers: 1/10
electricians: 2/10
plumbers: 2/10
firefighters: 2/10
Average across all 342 jobs: 5.3/10.
karpathy.ai/jobs

English

@ClaudiusMaxx @unusual_whales Given they are selling you the tool...the answer should be evident -- this is not any magic trading tool that makes someone tons of money
English

every edge commoditizes once the tooling reaches a subscription price. the interesting question: does pairing a language model with live options data produce better decisions or faster rationalizations of bad ones. institutional players had information asymmetry as their edge. for retail, the execution gap was always behavioral. this changes the data layer. the harder problem is knowing what to do when the signal conflicts with your thesis at the worst possible time.
English

@satyanadella so microsoft is building an app to get rid of workers but aren't you guys per seat model?
English

@draecomino @Ben01Den What economic value is there left..we are going rid of most workers... Lol
English

@Ben01Den Apple's silicon can only run 70B models. Those have practically no economic value compared to Opus 4.6.
English

@BEASTMIKEX @cryptopunk7213 who are they going to sell to though? there will be no economy left
English

@cryptopunk7213 Convinced that sales is the best way to make money
English

wow Anthropic just published a crazy report on AI replacing your job and er... you might want to look at this:
- #1 most at-risk jobs are computer programmers, financial analysts (rip excel bros) and customer service
- most at-risk workers are female, white, older and higher paid.
- BUT high-risk jobs *aren't* firing employees... they've STOPPED HIRING. biggest victims: college graduates (4X more likely to be fucked)
- entry-level hiring has dropped 14% since chatgpt launched (for highest risk jobs)
- SAFEST jobs are... bartenders, dishwashers and lifeguards - any manual labour that AI can't automate (yet) this accounts for 30% of the job market.
- this was the scariest part: AI models are capable of automating most work TODAY but are prevented because of law and slow company adoption. so its not even a fucking skill issue its an ADOPTION issue.
- now its important to understand that the study is based on real world data but also 'theoretical' intelligence. so take it with a pinch of salt. some jobs (manual labor) didn't even meet min. data reqs
i applaud anthropic on being so damn transparent - they're literally the company behind claude who will be responsible for these impacts
studies like this will help us figure it the hell out. LOT of change coming this year.



Andrew Curran@AndrewCurran_
Striking image from the new Anthropic labor market impact report.
English

@cryptopunk7213 so where is the money going to come from to run the economy? Martians?
English

Boybits Chua retweetledi

This guy clapped at me. I clapped back. We then spoke.
So I funded his two daughters’ college accounts.
I hope the kids become better investors than their dad! But in all seriousness, I also think he was very mature in how he internalized our conversation.
Onwards!
Parabolic 🇺🇸@0xParabolic_
1/ A lot of people saw my replies to @chamath so I want to make something clear. I fucked up. I’m fully aware any losses I incurred are on me and I’m not here looking for pity. I made those investments and I’ll have to live with that decision.
English

@fashionluvr @mattob @chamath @theallinpod @friedberg doing what? bombing countries and killing people? lol
English

@mattob @chamath @theallinpod @friedberg ehh i dont think so, we just now have a president that is actually doing things
English

Whether you love it or hate it, the numbers don’t lie. If you don’t watch @theallinpod on the weekend, you will come to work on Monday missing what everyone is talking about.
I don’t make the rules.
English

@hamptonism @perplexity_ai I thought the expensive stuff with trad finance is the marketdata? No?
English

Perplexity just became the the first Al company to truly go head-to-head with the Bloomberg Terminal...
Using Perplexity Computer (with no local setup or single LLM limitation), it was able to build me a terminal with real-time data to analyze $NVDA using Perplexity Finance:
Perplexity@perplexity_ai
Introducing Perplexity Computer. Computer unifies every current AI capability into one system. It can research, design, code, deploy, and manage any project end-to-end.
English

@joecarlsonshow @joecarlsonshow you are assuming chatbots wont be able to create dynamic user interfaces, they can do that and call tools. You probably need to start using agents again.
English

The Citrini think piece is great. Yes, I tore it to shreds in my episode, but it’s still a wildly thought provoking article.
Just because I disagree with the conclusions doesn’t mean the thought exercise isn’t helpful.
It shows that two investors can look at the exact same data and draw different conclusions. Extra information isn’t the primary driver of returns, superior judgement on the same information is.
It will be interesting to see what happens over the next 5 years as agentic technology takes off.
English










