bearishcat

@f_wintersberger @grok is this real

While one Fintwit account says Iran will win, the one below says Iran is toast already, but below is another one that lays out why Iran‘s military tactics will bleed out the US. But then again, below is one that says the US already won and I have no clue what‘s true, why these people should know and why they should know. Peak X, I guess.

@ragipsoylu Its like giving an umbrella when it doesn’t rain, only to take it when it does.

Exciting news: The European Investment Bank is set to formally return to Turkey on Friday by signing a statement on intent on a combined €200 million ($236 million) of loans for renewable-energy projects with Turkish Finance Ministry - Bloomberg

@WallStreet_Wire @MenchOsint Someone can’t stand the truth like a colonizer, oh wait 😆 you’re in his lands and mock him for being elsewhere knowing you caused it lmao

@MenchOsint Someone is obsessed, as an Arab living in Europe

"Trump’s newly named Fed Chair Kevin Warsh is listed in the Epstein files. He is the son-in-law of Ronald Lauder, the president of the World Jewish Congress" This is the people who seek to acquire Greenland through Trump.

@AndreasSteno @grok explain 😆

I love how everyone asks Grok to validate. I mean jesus christ.. didn't you get a word of what I wrote? :)

There are a lot of otherwise renowned pundits on X claiming that Trump took Venezuela because of deposits of cobalt, antimony, and rare earths, among other things, not because of oil. My impression is that this is a large language model hallucination that is now circulating as evidence, but it is utter bollocks. Sure, there are some “known” deposits, mainly because Chavez (and, to some extent, Maduro) have claimed so. But take a look at any credible mining database: there is no cobalt of any meaningful size in Venezuela. There is no antimony of any meaningful size in Venezuela. It may be that deposits can be discovered, but at this juncture it is a wildly speculative claim, and there is nothing backing it up. Sad, but it goes to show that an LLM is not yet bulletproof as a data researcher. It looks in words, which is a constraint for this kind of task. Venezuela is (at least for now) not a mining country for a lot of tech/defence related metals. That is why Greenland is next on the list for the Trump administration, because there we have known deposits of stuff that is needed.

@VolSignals It is insane how surpressed VIX is with this much commotion surrounding macro factors

I've seen 6800 as a critical point through the positioning for a while but I do think it was tested and left behind on Friday. My "GUN TO HEAD" guesses: WEDS VIX: 14.6 DEC AM EXPIRY: 6940 DEC PM EXPIRY: 6870 NFA. PROBABLY WRONG. There's something strange going on in the VIX / SPX cross-current and I'm starting to pick up on what it may be. There's a lot of macro risk to chew through this week, so EVERY. SINGLE. THING. is *more* conditional than usual. GL trading this week 🍻

@AuclairsDad That’s it, I’ve followed you for a long time and you became increasly more islamophobic. As a Muslim I cannot accept this kind of language anymore. You, out of all people, should know not to overgeneralize. Unfollowed.

@shortbelly @VolSignals Still can’t wrap my head around how muted it is

@VolSignals It barely moves! Printing a mere 17.2

"14.5 - 16 range" Is this the answer to the test- or do we still have to explore 700bps of VIX upside?

thank you for coming to my ted talk.

🧵1/19 Just dropped the most important thread of Q4. You’re not ready for this level of alpha 🔥🚨

@AuclairCapital I missed that one

@AuclairCapital Maybe I’m missing something; - job openings are up slightly - initial jobless claims are lower - ADP non-farm change is worse (going against my point) Would these development warrant another cut?

@AuclairCapital So if employment data isn’t worse and is marginally better-ish, inflation isn’t rising, I can’t grasp why the FED would cut rates?

@PharmD_KS So if employment data isn’t worse and is marginally better-ish, inflation isn’t rising, I can’t grasp why the FED would cut rates.

639 NQ for the stick save, so far. It’s a long night, and the night is dark, and full of terrors 😬 This really could go either way and 662-667 range may decide things later on. HAGN.

Holy cow the yen is in trouble

@VolSignals LLMs are summer interns in fancy suits

@TailThatWagsDog The issue is that it will give an answer based on what it has read before in its training dataset, and then it will just give an answer. It won’t go back and check if each item in that list has duplicates. Just like a summer intern. Any answer is better than no answer

@TailThatWagsDog You might want to try this experiment; provide an LLM a list of ~450 distinct strings with a space in between each like “stringone stringtwo” etc. Then ask it to conver to a list like [“stringone”, “stringtwo”, …]. Then ask it if it has duplicates. Chances are it says yes

Warning: Long Read ... but worth it if you're interested in understanding why I'm stressing the human-AI hybrid concept ... as a collaboration ... wherein each participant (the human and the AI) need to be functioning at the top of their game ... cross-checking one another for errors. To give you an example of how you really have to understand options and have to challenge the AI (Claude) when its conclusion doesn't look right ... here is what Claude told me when my databased expert system disagreed with his analysis. I'm wanting to examine net put delta in delta "buckets" (ranges). I want to look at deltas in the range where tail risk is a priority. Claude initially saw positioning in this range as overwhelmingly bearish. I disagreed. I saw nuance. So we had to walk through the difference between Notional Net Put Delta and Premium ... stepping through with a single strike ... checking each others' work. Finally, Claude agreed with me ... quantitatively and in the data's implications. _________________________ At one point, when I challenged Claude's conclusion, Claude said: "Wait, let me recalculate this because I think I have the sign convention backwards": Actually, looking at the data: Buyers paid: $651,895 Sellers collected: $760,040 Net Premium = Bought - Sold = $651,895 - $760,040 = -$108,145 This negative value means more selling than buying in terms of premium dollars. Correct Answer: More SELLING than BUYING, which appears BULLISH (reducing put protection), BUT: The Notional NPD of $614K represents the total gross bearish exposure that exists in the market from both sides. The fact that there was net selling (-227 contracts) suggests some traders are closing or reducing bearish hedges, which is a bullish signal. Bottom line: Net selling pressure (bullish signal) but substantial gross bearish exposure still exists in this delta range. _________________________ Sooooo .... you see? 1) You really have to understand your field of interest. 2) If you read generalizations by folks twittering on X about AI's value in options analysis ... be dubious. 3). You have to recognize when the AI is wrong. (OMG) 4) And step through the process ... ensuring that neither you nor the AI is making an error in any of the field calculations. But, in the end ... you end up with a specific prompt that *should* ensure that Claude and you don't make a mistake during the analytical process.

@TailThatWagsDog I wish I could disclose my work with AI but we try to catch financial fraud. A component of it is with open source intelligence

@readbearcat Really? It "... can literally make up numbers and conclusions ..." In my example and in the end, we both agreed. But I had to correct it. Was it making stuff up, initially? Curious, what's your experience with this ... if you're inclined to share?

@TailThatWagsDog When you give an LLM a dataset on positioning, it will know what the sentence format is you would want to read based on literature on positioning. Then it can literally make up numbers and conclusions as long as it fits the format of a sentence, irrespective of the actual data

@TailThatWagsDog Since the output of an LLM is a result of probabilistic calculations on the likelihood of the next word within a sentence, I wouldn’t trust it to “reason”. It doesn’t draw conclusions like we do. Until fairly recently LLMs struggled to find how many r’s were in strawberry