Vashistha Iyer @uptickr

An average active mutual fund beats its benchmark. But the one in your portfolio may not. Since 2013, the average active large-cap fund barely beat the index while small caps beat the benchmark with 4% higher CAGR and ~15% lower drawdown too. But here's the catch. You can't buy "the average fund."

x.com/i/article/2075…

A stock drops 20% on a rumour and is back to even within a year. Another drops 80% on confirmed fraud and never recovers. We pulled 20 years of NSE data and ran 22 governance events through the same framework—all the way from Satyam in 2009 to HDFC earlier this year. Our conclusion? Headline severity is a bad guide. The type of bad news predicts recovery far better than how big the initial drop looks. What matters is what kind of governance news you're looking at, and whether the business will still make money if the worst of it turns out to be true. We built this into a five-rule framework for the next governance scare that lands in your portfolio. Read about it here: capitalmind.in/insights/not-a…

Bessembinder new paper... Best way to improve returns of the market? Avoid the big stocks. Also, equal weight over market cap weight. Take for a spin! papers.ssrn.com/sol3/papers.cf…

keyboard_arrow_left Previous keyboard_arrow_right Next

Macros by Capitalmind - July 2026 Most investors watch markets, but informed investors watch what moves the markets. Every month, we track such macro signals (thread)

x.com/i/article/2065…

"Free the rupee. Let it go. It'll come back" An economist (@anupammanur) and a market investor (@deepakshenoy) come together to propose solutions to break out of the current SIP funding FII exit loop. It's a fun discussion, with some provocative points.

This, by @hadleywickham , is by far the clearest explanation of agents and harnesses that I've read so far. Zero jargon. Zero complications. tidydesign.substack.com/p/what-is-an-a…

x.com/i/article/2064…

Red Bull's F1 team ran 8 billion simulations before the crucial 2025 Qatar GP. Not to predict the race. But, to have a response ready for whatever showed up on track. Same works for investing too. @divyanshagnani writes capitalmind.in/insights/eight…

You can now lock your mutual fund folios. Switch on Folio Lock on MF Central and no units leave until you unlock it. Here's a step-by-step video showing how it works.

Most NRIs investing in India ask "which fund should I buy?" before they've asked where they're tax resident. That sequencing error causes real, expensive tax problems. We've looked at the major things NRIs and OCIs don't think about before investing in India: tax residency (and why your Indian passport has nothing to do with it), which investment instrument is most tax efficient as per your current tax-residency jurisdiction, and what getting that wrong actually costs in practice. The structure around your Indian investment often matters more than the investment itself. If you have Indian money to invest, or have Indian money already invested, this piece by @avijeet_sen worth a read. capitalmind.in/insights/nri-i…

I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem. As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)! I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work. It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results? 88ms => 1.5ms 150K allocs => ~500 allocs Incredible right? Nope. My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path. This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput. The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity. Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.

Why you shouldn't write like AI Because it's shit writing. Like absolutely bland. It's not this, it's that. Question mark? No, this is the answer. It's like eating, I don't know, sawdust. My apologies to sawdust. It's late at night and essays that I'm reading have the exact same crappy footprint. I've learnt to live with the long hyphen. It's garish. It's out there telling you that hello, uncle, this is AI, but you know what, it might be interesting anyhow. And I've tried to give it a shot, but somewhere along the line it's like the whole paragraph has been attacked by Dementors and they took all the soul from the written words and it's left with these alphabets that juxtapose just enough to make grammatical sense. Yes, your actual writing might be terrible. Or you don't have the patience to type endlessly into the ether, like I do, or like I love to do. I just love to write, so maybe I'm possessed with this demon that every piece of writing needs to have a soul. Maybe it's my unnecessary adulation, no, love, for prose. Too much prose, that too. Or to make that spelling mistak so you just know it's human. Hoo-man. We are all government documents now. Devoid of colour, bereft of whatever bereft means to be bereft of. AI has perhaps made writing easier for a lot of people. And for technical stuff, it's all fine. You're expecting to be bored to the bottom of your stomach, so some outrageous terms, no matter how often they're used, ease the pain a little. But that opinion piece which needs you to make the words be alive, it sucks when AI renders it into this communist level everyone-must-have-the-same-writing blandness. I don't think I want the extra long words - that is just showing off or winning at scrabble - but I do want a little punch in the sentence. The feeling that you said what you actually felt. The occasional slip, the authenticity. Maybe it's just me, but I don't care, this whole tweet is just me anyways. Whatever it is, I swear, I will be nicer when it's got that hoo-man in it. Even to the nasty trolls, who I will mute with a smile because they wrote it themselves. You have to give me one late night rant about nothing. I do spend most of the days telling governments and the RBI what to do, so I'll bloody well rant about what I want to read once in a while.

Has RBI been buying gold aggressively? India's gold reserves just hit a 25-year high at 17.3% of total forex reserves. Of the $72bn rise in gold's reserve value since end-2023, 93% came from price appreciation and only 7% from buying more gold. RBI bought only 73 tones in 2024, 4 in 2025, and nothing in 2026.

Watch @CalmInvestor talk about: - How to build winning portfolios - The million-dollar question: where is the "value"? On Air at @CNBCTV18Live Watch: youtube.com/watch?v=WQ2TBH… Mutual Fund investments are subject to market risks, read all scheme related documents carefully.

Watch @CalmInvestor on @CNBCTV18News talk about the markets, our funds and more: - Earnings of the large/mid/smallcaps - Cynefin 4 situation framework - How the @CapitamindMF Flexi Cap fund managed cash - The Multi-asset fund that goes beyond Gold and Silver - And more! youtube.com/watch?v=WQ2TBH… Mutual Funds are subject to market risks, read all scheme related documents carefully.

ESOPs are compensation. Hence, it will be taxed as income. That's not the problem. The timing i.e. tax liability arising upon exercise, creates a cashflow problem. A better way would be to shift that liability to the point of sale of the converted shares.

Alan @cbe_sam

2 months ago

India taxes ESOPs twice... I've observed this in 2 of my previous companies. Will be paying more tax when the listing happens and will sell it. I'm all in for paying tax, but this arrogant tax policy should end.

1 5 32 6K 4

Capitalmind Mutual Fund is now the first AMC to add "Portfolio Changes" summary in their factsheet. Earlier, I used to manually compare two months portfolio & share the changes. Now AMC itself is doing it. Great work @deepakshenoy @CalmInvestor @uptickr & team. 👏

More FMs should be doing this; the transparency and clarity regarding the Portfolio changes are a good addition, and more AMCs should follow Kudos to the team! @deepakshenoy @CalmInvestor

Deepak Shenoy @deepakshenoy

2 months ago

In the @CapitalmindMF Flexi Cap Fund last month, I'll post the factsheet notes on the summary of what we own and how it is: We have increased allocations over the month of April, from the 66% in March to around 85% in April:

13 12 182 37K 55