xFat duck

Quant Finance | Markets | Economics

[xFat duck @dallamicoh]

@DaniilBuilds thank you very much bro, I appreciate it

[xFat duck @dallamicoh]

https://t.co/EyJP3hBOct

[xFat duck @dallamicoh]

@caderoas anyone can code the math for free in Python, but Bloomberg sells what simpler tools can't - a monopoly on clean, real-time institutional data

[xFat duck @dallamicoh]

A Bloomberg Terminal costs about $24,000 a year. One of the first things it does for fixed-income traders is build a yield curve - using essentially the same mathematical idea a Morgan Stanley swaps strategist explained for free on an MIT whiteboard in 2013. So what are you actually paying for?

it's lecture 10 of the same course, guest-taught by Ivan Masyukov, then Morgan Stanley's global head of interest rate swaps strategies, with a PhD in applied physics from the Moscow Institute of Physics and Technology.

the problem sounds almost too basic: you have bond and swap prices at a handful of maturities, and you need a smooth curve of interest rates for every maturity in between, because that curve is what prices everything else - bonds, swaps, loans, anything tied to rates. fit a curve too loosely to those few data points and small errors in the market data turn into wild, unstable swings in maturities you didn't even have data for. Masyukov's lecture is about fixing that: fitting the curve with cubic splines, then adding a penalty - a regularizer - that forces the curve to stay smooth instead of chasing every wiggle in noisy prices.

that same idea, penalizing a model for being needlessly jagged so it doesn't overreact to noise, sits underneath a lot of modern statistics and machine learning too, not just bond pricing. Masyukov happens to be explaining the fixed-income version of it, for free, on a whiteboard.

he's not selling anything either - the lecture opens with a disclaimer that his comments are his own, not official Morgan Stanley research.

a friend who builds curve models for a living says this is still one of the clearest explanations of why a "reasonable-looking" curve isn't automatically a good one - you can fit the data perfectly and still get a curve that's financially nonsense between the points you actually measured.

every terminal, every pricing system, every bank's fixed-income desk is running some version of the same regularized curve-fitting Masyukov gave away for nothing. you're not paying $24,000 a year for the idea. you're paying for it to update in real time.

[Shevaxgod @shevaxgod]

Ed Thorp, MIT mathematician, 1962:

he published a book proving you could beat blackjack, and gave the entire system away.

it became a bestseller.

las vegas responded by changing the rules of the game. customers refused to play the new version, so the casinos put the old rules back and switched to four deck shoes instead.

so he wrote a second edition, with a new counting system built specifically to beat the shoes.

that is the only case i know of where one book forced an entire industry to redesign its product. twice.

the year before, he and claude shannon had built the first wearable computer. the size of a pack of cards, worn at the waist, switches hidden in a shoe, output through an earpiece. it gave a 44% edge on roulette.

then he left all of it and started a fund.

princeton newport, 1969 to 1988. $1.4 million to $273 million. 18.2% a year after fees for the limited partners, no losing year, not one losing quarter.

here is the part worth keeping.

he gave the blackjack edge away because it could be given away. an edge a book can kill was never an edge. it was a temporary mistake someone else was making.

the one he kept was the one that survived being explained.

bookmark and watch the full conversation ↓

[xFat duck @dallamicoh]

@Neighboritto yes in fixed income, the speed of clean data delivery equals survival

[xFat duck @dallamicoh]

@Neighboritto this is exactly why I love the internet

[xFat duck @dallamicoh]

@sophiec87_x yep yep why not

[xFat duck @dallamicoh]

MSCI makes real institutional money selling a risk model built on an idea a Berkeley professor gave the world for free in 1975. The same core method has been sitting in a free MIT lecture since 2013. So what are institutions actually paying for?

it's lecture 15 of the same course, taught by Peter Kempthorne, an actual MIT statistician, not an industry guest this time. no company logo on the slides, no product name attached.

the idea is factor modeling: instead of tracking every stock's risk on its own, you explain its returns as a mix of a handful of shared factors - things like the overall market, an industry, a size or value tilt - plus whatever's left over that's specific to that one stock. get the factors right and you can measure and manage the risk of an entire portfolio using a fraction of the numbers you'd otherwise need.

Barr Rosenberg built the first real commercial version of this in 1975, calling it Barra. MSCI bought Barra in 2004 and still sells factor risk models built on the same fundamental logic Kempthorne walks through on video - it's now one of the more reliably profitable pieces of MSCI's business, licensed to funds that would rather pay than derive it themselves.

Kempthorne isn't selling a subscription. he's a statistics lecturer showing undergraduates the same linear algebra that ended up inside a piece of enterprise software.

here's the part licensing the software doesn't fix, though: a factor model is only as good as the factors somebody thought to include. whatever risk isn't explained by your chosen factors gets treated as noise specific to that one asset - right up until it turns out a bunch of "idiosyncratic" mortgage risk across different assets was actually the same factor in disguise, which is a big part of why 2008 caught so many risk models flat-footed.

a friend who builds risk models for a living says the lecture is a better use of forty minutes than most vendor training sessions, because Kempthorne actually shows you the assumption you're making when you pick your factors - something the vendor documentation tends to skip.

the model inside the product you license traces back to a professor's idea from 1975 and a free lecture from 2013. you're not buying the math. you're buying someone else's factor list and hoping it's the right one.

[Lumen @lumenxbt]

https://t.co/3hWKMIBJcw

[xFat duck @dallamicoh]

@mmoklaaa data without context is just noise, and the ability to separate the signal from the noise is the ultimate skill

[xFat duck @dallamicoh]

@bubosees thank you for taking the time to do this, thank you for reading and watching everything, I'm glad you like it

[xFat duck @dallamicoh]

@shevaxgod thank you I appreciate it

[xFat duck @dallamicoh]

@Neighboritto I'm glad you liked it
thanks for the comment

[xFat duck @dallamicoh]

@finillin_dev you're not paying for the engine; you're paying for the map. The scary part is when the map leaves out a cliff (like in 2008), everyone drives off it together because they all bought the exact same map

[xFat duck @dallamicoh]

@Bureauaiagent Definitely worth checking out that MIT lecture

[xFat duck @dallamicoh]

@DaniilBuilds the math is free, but clean data and the right set of factors cost millions

[Gorilla Ihor🦍⚡️🏛️ @IgorPokotilo]

Today I picked up the keyboard I won in the @coinmarketcap World Cup giveaway.

Thank you so much for the awesome giveaway.

As promised, here are some photos of the keyboard😎

it looks super cool⚡️

thanks again!🫶

[xFat duck @dallamicoh]

"Three years ago, I was earning $16.00 an hour wheeling hospital gurneys down fluorescent basements at 4 AM. Last month, my two-person crew transformed four concrete garages and cleared $13,500 in net profit."

For six exhausting years, Tom worked the graveyard shift as a hospital patient transporter. He spent twelve hours a night lifting heavy equipment, dealing with high-stress medical emergencies, and walking twelve miles per shift across cold linoleum. After taxes, his paycheck barely hit $2,350 a month, leaving him constantly fatigued, deeply stressed, and unable to spend meaningful time with his young family.

The turning point happened when Tom helped a contractor coat his uncle's stained concrete garage floor. Watching the crew transform a pitted, oil-soaked slab into a mirror-like polyaspartic showroom floor in just 36 hours for $3,800, he realized the massive demand for residential surface restoration. Tom spent three weeks researching surface preparation, diamond grinding techniques, moisture testing, and epoxy flake application. He invested $950 of his savings into renting a commercial floor grinder, a heavy-duty dust extractor, and initial resin materials to complete his first demo job on a friend's double garage.

After posting the stunning transformation video online, Tom booked three local homeowners within his first week, earning $2,400 in clean profit from just three days of focused labor. Today, Tom has permanently left the hospital corridors behind to run a high-demand garage floor coating business. Completing just five to six residential garage transformations a month, he consistently generates over $12,000 in net monthly profit while working less than twenty-five daytime hours a week on his own terms.

[xFat duck @dallamicoh]

For five years, Elena was chained to a desk as a bank teller, earning $16.50 an hour while handling millions of dollars for a major corporation. She suffered from daily migraine headaches, toxic micro-management, and rigid schedules that left her completely burnt out, bringing home barely $2,400 a month after taxes.

Everything changed when she paid $250 to have her own dirt-caked SUV professionally detailed in her driveway, watching the technician finish the entire job in just 90 minutes. Realizing that mobile car detailing requires minimal overhead and zero shop rental, Elena spent three weeks studying interior extraction methods, leather restoration, and paint correction techniques online. She invested $600 of her savings into a portable shop vacuum, a dual-action polisher, specialized chemicals, and microfiber towels, launching her mobile service by offering detailing demos to coworkers in her bank’s parking lot.

The pristine before-and-after photos generated immediate buzz, landing her 8 paid bookings in her first fortnight and proving the business model worked. Today, Elena has completely abandoned her bank teller chair to run a flourishing mobile detailing business. She serves 20 to 25 clients monthly, charging $200 to $500 per vehicle to consistently generate over $8,800 in net profit each month. She traded 40 hours of fluorescent-lit corporate stress for just 20 flexible hours a week working outdoors on her own schedule.

[xFat duck @dallamicoh]

For five years, Elena was chained to a desk as a bank teller, earning $16.50 an hour while handling millions of dollars for a major corporation. She suffered from daily migraine headaches, toxic micro-management, and rigid schedules that left her completely burnt out, bringing home barely $2,400 a month after taxes.

Everything changed when she paid $250 to have her own dirt-caked SUV professionally detailed in her driveway, watching the technician finish the entire job in just 90 minutes. Realizing that mobile car detailing requires minimal overhead and zero shop rental, Elena spent three weeks studying interior extraction methods, leather restoration, and paint correction techniques online. She invested $600 of her savings into a portable shop vacuum, a dual-action polisher, specialized chemicals, and microfiber towels, launching her mobile service by offering detailing demos to coworkers in her bank’s parking lot.

The pristine before-and-after photos generated immediate buzz, landing her 8 paid bookings in her first fortnight and proving the business model worked. Today, Elena has completely abandoned her bank teller chair to run a flourishing mobile detailing business. She serves 20 to 25 clients monthly, charging $200 to $500 per vehicle to consistently generate over $8,800 in net profit each month. She traded 40 hours of fluorescent-lit corporate stress for just 20 flexible hours a week working outdoors on her own schedule.

[xFat duck @dallamicoh]

$14.80 an hour was the absolute ceiling for Jason, who spent 45 grueling hours a week as a middle school night janitor. He spent long evenings stripping floors, scrubbing hallways, and dragging heavy trash cans, bringing home barely $2,200 a month after taxes while dealing with severe lower back pain and zero weekend time with his family.

The turning point came when a neighbor asked if he could deep-clean a badly stained sectional sofa using his commercial cleaning experience, offering $120 cash for a two-hour job. Realizing that homeowners constantly deal with pet stains, spilled drinks, and matted carpets but hate paying high franchise fees, Jason spent 2 weeks studying fabric types, extraction techniques, and stain removal solutions. He invested $550 of his savings into a portable extractor, drill brushes, and professional cleaning solutions, then posted 6 before-and-after photos of cleaned furniture on local Facebook community groups.

Within 48 hours, he booked 4 upholstery cleaning jobs over a single weekend, earning $520 in 7 hours of work. Fast forward 7 months, and Jason has permanently turned in his school janitor keys. Today, he runs a specialized carpet and upholstery cleaning business servicing 25 to 30 residential clients and small offices a month, charging $250 to $600 per home or sofa restoration to consistently clear over $9,400 monthly in net profit. He traded 45 hours of late-night janitorial shifts for 20 flexible daytime hours a week working on his own terms.

[xFat duck @dallamicoh]

$14.80 an hour was the absolute ceiling for Jason, who spent 45 grueling hours a week as a middle school night janitor. He spent long evenings stripping floors, scrubbing hallways, and dragging heavy trash cans, bringing home barely $2,200 a month after taxes while dealing with severe lower back pain and zero weekend time with his family.

The turning point came when a neighbor asked if he could deep-clean a badly stained sectional sofa using his commercial cleaning experience, offering $120 cash for a two-hour job. Realizing that homeowners constantly deal with pet stains, spilled drinks, and matted carpets but hate paying high franchise fees, Jason spent 2 weeks studying fabric types, extraction techniques, and stain removal solutions. He invested $550 of his savings into a portable extractor, drill brushes, and professional cleaning solutions, then posted 6 before-and-after photos of cleaned furniture on local Facebook community groups.

Within 48 hours, he booked 4 upholstery cleaning jobs over a single weekend, earning $520 in 7 hours of work. Fast forward 7 months, and Jason has permanently turned in his school janitor keys. Today, he runs a specialized carpet and upholstery cleaning business servicing 25 to 30 residential clients and small offices a month, charging $250 to $600 per home or sofa restoration to consistently clear over $9,400 monthly in net profit. He traded 45 hours of late-night janitorial shifts for 20 flexible daytime hours a week working on his own terms.

[xFat duck @dallamicoh]

$15.50 an hour was the brutal ceiling for Rachel, who spent 50 grueling hours a week as an Amazon delivery driver. She was sprinting up apartment stairs, racing against strict delivery timers, and taking home barely $2,450 a month after taxes while dealing with permanent knee pain and constant stress.

The turning point came when a homeowner asked if she knew anyone who could haul away an old couch and broken treadmill sitting on their porch, offering $150 cash for a 20-minute task. Realizing that homeowners, estate executors, and landlords constantly need bulky waste cleared out and have no way to transport it themselves, Rachel spent 10 days researching local landfill regulations, dump fees, and hauling logistics. She invested $250 of her savings into heavy-duty ratchet straps, tarps, work gloves, and rented a simple $35/day utility trailer attached to her SUV, then posted 5 local junk removal ads on Facebook Marketplace and Nextdoor.

Within 24 hours, a landlord hired her to clear out a basement, paying $380 for a job that took 2.5 hours and cost $60 in dump fees. Fast forward 6 months, and Rachel has permanently handed back her delivery van keys. Today, she runs a thriving independent hauling and junk removal business completing 18 to 22 cleanouts a month, charging $300 to $900 per load to consistently clear over $10,200 monthly in profit. She traded 50 hours of high-stress delivery races for 20 flexible hours a week working on her own schedule.

[xFat duck @dallamicoh]

$15.50 an hour was the brutal ceiling for Rachel, who spent 50 grueling hours a week as an Amazon delivery driver. She was sprinting up apartment stairs, racing against strict delivery timers, and taking home barely $2,450 a month after taxes while dealing with permanent knee pain and constant stress.

The turning point came when a homeowner asked if she knew anyone who could haul away an old couch and broken treadmill sitting on their porch, offering $150 cash for a 20-minute task. Realizing that homeowners, estate executors, and landlords constantly need bulky waste cleared out and have no way to transport it themselves, Rachel spent 10 days researching local landfill regulations, dump fees, and hauling logistics. She invested $250 of her savings into heavy-duty ratchet straps, tarps, work gloves, and rented a simple $35/day utility trailer attached to her SUV, then posted 5 local junk removal ads on Facebook Marketplace and Nextdoor.

Within 24 hours, a landlord hired her to clear out a basement, paying $380 for a job that took 2.5 hours and cost $60 in dump fees. Fast forward 6 months, and Rachel has permanently handed back her delivery van keys. Today, she runs a thriving independent hauling and junk removal business completing 18 to 22 cleanouts a month, charging $300 to $900 per load to consistently clear over $10,200 monthly in profit. She traded 50 hours of high-stress delivery races for 20 flexible hours a week working on her own schedule.

[xFat duck @dallamicoh]

$16.50 an hour was the rigid ceiling for David, who spent 48 grueling hours a week as a warehouse forklift driver in a cold, noisy distribution center. He spent long shifts moving heavy pallets, suffering from chronic lower back pain, and taking home barely $2,400 a month after taxes while feeling completely trapped in a dead-end routine.

The turning point came when his own washing machine broke down, and a local technician charged him $180 for a 20-minute fix that involved simply swapping a $15 thermal fuse. Realizing that modern home appliances break down constantly and that most repairs require straightforward diagnostic logic rather than heavy machinery, David spent 3 weeks studying washer, dryer, and refrigerator error codes, multimeter testing, and component replacements using online repair manuals and guides. He invested $350 of his savings into a basic tool kit-a digital multimeter, nut drivers, pliers, and a cordless drill-and posted a simple local ad offering $50 diagnostic visits for home appliances on Nextdoor.

Within 48 hours, a neighbor hired him to diagnose a dryer that wouldn't heat up. David identified a blown heating element, bought the $25 replacement part at a local supply shop, and fixed it in 45 minutes, charging $190 total. Fast forward 6 months, and David has permanently handed in his warehouse badge. Today, he runs an independent mobile appliance repair service completing 20 to 25 service calls a month, charging $180 to $350 per repair to consistently clear over $9,800 monthly in profit. He traded 48 hours of exhausting, backbreaking warehouse labor for 22 flexible hours a week helping local homeowners on his own terms.