Josh Foster

Be honest, Iowa, you made that app out of corn.

I'm not happy Rush Limbaugh has lung cancer. Nobody should be. I've had family members die from lung cancer and it's awful. My grandfather had lung cancer and when he died, he coughed up so much blood on the bathroom walls that homicide had to investigate. I wish Rush the best.

It's not all that dumb to confuse KC, Kansas for KC, Missouri. But most of us would have had that little doubting voice in our heads telling us to double-check before posting it to our 72 million followers. Trump lacks that voice, which is part of the mixed-blessing of narcissism

Goalie fights, no matter how ridiculous, are the best things in hockey, bar none. I'll die on this hill.https://twitter.com/HockeyCentraI/status/1223833041891733504 …

This dude's the real deal. Can't wait to see him play in person.https://twitter.com/PelicansNBA/status/1223427087450677248 …

I wonder if any of the GOP senators arguing that what Trump did was wrong but not bad enough to remove him from office voted to remove Clinton from office in 1999.

Of course, in real life my BF-hacked prior would be absurd. And, of course, nobody should adjust priors just to attain bigger BFs. But I've found it useful to experiment with different parameters to get a feel for how this whole thing works.

A kinda fun way to get a feel for Bayes stats is to BF-hack a result by adjusting location/scale of prior distribution. My best result so far (effect, d = .09) is going from a BF = .93 (Cauchy, location = 0, scale = .707), to BF = 22.70 (Normal, location = .09, scale = .01).

I'm a Bayesian noob, but I'm trying. Anyways, at the risk of embarrassing myself...I did a study based on a previous study that reported a Bayes factor. Can I use that BF to inform my prior (e.g., adjust location/scale of prior distribution)? Recommendations for practical guides?

The major difference is that your friend got arrested whereas @RandPaul will get reelected.

Remember when your idiot friend was convinced that if you asked an undercover cop if they were a cop they'd have to tell you? That's basically what @RandPaul tried today. #RandPaulResign

Everyone is focusing on the genetics "finding" in this study, which looks unconvincing. But as a father with two boys who play soccer, it's the heading finding that's the real story here. And that one looks much harder to explain by randomness.https://twitter.com/ShaiCarmi/status/1222593017317601283 …

It was particularly devastating because it killed young adults at some of the highest rates. It appeared to trigger an out of control immune response (cytokine storm), which made it particularly bad in young healthy people.

About one-third of the global population was infected and as much as 6% died.

Efforts by warring nations to cover up the severity of the outbreak led to tens of millions of infections and deaths. A parade that attached 200k people was held in Philadelphia despite warnings (to avoid showing weakness). Within 6 weeks, 12k people were dead.

It was so bad that life expectancy in the US dropped a staggering *12* years in the first year of the pandemic.

In light of current events, I highly recommend this fascinating podcast on the "Spanish" Flu pandemic. (It should have been called the American Flu because it started in Kansas, but no country other than neutral Spain reported cases because of WW1.) https://youtu.be/9_ann3OeiZA 

When I get weighed at the doctor's office the nurse always comments that I don't look like I weigh that much. But today she just said, "OK."

In hindsight, I think we should have subpoenaed Bolton et al. & let it work its way through the legal system. At this point, I think we should trade Hunter Biden for Bolton & let the chips fall where they may. And with that I'll cede the rest of my time to the gentlewoman from