@cannibalstocks Nice.. You need to look at Risk free rates in Kazak.. I think it is north of 15%.. just a small detail that makes big difference in evaluating potential.
English
Yash Gurjar
286 posts
Yash Gurjar
@ys_gurjar
Aspiring Swing trader. Here to document the journey and connect. Not wise enough to advise.
Denmark Katılım Haziran 2014
254 Takip Edilen25 Takipçiler
The Peer Multiple Massacre
Screenshot this. Tape it to your monitor.
MercadoLibre. Latin America super app. 45 times earnings. $MELI
Sea Limited. Southeast Asia super app. 35 times earnings. $SE
Adyen. European payments machine. Around 30 times earnings. $adyey
Kaspi. Higher margins. Higher ROE than all of them. Stronger moat than all of them.
The dot at the end of the URL is the only reason you can still buy this at 7 times earnings. $KSPI @MohnishPabrai
cannibalstocks.com/p/while-everyo…
English
我的trading系统压成四句话:
1. 只看leader。RS<90的票连看都不看。
如果你跑不赢90%的股票，不好意思，
你不配在我的watchlist里
2. 只在pivot买。base收敛、缩量、tight,
我喜欢在这里买，因为我可以止损设置紧到不能再紧
我喜欢寻找机构把散户洗掉的痕迹
3. 回踩10EMA加仓。不跌破证明机构在防守。
Thesis confirmed.
4. 跌破50DMA走人。
leader一旦失去50日线的support,故事就结束了
不解释、不期待、不补仓。
赢的时候让仓位帮我赚钱, 输的时候让纪律帮我活下来。
Easy, simple, just like that.
中文
Yash Gurjar retweetledi
$EU +2.19 ✅
Rango alcista 4h confirmado
Entrada en m15 en la apertura de las 3NY por encima de la parte media del rango
#LCTradingModel🍯
Documentado en La Colmena en riguroso directo
Español
@MatiasScalbi Great paper ! Espcially the angle of using market breadth as a barometer !
English
🔥 "Optimizing Portfolio Performance" El paper que demuestra que combinar tres estrategias simples le gana a la mejor estrategia individual
20.20% anual durante 19 años. El SPY mismo período: 8.27%
Daniel Requejo es un ingeniero en sistemas que se enseñó solo a operar el mercado. Agarró tres estrategias clásicas, las combinó en proporciones distintas, y backtesteó todo entre 2005 y 2023
Las tres estrategias por separado:
- Swing Trading: 19.7% anual
- Asset Allocation: 17.8% anual
- Lazy Portfolio: 15.4% anual
Nada del otro mundo
Pero cuando las combinás 50 - 50 con un filtro de riesgo de mercado, los números se ponen feos:
Drawdown máximo de -12.91%. El SPY se comió un -56% en 2008
Sharpe de 1.66 vs 0.54 del SPY
USD 100 en 2005 se convertían en USD 3,234. Los mismos USD 100 en SPY: USD 381. Casi 9 veces más plata, con un drawdown 4 veces menor
Lo más loco: el portfolio óptimo según Markowitz tiene Sharpe 1.78. El Portfolio 33 - 50 del paper llega a 1.70. Prácticamente clavado a la frontera eficiente, pero armado con tres estrategias simples que cualquiera puede replicar
La moraleja rompe el mito del trader que busca toda la vida LA estrategia perfecta. Tres estrategias decentes bien combinadas le sacan ventaja a cualquier sistema individual
Link al paper en el primer comentario
Matias Scalbi@MatiasScalbi
🔥 "The Alchemy of Multibagger Stocks" Paper increíble para entender qué acciones hacen 10x Analizan 464 acciones del NYSE y Nasdaq que se multiplicaron por diez entre 2009 y 2024 y modelan qué factores predicen ese tipo de retorno con econometría seria Lo más interesante es que rompe el mito principal: el crecimiento de ganancias (EPS, ventas, profit) NO predice el retorno futuro, algo que casi todos los inversores tratan como verdad absoluta El driver más fuerte resultó ser el Free Cash Flow Yield. Sumado a small cap, alto book to market, buena rentabilidad y una inversión agresiva pero respaldada con crecimiento de EBITDA Te dejo el paper en el primer comentario
Español
@fedex774 Great thesis overall. I am trying to dig into it since this post was published. I see most of the reports in french. I am wondering how did you pull such a nice and detailed info in English.
English
🧠 Why this is interesting right now:
Stricter environmental regulations → waste heat recovery is becoming mandatory
High energy prices → ENOGIA's ROI for customers is better than ever
Hydrogen + fuel cells boom → their ORC + compressor combo is uniquely positioned
English
$ALENO — ENOGIA is a French micro-cap that converts waste heat into electricity using patented ORC micro-turbines.
Nobody is talking about it.
Yet revenues have doubled in 2 years, the order book is exploding, and the company just turned profitable for the first time.
A thread 🧵
English
Yash Gurjar retweetledi
I want to clear up what I see as an important misunderstanding in the investment community: It has become fashionable to say that portfolio returns are “power law distributed” or that only a few companies drive portfolio returns.
But this is only true in a dollar sense, and even then it is somewhat tautological.
As portfolios compound over decades, the largest winners naturally dominate the ending dollar gains simply because they are sitting on top of all the prior compounding that came before them. The earlier “stepping stone” winners that helped build the capital base become forgotten with time.
In other words, when people look back after 40 years and say “these 3 stocks drove the portfolio,” they are often ignoring that those stocks only became so economically dominant because dozens of prior gains compounded into them beforehand.
What matters more is the underlying distribution of percentage returns and the weighting schema applied across those returns over time.
Percentage returns are not power law distributed. They are closer to a normal distribution with an obvious right-tail skew because losses are bounded at -100%.
The real investing skill is not some romanticized notion of “finding one giant winner,” but consistently generating favorable return distributions and allocating capital effectively across them through time.
Below is a hypothetical example of an active portfolio invested over 40 years.
English
@MatiasScalbi Interested to know what is the setting of ATR they used
English
🔥 "A Century of Profitable Industry Trends" Ganador del Charles H. Dow Award 2025, este paper te muestra cómo ganarle al mercado durante 100 años con una sola idea
Aplican trend following solo long a 48 industrias del mercado USA desde 1926 hasta 2024
Los números son una bestia: 18.2% anual con Sharpe 1.39, casi el doble del buy and hold del mercado (9.7%) y con MENOS volatilidad
Lo más impresionante es la robustez: funcionó en TODAS las décadas desde 1926. Gran Depresión, Segunda Guerra, stagflation, crash del 87, puntocom y 2008. Cada época tuvo industrias ganadoras distintas y la estrategia las capturó
Link al paper en el primer comentario
Matias Scalbi@MatiasScalbi
🔥 "The Alchemy of Multibagger Stocks" Paper increíble para entender qué acciones hacen 10x Analizan 464 acciones del NYSE y Nasdaq que se multiplicaron por diez entre 2009 y 2024 y modelan qué factores predicen ese tipo de retorno con econometría seria Lo más interesante es que rompe el mito principal: el crecimiento de ganancias (EPS, ventas, profit) NO predice el retorno futuro, algo que casi todos los inversores tratan como verdad absoluta El driver más fuerte resultó ser el Free Cash Flow Yield. Sumado a small cap, alto book to market, buena rentabilidad y una inversión agresiva pero respaldada con crecimiento de EBITDA Te dejo el paper en el primer comentario
Español
Yash Gurjar retweetledi
🔥 "Does Trend-Following Still Work on Stocks?" Paper de 2025 buenísimo para entender por qué tu bot necesita aceptar perder y que existen estrategias MUY RENTABLES con bajo win rate
75 años de data (1950-2024), 66.000 trades simulados en TODAS las acciones líquidas USA. Dataset completo, sin trampa
La estrategia es brutalmente simple: comprás cuando una acción hace nuevo máximo histórico, salís con trailing stop ATR. Long-only
Los números rompen mitos: win rate de 43.9% (perdés 56% de las veces), pero menos del 7% de los trades genera TODA la rentabilidad. El secreto NO es ganar mucho, es dejar correr a los pocos ganadores que explotan
Promedio ganador: +1.90R. Promedio perdedor: -0.70R. La asimetría hace toda la diferencia. La mayoría de los traders fallan acá: cortan ganadores temprano y mantienen perdedores esperando recuperarse
Link al paper en el primer comentario
Matias Scalbi@MatiasScalbi
🔥 "A Century of Profitable Industry Trends" Ganador del Charles H. Dow Award 2025, este paper te muestra cómo ganarle al mercado durante 100 años con una sola idea Aplican trend following solo long a 48 industrias del mercado USA desde 1926 hasta 2024 Los números son una bestia: 18.2% anual con Sharpe 1.39, casi el doble del buy and hold del mercado (9.7%) y con MENOS volatilidad Lo más impresionante es la robustez: funcionó en TODAS las décadas desde 1926. Gran Depresión, Segunda Guerra, stagflation, crash del 87, puntocom y 2008. Cada época tuvo industrias ganadoras distintas y la estrategia las capturó Link al paper en el primer comentario
Español
Yash Gurjar retweetledi
A股十大热门股速览（3.20）
1. 中际旭创(300308)：光模块龙头，净流入21.52亿，算力需求放量，机构抢筹
2. 阳光电源(300274)：光伏逆变器龙头，净流入18.74亿，海外订单增长
3. 新易盛(300502)：光模块出口龙头，净流入15.4亿，8.32%涨幅，业绩预期强
4. 宁德时代(300750)：锂电龙头，净流入13.03亿，3.12%涨幅，储能业务爆发
5. 上能电气(300827)：储能逆变器龙头，净流入10.63亿，新能源转型受益
6. 亿纬锂能(300014)：锂电二线龙头，净流入9.09亿，5.58%涨幅，储能动力双线发力
7. 隆基绿能(601012)：光伏硅片龙头，净流入8.85亿，行业景气度回升
8. 美利云(000815)：数据中心+造纸，净流入5亿，10.02%涨停，低位启动获游资爆炒
9. 科大讯飞(002230)：AI应用龙头，星火大模型V5.0落地，教育医疗场景商业化
10. 工业富联(601138)：AI服务器代工龙头，算力硬件核心，成交额居前
核心逻辑：资金聚焦AI算力与新能源两大主线，机构主导大票，游资活跃小票，结构性行情显著。
中文
Yash Gurjar retweetledi
This 30-minute speech by the Head of Anthropic "Coding Agents" researcher will teach you more about vibe coding than 100 paid courses.
Bookmark it & give it 30 minutes today. This video will change the way you use AI forever,
Movez@0xMovez
This weather bot turned $300 → $122K on Polymarket weather markets in 3 months I fully decoded algo and built a self-learning Hermes weather trading agent using weather APIs + Opus 4.7, the bot runs 5-min scans & searches mispricings on Polymarket run your agent in 5 steps: • set up a VPS server on Hetzner - $6 • create a weather API on {visualcrossing} - free • set up Hermes agent using one-liner code - free • connect Telegram bot + Opus 4.7 • send {weather trading logic} from article to agent started my agent 2 days ago with a test sum and already having 40% profit agent already caught 2 traders with +400% ROI on Seoul & Chicago weather markets bot used for logic: @coldmath?via=following" target="_blank" rel="nofollow noopener">polymarket.com/@coldmath?via=…
my bot test wallet: @hermesweather?via=following" target="_blank" rel="nofollow noopener">polymarket.com/@hermesweather… Hermes bot is a self-learning agent so give him enought trades {100+}, to build his own logic. start small English
Yash Gurjar retweetledi
@TheDopeNobody Wow ! I never thought like this. Will do for sure. Many thanks
English
@ys_gurjar Make an honest list of your biggest struggles and tailor a strategy to YOU.
English
Never EVER EVER EVER pay large amounts of money for trading education.
What you think is behind the paywall is not there, I promise you.
There is only ONE profitable strategy in trading which is
Size appropriately
Manage risk
That’s literally it.
English
Yash Gurjar retweetledi
This 1 hour podcast with the Head of Claude Code will teach you more about building software with AI than most Google Engineers learn in 6 months of actually doing it.
Bookmark this & give it 1 hours, no matter what. It’s the most productive start you can give your mounth. Then read the article below.
Hanako@hanakoxbt
English
Yash Gurjar retweetledi
This 2 hour Stanford lecture shows exactly how Stanford trains it's engineers to build AI systems. It's more practical than every Claude tutorial & prompting threads you've seen.
Bookmark & give it 2 hours, no matter what. It'll be the most productive thing you do this weekend.
English
Yash Gurjar retweetledi
🇫🇮Finnish satellite company Iceye revenue:
2024 $120m
2025 $250m
2027 target published today: $1,2 billion (Rheinmetall jv starting etc)
Closest public market peer $PL 🇺🇸 Planet:
2024 $220m
2025 $244m
2026 (est) $297m
2027 (est) $377m
Iceye is profitable, with positive cash flows. PL is heavily unprofitable.
Both sell satellite constellations / image data that are long term strategic assets for governments and companies.
PL faces more competition (more competition in optical sats than SAR sats) and Iceye is heavily skewed on "looser" and more stable government / defence contracts, which I'd guess is the main reason for the profitability difference.
Iceye latest round valued at $2,8b = 2,3 x 2027 target
Planet current valuation $8,5b = 23 x 2027 est
Sure PL likely overvalued currently. But still, thats a 10 x difference. And private markets do like to use public market peers as measuring stick (ofc only in case of discount).
Iceye starting to look cheap if the 2027 target is even at ball bark level. And not just by comparing to PL.
Iceye managed to achieve profitability in this high growth phase of 2025, so could assume at least pretty good economics/margins long term (tho we don't know yet exacts).
$SSIT 🇬🇧 Seraphim holds 39% of NAV in Iceye (and trading currently pretty much exactly at NAV).
SSIT did raise their NAV calculation of Iceye +34% in latest February release "using public‑market comparables", which already factors in some of that discount.
English
Yash Gurjar retweetledi
Guy makes $250,000 per month with a VERY simple app.
Keep. it. simple. stupid!
@Samarsky and I talked for a few hours. Here’s the top 4% of our chat:
> His simple app that generates $3m/year (1:30)
> The ONE thing $0 MRR founders ignore (3:44)
> How to build simple apps in 2026 (7:20)
> The tools that power his $250k MRR app (10:55)
> The steps he would take if he had to start over today (12:08)
English
Yash Gurjar retweetledi
Yash Gurjar retweetledi
Quantization can make an LLM 4x smaller and 2x faster, with barely any quality loss. But what *is* it? @samwhoo crafted a beautiful interactive essay explaining it from first principles, aimed at coders, not mathematicians.
ngrok.com/blog/quantizat…
English
@ProStockAmateur Amazing ! Very rare to come across quality material on non US stocks
English
