suryap

[Hieu @hieutrades]

Entry Models // Failed auction above VAH

1. Price pushes above yesterday's VAH
2. Buyers lift the offer but there's no follow-through: delta is positive and price stalls
3. A bar closes back inside value
4. Short the retest of VAH from below
5. First target VPOC, second target VAL

Invalid if price accepts back above VAH.

[CV @christianvctrs]

Sebenarnya saya agak bingung aja

- suku bunga global naik setelah 3 tahun tidak ada kenaikan
- suku bunga BOJ juga naik setelah 30 tahun tidak ada kenaikan. Sekali lagi, 30 tahun
- berita perang memanas dengan embel embel WW3
- isu inflasi global di banyak negara

Tapi bitcoin:native terbang sebegitunya

Ya emang sih karena ada faktor liqudity, tapi triggernya itu lo apa

Seolah olah basic teori fundamental dan tekninkal analisis gak ada gunanya lagi gitu

[Quant Science @quantscience_]

Wall Street's most valuable equation in microstructure trading:

θ*(θ* − ϕ) = s²G

The entire edge comes from the option value of patience.

Bookmark this (26 page pdf):

[Menthor Q @MenthorQpro]

Jim Simons explaining alpha.

[Spicy @spicyofc]

how to find trading alpha

↓

[kvro @0xkvro]

this quant paper is f*cking insane

it's basically a full blueprint for separating real alpha from the market noise most traders mistake for signal

the framework is all there: market beta, sector moves, factor exposure, liquidity shocks, volatility regimes, residuals, out-of-sample testing

the same logic I break down in the article below — strip the common drivers, isolate what remains, then stress it under costs and changing regimes

the crazy part is most traders never get past the raw return

they see a stock move, build a story around it, and never ask how much of that move the market already explained

the edge is not the story

the edge is what survives costs, time, and regime change

read the paper first, then read the article

bookmark before this thread gets buried

[Crypto Rover @cryptorover]

BOOM 🚀 Bitcoin Coinbase Premium just flipped GREEN for this first time in over 3 months.

US institutions are spot buying again! 🤯

[Penguin X @ThePenguinBTC]

Bitcoin 82 bin doları neden geçemiyor? Cevap yarın sona erecek 6.4 milyar dolarlık sözleşmede.

Bitcoin bir haftada 62 binden 80 bine çıktı ve 82 bine her yaklaştığında geri döndü.

Bunun sebebi grafikte görünmüyor. Opsiyon piyasasında duruyor.

Anlatıyorum.

Bitcoin şu an iki opsiyon seviyesinin arasında.

Üstte 82 bin dolar.

Büyük Bitcoin sahipleri, ellerindeki Bitcoin'i 82 binden satmayı kabul edip karşılığında prim aldı.

Bu seviyede piyasa yapıcılar, yani dealer'lar, fiyat 82 bine yaklaştıkça riski dengelemek için Bitcoin satıyor. Bu yüzden her yükseliş denemesi aynı yerde kesiliyor.

Altta 75 bin dolar.

Aynı mantık tersine işliyor.

Büyük oyuncular 75 binden Bitcoin almayı kabul edip prim aldı. Dealer'lar fiyat 75 bine yaklaştıkça Bitcoin alıyor. Bu yüzden düşüşler aynı yerde tutunuyor.

Ortada 80 bin dolar.

Opsiyonların en yoğun olduğu seviye.

Fiyat 80 binin üstüne çıkınca dealer'lar Bitcoin satıyor, altına inince alıyor.

Bu yüzden Bitcoin 3 gündür 80 binin etrafında dolaşıyor.

Yarın bu düzen değişiyor.

81.700 Bitcoin opsiyonu sona eriyor. Toplam 6.44 milyar dolar, Deribit'teki açık pozisyonların yaklaşık beşte biri.

Bunların en büyük kısmı 75 bin ve 80 bindeki alma hakları. Yani 80 bini tutan pozisyonların büyük kısmı yarın ortadan kalkıyor.

Piyasa neye hazırlandığını gösteriyor.

Bir hafta önce yatırımcılar düşüşe karşı koruma alıyordu.

Bu hafta yükseliş hakları, düşüş haklarından daha pahalı işlem görmeye başladı.

Somut örneği de var.

4 Eylül vadeli, 82 bin üstü için 2.000 sözleşmelik işlem görüldü.

Fiyat 82 bini geçerse, o seviyede yükseldikçe satış yapan dealer'ların satacak bir şeyi kalmıyor. Direnç ortadan kalkıyor.

Bana göre Bitcoin yarından sonra 82 bini geçecek.

Geçildiğinde bir sonraki seviye 85 bin.

Bu benim görüşüm, yatırım tavsiyesi değil.

Gelişmeleri sizinle paylaşmaya devam edeceğim.

[套利豪仔🗽 @pritipatelor]

又测了一个有意思的量化因子，叫做波动幅度扩散因子（Volatility Range Expansion Factor）（下面会放测试图）

逻辑依旧很直觉：市场的真实波动，并不只是单纯看当日的高低价差，而是要同时考虑「跳空」带来的隐含波动

因此，我们将三个维度纳入考量——当日振幅（HIGH-LOW）、昨收与最高价的距离、以及昨收与最低价的距离，取其中最大值，来刻画当日的真实波动范围

因子公式如下：

-MEAN(MAX(MAX((HIGH-LOW),ABS(DELAY(CLOSE,1)-HIGH)),ABS(DELAY(CLOSE,1)-LOW)),12)

本质上是在捕捉市场在隐含波动扩张下的状态转变。当波动快速放大时，通常伴随着非理性交易与过度反应，这也为后续的价格修复或趋势延续提供了潜在的交易机会

测试及其简单，两行代码跑完整回测，大家可以去测试看看
https://t.co/cvZgUElpJA

[Quant Science @quantscience_]

159 page PDF download.

The best examples of how machine learning is used in finance and algorithmic trading.

Grab the paper here:

[venus @RitOnchain]

Jane Street fired this guy who leaked their method to build regime aware trading agents.

he openly shared this guide on reddit that got 40k upvotes.

Bookmark this before its lost.

[Quant Science @quantscience_]

👉 10 Fastest Growing Python Libraries in Algo Trading

Here's the Full List:

[Tatara @0xTatara]

this FinDPO paper is f*cking insane

instead of training an LLM to simply label financial news as positive or negative, the model is optimized around human preferences for which interpretation is actually more useful

the result: 67% annual returns with a 2.0 Sharpe after 5 bps transaction costs, while improving sentiment accuracy by 11% over supervised baselines

the crazy part is the edge does not come from predicting prices directly

it comes from turning messy financial language into a ranked signal that can actually be traded

most finance LLMs learn to classify text

this one learns which interpretation is worth putting capital behind

bookmark before this gets buried

[Hieu @hieutrades]

If you struggle with identifying key levels

Start with value area highs and lows

[Quant Beckman @quantbeckman]

Measure float turnover, breakout strength, and how much post-breakout trading occurs while price remains above VWAP. If turnover is extreme but VWAP retention and marginal price progress deteriorate, classify the move as probable distribution and reject the breakout.

[Chart Champions @ChartChampions]

You asked. We delivered.

Our official Chart Champions TradingView channel is now LIVE

And we’re kicking things off with our first free indicator: Single Prints by Chart Champions

Automatically mark Single Prints on your chart the same way we identify them, with plenty more indicators still to come.

Follow us on TradingView and stay tuned. https://t.co/D7l2gFhEyB

[kvro @0xkvro]

this quant paper is f*cking insane

it shows why most traders overestimate how many real bets they are actually making

30 different tickers can still be the same trade wearing different clothes

the better approach is:

map the common forces
hedge what you do not want
study what survives
only trust signals that keep working under pressure

the crazy part is where most strategies really die

not on normal days
but during liquidity holes, correlation spikes, and regime shifts

a calm backtest can hide massive tail risk
a “diversified” book can still collapse together
and one lucky period can make a weak system look elite

most traders try to make the model look smarter

real quants build a pipeline designed to reject bad ideas

because the goal is not to protect the strategy

the goal is to stop fake alpha from ever touching capital

bookmark this before you trust your next signal

[Spicy @spicyofc]

Traders can "iceberg" their orders so that they can enter/exit larger positions in a less obvious way

visualized ↓

[kvro @0xkvro]

this paper is f*cking crazy

it argues that most traders are not really trading signals at all

they are trading stories layered on top of shared market movement

the framework is brutally simple:

identify the force behind the move
compress the noise into a few common drivers
hunt the leftover
try to kill the idea before capital ever sees it

the crazy part is how often “skill” is just randomness wearing a good costume

ten trades tell you almost nothing
a green month can be noise with good marketing
and a smooth strategy can simply be short disaster

most traders ask how good the backtest looks

real quants ask how hard the strategy is to break

the edge is not the model

the edge is the rejection machine around it

bookmark this before you trust another pretty equity curve

[suryap @ibgsuryap]

@draftanakunpad5 Halo kak saya mau cek turnitin