I'm using this thread as a place where we can thoughtdump stuff on interpreting market structure for a better read on what's going on. I am not good at this so could use some help or any input. A lot of fintwit retards are saying the recently SPY rally was 'obvious', and I didn't see it coming in so fast and sharp, so I'm looking into some means of better understanding underlying market structure to make better predictions. Some examples:>inference of dealer positioning (GEX, etc) and understanding of the dynamics that follow from their positioning (eg vanna, charm, etc)>institutional flows (JHEQX, JPM collar, CTAs, corporate buybacks, etc)>expiry dates (OPEX, vixperation, etc)>sec filings >hedging flows, put call parity, etc
>>62136320markets dont like unexpected events so they sell off when they occur. over time markets learn about the situation and often start to recover. see covid, tariffs, and iran as recent examples.
market crabbing = house taking all bets in waiting for imbalance to occurmarket not crabbing = imbalance occurringsimple as>vanna, charmnot simple asnow why that imbalance (or no imbalance) is occurring and who the players are causing said balance/imbalance is a different conversation... ... ... (niggers & jews)
>>62136340yes. but this is due to exogenous events, what I'm looking to better interpret is endogenous market structure. >>62136346How are you observing said imbalance? Are you reading L2 order book data? What are you intending to imply by your comment on vanna and charm?
>>62136398>What are you intending to imply by your comment on vanna and charm?i dont make trading decisions based on any greeks other than delta. especially not the second order derivatives. >How are you observing said imbalance?i just "feel it in my bones" ya know?
>>62136430>i dont make trading decisions based on any greeks other than delta. especially not the second order derivatives.Why do you not see any value in gamma / gex for example to infer dealer positioning at given price levels?>i just "feel it in my bones" ya know?no, I dont' know
>>62136320No hate, but are you the same guy from this>>62130967thread?
Random walk with positive drift
>>62136544indeed I am>>62136555yeah but the walk is influenced by positioning and flow of capital, which is what I'm trying to better codify here
>>62136346jesse what the fuck are you talking about>>62136320just pay attention to what the FED and institutional banks are doing, that's usually all you need to do
>>62136398>i am not good at this so could use some help or any input. A lot of fintwit retards are saying the recently SPY rally was 'obvious'>markets dont like unexpected events so they sell off when they occur. over time markets learn about the situation and often start to recover. see covid, tariffs, and iran as recent examples.>what I'm looking to better interpret is endogenous market structure.ok buddy lol. dont say people didnt try to help you.
>>62136570I'm trying to improve my ability to read mechanical market state, like reading what the market is telling me. The FED and banks are exogenous factors that the market reacts to and should be monitored closely of course but are outside the endogenous mechanics of the market.The FED might make a decision that causes a reposition of the major players in the market, and if you can read the market well you can infer what it expects now, for example, and that can indicate volatility and direction.
>>62136556My assumptions on you based on what you wrote:>pretty high IQ>good at what he does, decent trader>turbo autism>virtually zero ability to understand what motivates people>likes numbers/data>tunnel vision, overfocusing>no idea what is going on in the worldHow close am I?Personally I would love to hear what your guess is on why this war is happening.
>>62136592i seewell good luck
>>62136588you're not helping though. everyone knows the market derisks under increasing uncertainty and violently sells off during 'black swans'. this is common knowledge and doesn't relate to what's being asked in the OP.You're trying to point out that once we had more information regarding the outcome of the situation in the Middle East, the market rebounded due to the increasing certainty that came from taking wild scenarios (WW3, etc) off the table. That was expected. What was not expected was the slingshot to $710, and this largely a mechanical move due to the market being overhedged (short squeeze) and a lot of gamma around $685 due to the JPM collar. That's my opinion after the fact anyway, I was not plugged in enough during the rally to see the mechanics unwind and that was a mistake I'm trying to fix.
>>62136601>virtually zero ability to understand what motivates people>tunnel vision, overfocusingWhat do you mean here? Are there things I'm not factoring in that I should be? >How close am I?I do a lot better trading purely quantitatively / algorithmically. That's usually what I do, but it's slow to build out infra so I don't have all the strucutural/mechanical/microstructure stuff I'd like yet and the market lately has been dominated by geopolitics and Truth social post that aren't as friendly to that kind of stuff. The people that *did* have the ability to read the mechanical state of the market during the rally were seemingly in a pretty good spot to catch a once in a lifetime move, so the opportunity cost of not being setup to catch those moves or at least hedge against explosive upside was huge during this, so I'm prioritizing it a lot more. I think a lot of it is probably very nuanced too and it'd take a fuckton of historic data to backtest everything required (as in, beyond just market data but including JPM collars, put call parity, institutional inflows, etc - it'd be a pretty extreme task and could take years to get to the point), so I'm trying to learn to do subjective, interpretive market reads to be a better discretionary trader or influence my automated system using some discretionary influence.If you could've read the market well, you could've bought a far OTM call going into the 4/17 OPEX, during the first week of April and it would've absolutely printed. I don't really know what's going on in the war aside from the strait being closed throughout and I doubt Israel is going to let us off the hook so easily, Trump seems to want a graceful exit but idk if Iran or Israel will give him that. But he's typically ok with just lying and forgetting about stuff like that. Mechanically OPEX reset so we're looking at fresh dealer positioning and I'm betting on a move down to ~$695-700.
Learn ICT trading concepts. ICT is the SICP of trading.
>>62136648lol good one retard
>>62136638>What do you mean here? Are there things I'm not factoring in that I should be?I very much like the idea of automation so I'm rooting for you but your desire to separate the data/numbers from the external world is going to miss things in times like these. You're not incapable of incorporating outside factors into an algorithm, you might want to make a system that helps you track what companies are pro or anti this or that, but sometimes people intentionally lie for ulterior motives.Maybe just design the system so it is able to identify pattern-breaking moments like these. Then label it an "anomaly" even though in reality the concept of anomaly basically just means "something I don't understand."On most days I don't believe in free will so I believe the majority of people not only don't understand what's happening, even if you told them they would just reject it and it would actually be impossible for them to accept it.
>>62136729>You're not incapable of incorporating outside factors into an algorithmIt's a long term goal of mine, it just introduces complexity. I don't even know what backtesting that would really even look like either or how you'd design a learner over that. What I can do though is incorporate the underlying mechanical forces that result from these real world events, and feed that to the learner, so I'm trying to focus that down. I think that's more fundemental and generalizable too. Where I think solid edge would lie for real world event interpretation would be more in live dialogue during FED meetings or earnings, since you could get in before the market fully restructures. I just don't know how you'd design a learner that could navigate something novel like the Iran war unfolding combined with a president actively contradicting things. It seems like it'd be easier to just design something that follows the trend soon after he posts and hope it's correct most of the time.My automated system actually performed decently enough through this whole thing, but it mostly does intraday only, so it wasn't really capable of capturing that longer term trend which would've been extremely profitable. My discretionary trading was bad though and that's what I need to iron out first. This rally minted many new millionaires and I wasn't one of them, but I feel if I had both more skill in interpreting the mechanical factors driving the market as well as the infra to display it, I could've done significantly better. Allotting even $2k to hedge insane upside early April on SPY $700 4/17 calls would've been an insanely profitable and reasonable play if you were aware of the squeeze potential and gamma positioning around $685/700 and the heightened volatility that goes into the monthly OPEX, imo. It's something that crossed my mind repeatedly, I just wasn't organized enough mentally or workflow wise so I'm spending this week building out some stuff to help there.
