@JamesEvans CNN reporting something similar: McConnell believes impeachment push will help rid Trump from the GOP, but has not said if he will vote to convict

Several GOP sources said on Tuesday that if McConnell supports conviction, Trump almost certainly will be convicted by 67 senators in the impeachment trial.

Some troubling events last night: - [7 people shot at the Louisville, Kentucky protests](https://www.nytimes.com/2020/05/29/us/louisville-protest-shooting-breonna-taylor.html): It sounds like none have died yet. No officers discharged their weapons. - [Shots fired during a rally in Denver, Colorado](https://denver.cbslocal.com/2020/05/29/panic-denver-shots-fired-during-george-floyd-rally-outside-colorado-capitol/): No injuries. Police don’t have a suspect. - [Armed vigilantes “defending” stores](https://nypost.com/2020/05/28/armed-rednecks-stave-off-loo...

The latest reports are saying he wasn’t put on any type of ventilation (neither invasive nor non-invasive). I’ve switched to the overall ICU base rate (44%) and updated my prediction to an 89% chance of recovery.

As @KrisMoore pointed out, I've been working on an [exponential smoothing model](https://github.com/beala/covid_forecasts/blob/master/Exponential%20Smoothing%20Model.ipynb). - If I apply exponential smoothing to the US totals, I get a prediction of 1.09M (~21,080 cases/day) - If I apply the model to each state's total, and then sum the predictions for a nationwide total, I get a prediction of 1.11M (~25,300 cases/day) - As others have mentioned, a linear model gives around 1.16M (~33,600 cases/day) So that's quite the spread! Here's what I think: - L...
A good round up of all of Trump's legal liabilities: The Economist: [Donald Trump faces an array of legal trouble when he leaves office](https://www.economist.com/united-states/2021/01/17/donald-trump-faces-an-array-of-legal-trouble-when-he-leaves-office) > Potential charges, if evidence is found, could include scheming to defraud, falsification of business records, insurance fraud and criminal tax fraud. Some of these are felonies carrying penalties of up to 25 years. > > ... > > Other troubles await Mr Trump in New York that could cost him money, if...

Trevor Bedford, an epidemiologist who has been studying the spread of COVID, tweeted his arguments against the hypothesis that the virus originated in a lab: https://twitter.com/trvrb/status/123063413610…

I’ve updated to 15%: - I agree with @Tamay that prison riots are unlikely to be deadly enough to trigger this forecast, but I think there has been an uptick in frequency, so I’m updating slightly. [One paper](https://journals.sagepub.com/doi/10.1177/003288558306300104) I found implies a rate of 2.86 prison riots/year. There have already been 2 COVID related riots this year, and there are 8 months left. - Incidents of unrest around the world related to COVID, [mostly](https://www.aljazeera.com/news/2020/04/fire-tears-greece-refugee-camp-coronavirus-prote...
[Exclusive: U.S. plans massive coronavirus vaccine testing effort to meet year-end deadline](https://www.reuters.com/article/us-health-coronavirus-usa-vaccine-exclus/exclusive-u-s-plans-massive-coronavirus-vaccine-testing-effort-to-meet-year-end-deadline-idUSKBN22Y2L3) > The United States plans a massive testing effort involving more than 100,000 volunteers and a half dozen or so of the most promising vaccine candidates in an effort to deliver a safe and effective one by the end of 2020, scientists leading the program told Reuters. > > ... > > Candidate...
New post by Gelman on The Economist’s model: [So, what’s with that claim that Biden has a 96% chance of winning? (some thoughts with Josh Miller)](https://statmodeling.stat.columbia.edu/2020/11/02/so-whats-with-that-claim-that-biden-has-a-96-chance-of-winning-some-thoughts-with-josh-miller/) > our backfitted predictions seem to be overconfident . . . so I don’t fully believe our numbers either. > > Our model was not designed to make sense in the tails; we didn’t think through what would happen if the vote share needed for Trump to win became a tail even...
This is a good overview of the events leading up to Twitter’s new approach to handling Trump’s tweets. It’s paywalled, so I will quote liberally. [Inside Twitter’s Decision to Take Action on Trump’s Tweets](https://www.wsj.com/articles/inside-twitters-decision-to-take-action-on-trumps-tweets-11590866833) > The moves marked a sharp reversal for Twitter, which for years has faced criticism from users for what they see as inaction and inconsistency in policing its own platform, and turmoil among its employees over how it has managed its most prominent use...

Interim analysis results posted here: https://www.thelancet.com/lancet/article/S014…

And some commentary from Natalie Dean:

Out in @TheLancet, results from the Oxford/AZN trials, including more detail on the low dose results. Notably, the low dose recipients "received their second dose after a substantial gap." Only 0.8% received a second dose within 8 weeks of the first. 1/5

This study from China sets the base rate for survival given that Boris has been put on "non-invasive ventilation" (I assume CPAP counts) to 21% (https://www.thelancet.com/journals/lanres/article/PIIS2213-2600(20)30079-5/fulltext ). The reason my forecast is so optimistic is that I believe Boris will have access to cutting edge treatments, such as convalescent plasma (plasma donated from someone who has recovered). In studies on SARS, it increased the odds of survival 8:1 (https://www.ncbi.nlm.nih.gov/pubmed/15616839). In a small study on COVID, all 10 p...
Video interview: [Moderna CEO forecasts 2021 as the earliest for an approved Covid-19 vaccine](https://www.cnbc.com/video/2020/04/14/moderna-ceo-forecasts-2021-as-the-earliest-for-an-approved-covid-19-vaccine.html) - The interviewer mentions a plan to have vaccines for high risk groups in the fall. [March 23rd article on this](https://techcrunch.com/2020/03/23/moderna-could-make-experimental-covid-19-vaccine-available-to-healthcare-workers-by-fall/). - They won’t wait on the full Phase I dataset before starting Phase II. They will begin as soon as they ...

I’m curious about how much time people spend on their forecasts. I’ve been casually interested in forecasting for years, but only recently started doing it in earnest. I can spend days trying to answer one forecasting question. Is that normal? Or am I way past the point of diminishing returns?

@ege_erdil I definitely feel like I've learned a lot.

@MaciekK

Even a booster should count, IMO.

This question is primarily asking about updated vaccines. In the fine print:

The guidance must recommend revaccination with a different vaccine than the one that was originally administered.