All you need to know about long Covid - a big study compared outcomes 1 month to 1 year later in folks who had Covid with a matched group who didn’t.
The broadest measure of serious outcomes, hospitalizations, was EXACTLY the same in the groups.
Because long Covid IS NOT REAL…
Except for people who were sick enough to be hospitalized with Covid, especially those who required intensive care. Those people may have a long recovery.
But people with less serious Covid have basically no long-term damage. Even the authors ~admitted it:
1/3: Let's be clear, @MarionKoopmans: @NIHDirector_Jay is right. You absolutely pushed lockdowns.
In 2020, you and everyone at the top of @WHO (along with Tony Fauci) were caught between China and your secret fear that Chinese scientists were responsible for a dangerous virus...
2/3: Instead of telling the truth and letting the world navigate a few weeks of uncertainty, you pushed a lockdown and containment strategy - contrary to every previous plan for respiratory viruses - even as it became clear that Covid posed almost no risk to healthy people...
3/3: I know. I was there. And I didn't need a PhD in epidemiology to understand what was happening. Neither did anyone else.
You and the gang did extraordinary harm to public trust in science and medicine. Get yourself a less ridiculous avatar photo and apologize to the world.
Professional leagues basically need to suspend most prop betting (that is, betting on individual players's statistics) and in-game betting, but they won't, because those bets make far more money for betting companies than the topline game spreads.
In-game and prop bets are more profitable for several reasons
1: Traditional sports bets take hours to resolve. In-game bets win or lose in seconds and facilitate loss-chasing.
2: Prop bets let the betting companies use their massive databases to set more favorable odds...
3: In-game and prop bets facilitate parlays (multiple simultaneous bets that win or lose as one bet), which have much larger payouts (sometimes lottery-like) but also offer a much larger and hidden house edge (because the house has an edge on EACH leg of a parlay)...
1/ No, I don't trust a Chinese company to produce "safe" AI.
But the DeepSeek breakthrough is fantastic for both practical and philosophical reasons.
Here's what I mean: we've never had a tech hype bubble bigger than AI - not even the early Internet...
2/ And a key element of the hype the huge complexity and expense of the systems these companies are building to produce AI.
We're gonna need a whole new electrical grid! But it'll be worth it, because reasons. We promise!
3/ The race for complexity and cost has taken on its own logic. You're paying a guy who writes transformer code $500,000? Screw it, we'll pay him $700,000! You're buying $20 billion in Nvidia chips this year? We'll go $30!
There aren't just longitudinal studies, there are longitudinal studies examining both causation and reverse causation (looking at use and psychosis over multiple periods to see if psychosis in N is causative for cannabis in N+1. It's not)...
2/ The associational data is incredibly strong, with unadjusted ratios on the order of 10x.
There is clear biological/cellular level data showing that THC use dysregulates the cannabinoid system and that heavy users have changes in brain morphology...
3/ Many cannabis users (and practically all heavy users) have some experience with cannabis paranoia - which looks a lot like prodromal psychosis, why is everyone laughing at me? This is so common users joke about it - and have strategies to deal with it...