frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Dan Wang's Breakneck

https://hollisrobbinsanecdotal.substack.com/p/dan-wangs-breakneck-review
1•HR01•51s ago•0 comments

YouTube is rolling out likeness detection tool to combat deepfakes

https://www.engadget.com/entertainment/youtube/youtube-is-rolling-out-likeness-detection-tool-to-...
1•bobertdowney•1m ago•0 comments

The China Tech Canon

https://asteriskmag.com/issues/12-books/the-china-tech-canon
2•jger15•1m ago•0 comments

The natural clocks that can pinpoint someone's time of death

https://www.bbc.com/future/article/20250926-the-natural-clocks-that-can-pinpoint-the-time-of-death
1•pseudolus•2m ago•0 comments

Tomorrow Corporation Tech Demo [video]

https://www.youtube.com/watch?v=72y2EC5fkcE
1•nmilo•4m ago•0 comments

Why more people are going to gigs, festivals and clubs alone

https://www.theguardian.com/music/2025/oct/21/why-more-people-going-gigs-festivals-clubs-alone
1•pseudolus•6m ago•0 comments

Suzanne Somers' widower is trying to recreate her with AI

https://ew.com/suzanne-somers-widower-is-trying-to-recreate-her-with-ai-11834409
1•geox•8m ago•0 comments

Perplexity at Work: A Guide to Getting More Done [pdf]

https://r2cdn.perplexity.ai/pdf/pplx-at-work.pdf
1•theonionspeaks•9m ago•0 comments

Illinois utility tries using electric school buses for bidirectional charging

https://insideclimatenews.org/news/03102025/illinois-electric-school-bus-vehicle-to-grid-program/
2•PaulHoule•15m ago•0 comments

Chris Paik: A Taxonomy of Innovation – Capacity, Coordination, and Connectivity

https://docs.google.com/document/d/1unzEKQG0MD85JjZCPtjvwtvjZI8Sen8mEp1-Gn3JHDY/edit?tab=t.0
1•jger15•16m ago•0 comments

Show HN: LunaRoute – a high-performance local proxy for AI coding assistants

https://github.com/erans/lunaroute
2•erans•16m ago•0 comments

ShieldAI Launches the X-Bat

https://www.cnbc.com/2025/10/21/exclusive-first-look-at-shield-ais-x-bat-ai-piloted-fighter-drone...
1•robgourley•21m ago•0 comments

No Appointments, No Nurses, No Private Insurance Needed

https://www.theatlantic.com/health/2025/10/medicaid-mango-house/684488/
2•JumpCrisscross•24m ago•0 comments

AI Video Maker – Generate videos from text or images using Sora, Veo, and more

https://apps.apple.com/rs/app/ai-video-generator-aivideo/id6736862905
3•virtualminds•27m ago•2 comments

15M Budget Line That Doesn't Exist – Certificates

https://blog.axonshield.com/the-15-million-budget-line-that-doesnt-exist/
2•dc352•29m ago•0 comments

A billionaire has rebuilt downtown Detroit

https://www.economist.com/united-states/2025/10/15/a-billionaire-has-rebuilt-downtown-detroit
3•yakkomajuri•34m ago•1 comments

Knowledge Transfer from High-Resource to Low-Resource Languages for Code LLMs (2023)

https://arxiv.org/abs/2308.09895
1•peatmoss•34m ago•0 comments

C++26: Printing `std:tuple` with expression templates

https://www.cppstories.com/2025/tuple-iteration-cpp26/
2•ashvardanian•37m ago•0 comments

"Butt breathing" might soon be a real medical treatment

https://arstechnica.com/science/2025/10/butt-breathing-might-soon-be-a-real-medical-treatment/
4•zdw•39m ago•1 comments

The future of Python web services looks GIL-free

https://blog.baro.dev/p/the-future-of-python-web-services-looks-gil-free
1•birdculture•42m ago•0 comments

A Furious Debate over Autism's Causes Leaves Parents Grasping for Answers

https://www.nytimes.com/2025/10/14/health/autism-parents-genetics-vaccines-tylenol.html
1•walterbell•42m ago•0 comments

Active mechanical logic gates – Capstan Lever Logic (CLL) [video]

https://www.youtube.com/watch?v=Lt0alWQzkWY
1•karussell•45m ago•0 comments

Git considers SHA-256, Rust, LLMs, and more

https://lwn.net/SubscriberLink/1042172/869c5b0722123b13/
2•chmaynard•50m ago•2 comments

Experts say the high failure rate in AI adoption isn't a bug, but a feature

https://fortune.com/2025/10/21/ai-adoption-failure-rate-bug-feature-amy-coleman-jessica-wu-karin-...
1•minifyre•50m ago•3 comments

My Data Career Journey So Far

https://datamethods.substack.com/p/my-data-career-journey-so-far
2•zekrom•53m ago•0 comments

It's Not Just You – The iOS Keyboard Is Broken [video]

https://www.youtube.com/watch?v=hksVvXONrIo
7•quarkw•55m ago•3 comments

California cracks down on water theft but spares data centers

https://www.latimes.com/environment/story/2025-10-14/newsom-ai-data-center-water
1•1vuio0pswjnm7•57m ago•0 comments

The Hidden Engineering of Niagara Falls

https://practical.engineering/blog/2025/10/21/the-hidden-engineering-of-niagara-falls
2•chmaynard•57m ago•0 comments

You Probably Don't Need to Switch from Pandas to Polars

https://datamethods.substack.com/p/you-probably-dont-need-to-switch
3•zekrom•58m ago•0 comments

Uber: Rebuilding Uber's Apache Pinot Query Architecture

https://www.uber.com/blog/rebuilding-ubers-apache-pinot-query-architecture/
1•ankitsultana•59m ago•0 comments
Open in hackernews

The Lottery-fication of Everything

https://www.dopaminemarkets.com/p/the-lottery-fication-of-everything
46•_1729•2h ago

Comments

decimalenough•1h ago
> Sportsbook hold has doubled from 6% when parlays were just introduced to over 12% today.

> Options were 26% of Robinhood’s revenue in 2024 with an implied gross margin of over 90%.

Wow. For comparison, slot machine RTP (payout ratio) usually hovers around 91-93%, meaning a "hold" of 7-9%.

dmurray•52m ago
"Gross margin" is not the same as "hold" here.

Options pricing is reasonably competitive. Even a gambly thing like a Tesla zero day option has a spread of 1-2%, so someone trading it at random loses 0.5-1% per trade. And Robinhood is a brokerage, not an options market maker, so it doesn't capture all of that 0.5-1%.

You'd have to read Robinhood's financials to see what they mean by gross margin. Possibly it means if a customer deposits $1,000 and trades options, the customer eventually on average loses $900 of it? Even that seems too much TBH.

verteu•47m ago
Does anyone know why options broker fees are so high? A typical cost is ~$0.25/contract for a $1.00 SPY 0DTE, eg: https://www.interactivebrokers.com/en/pricing/commissions-op...

Does this reflect brokers' cost of technology (many tickers to keep track of)? Regulatory fees? Lack of competition?

edit: Perhaps it's largely profit. Seems https://public.com/invest/options-trading is trying to undercut on price.

boringg•29m ago
Profit and settlement risk
codeulike•1h ago
Zero day options rose from 5% of total options volume in 2016 to 61% by May 2025
Bjartr•1h ago
This article mentions describes financial instruments that were niche a few years ago that now dominate. I wonder what instruments are niche today that might have a similar trajectory.
codeulike•1h ago
Bartering for food
jbjbjbjb•1h ago
> The world is getting stranger

In the U.K. I was betting 5 minute binary options back in 2008 and parlays or accumulators as we call them (accys for short) have been popular for a while too.

giobox•49m ago
Rightly or wrongly, The Gambling Act 2005 put the UK literally decades ahead of places like the US in terms of creating a legal framework for sports betting/gambling in general.
shermantanktop•16m ago
And that's how you get a Paddy Power next to a William Hill while the rest of the high street is shuttered.
nly•4m ago
And Betfred just threatened to close 1,300 high street betting shops if Reeves increases gambling taxes in the budget next month.

https://www.ft.com/content/3641f944-38cc-4ed5-853a-03fab3ac4...

verteu•1h ago
> I had an absolutely disgusting thought today: Robinhood should offer parlays. Sell customers a call option on multiple assets. For example a call option to buy Apple at $250, NVDA at $190, GameStop at $25 and Bitcoin at $120k, but only if ALL of them are in the money. Robinhood could buy offsetting calls on each individual asset, then sell the parlay “bundle” to retail. Risk-free profit for Robinhood, and their customers would love it.

I don't see how this is risk-free - surely it involves some opinion on the correlation between the assets?

lordnacho•54m ago
It requires the correlation if you are going to price it accurately. But with a parlay people just look at the high payout and estimate the chances wrong. By like, a lot.

It's the fact that's easy to sell that makes it attractive, not that it's easy to price.

alasarmas•52m ago
I think you’re correct. There is a saying something like: in a crisis, all correlations go to 1. I believe it’s likely one of those things that’s okay most of the time and then, every once in a while, causes extreme and systemic problems. Therefore, Robinhood will probably do it because the incentives are aligned.
muxl•50m ago
It is truly risk free. You always buy the call using the customer's money but you only give them the call if every part of the parlay is correct. Assuming they charge a commission in addition to the asset price to cover transaction processing they shouldn't lose money

Edit: I don't really know how pricing these things usually works but I could see taking some risk on to price these attractively

toast0•36m ago
I get that Robinhood's business model is stealing from the poor and giving to themselves, and their customers are mostly unsophisticated, but why wouldn't the customers just buy the underlying call options if the price to buy the parlay is the sum of the underlying options?
sdwr•34m ago
Why would anyone ever take that bet? It has to be leveraged, which is where the risk comes from.
lordnacho•59m ago
Parlays: the problem is really just spread. For every leg, you are going to get a shitty spread, meaning you lose a bit from the "real value" of the trade. Say you are flipping two coins (eg two very evenly matched games) and so the outcomes should be 50-50. Well, if the market ends up showing 48-52, you are losing two points in edge. Compound it and you are losing even more.

The great thing about parlays is that when people win, they win by a big multiple. So they feel they have won. But when they win, they are actually getting less than they should have gotten on their winner. The example above should 4x your money since the real chance is 25%. You punter might end up with say 3.5x, so even though he feels great when he wins, he isn't winning enough to make up the loss the other times.

Perps: Traditional markets have futures that settle on a specific date. For instance, S&P futures. This presents a couple of issues for the uninformed.

First, the settlement means your bet ends on a certain date. People want to avoid having to sell their position and put it on again in the next expiry. It also just seems like a fee grab by the exchange.

Second, the futures price differs from the index, due to financing rates being different between the assets. Remember a future is a promise to exchange at a later date, so you have to take into account the time value of money, aka interest rates. Well, early crypto didn't have a bitcoin interest rate, and so any gap between the future and the index would be an implied rate that you were punting, which nobody would understand if they didn't work in finance. In any case, there would be questions to the customer service desk about the deviation. Much better to hide the financing rate inside the perp rate adjustment that happens every 8 hours, and presents the price of the perp as more or less equal to the price of the underlying bitcoins. Early Bitmex also had the entertaining wrinkle of not being able to trade against a stablecoin, so actually you had inverse perps that settle against the crypto in kind. This creates a weird non-linearity vs the dollar, but meh whatever, number go up. These are things that the market maker understands, but the public doesn't.

Third, the automatic settlement and liquidation system is actually pretty innovative. You can give people massive leverage because you know exactly who has what position at the exchange level, in real time. Traditional markets still settle on a daily cadence (often T+2/3), meaning you could do funny shit that your PB would have to build a system to look at.

0 days: Just another way to get fleeced on spreads. There are models that estimate how likely some price is to be over a line at the settlement time. You definitely want to use statistics to determine this kind of thing, but people think they are smarter. The market maker isn't going to care terribly much, as long as he is reasonably hedged. There are well-known ways to spread your risk across options, and that's what the market maker does.

Leverage is the common denominator here. These are all bets where you can make a lot of money with a little bit of capital. It's an age-old story that people will blow themselves up with leverage. After paying fees to put on a bad bet.

shermantanktop•6m ago
...when your business plan is based on attracting users who failed the marshmallow test but think they are above average at doing so.