mathbabe

I’ve spent a bunch of time worried about Florida and COVID in the past few months, partly because my grandma lived there for a number of years and so I spent a bunch of time there growing up. It’s a really vulnerable place, in some ways more so than Manhattan. And according to the data, and some reckoning, I figure the daily death counts will soon hit 600. I explain why in my new Bloomberg column:

Florida’s Covid-19 Deaths Might Rival New York’s

The state’s daily fatality count could hit 600 in a few weeks.

More of my Bloomberg columns can be found here.

You guys might have been wondering what happened to me! Well the answer is I moved to Somerville, MA temporarily, and it takes a TON of work to move, especially when you haven’t moved in 15 years!

Side note: I discovered I am a major hoarder in the categories of clothing, shoes, and yarn. This is something that is easy to deny when you don’t have to empty out large closets but is impossible to deny when you do.

OK so, and this is dark, I wrote a Bloomberg piece about my definition of quantified dread, which loosely speaking when things are getting worse and the rate at which they’re getting worse is getting worse:

America Is Being Way Too Calm About Covid-19

This is a case where optimism may be an existential threat.

You can read more of my Bloomberg columns here.

Here’s my newest Bloomberg column, in which I argue that we’ve already done the data work behind defunding the police, because “crime risk scores” predict police, and not in a good way:

Here’s an Algorithm for Defunding the Police

Crime-risk scores reveal the problems that society has shunted onto law enforcement.

Read more of my Bloomberg columns here.

Here’s my newest Bloomberg column regarding the state of facial recognition.

Amazon Can’t Make Facial Recognition Go Away

That would take an act of Congress.

Read more of my Bloomberg columns here.

Here’s my newest Bloomberg column, in which I discuss the darkest, scariest kind of data, namely missing data. We are getting some of those holes filled in when it comes to police misconduct, and we need more.

Don’t Let the Police Hide Their Bad Behavior

Fixing law enforcement will require better data.

Read more of my Bloomberg columns here.

My newest Bloomberg column, in which I examine how Sheryl Sandberg’s “Lean In” philosophy might be guiding her during the current Facebook shitstorm:

Maybe Sheryl Sandberg Should Be Leaning Out

Facebook needs better moral leadership.

Read more of my Bloomberg columns here.

In my newest Bloomberg column, I make the case, using new research by Measures for Justice, that mass incarceration, inequality, and racism cause epidemics both here and worldwide:

Maybe Racism Caused the Covid-19 Crisis

Mass incarceration and other social ills made the world more vulnerable.

You can read more of my Bloomberg columns here.

I’ve got a new Bloomberg column out today, about the game of chicken that colleges are playing with students and their parents:

Covid-19 Will Make Colleges Prove Their Worth

Online education should come at an online price.

See more of my columns here.

In my new Bloomberg column I suggest that R(t), which is a hyperparameter in most Covid-19 models, is a much better and more trustworthy figure to follow than any other particular data set.

One reason, which didn’t get into the column, is that R(t) can be estimated from most other daily data sources like hospitalizations, cases, or even deaths, albeit with lags. That means that we can piece together a trustworthy patchwork quilt of R(t)’s that might be more trustworthy than any particular version.

Moreover, R(t) is insulated from the bias we know exists in these figures (due mostly to not enough tests) and only cares about trends, so as long as the bias is consistent we don’t care about it.

The caveat here is that we’ve seen many states performing Covid-19 data manipulation (Texas, Florida, and Georgia for example) in order to open up sooner than they honestly should. Basically, they’re juicing the numbers. That’s a kind of political bias we cannot overcome easily (unless they forget to manipulate some of the data!).

Anyway, that’s a nerdy postscript on the following:

Here’s a Covid-19 Number Worth Watching

My other Bloomberg columns are available here.

I wrote a Bloomberg columns which heavily relied on Jarod Alper‘s recent YouTube talk:

Meet the Covid Models That Are Running the World

See other Bloomberg columns I wrote here.

Hi all,

On my walk to work I realized that the new FEMA model – which was used to strong-arm the Arizona governor into opening early – is a WMD, i.e. important, secret, and destructive:

A Secret Algorithm Is Deciding Who Will Die in America

Decisions on reopening should involve public data and debate.

See more of my columns here.

Hi all,

Two more Bloomberg column posts to share with you. First, one from Monday:

Let’s Make a Deal With Our Robot Overlords

and here’s on from yesterday:

I’ll Miss the New York City Subway

More columns are here.

Guys I’m sorry I forgot to blog last Friday about a piece I wrote:

Trump Is a Machine-Learning Algorithm Gone Wrong

We Can’t Get Together Until Tests Get Better

More columns are here.

I was happy to connect with my friend Ashton Applewhite, an ageism activist whom I met at TED, to discuss aging in the time of Covid-19. It led to this new Bloomberg column:

Pandemic Data Could Be Deadly for the Old

See other columns I wrote here.

An empirical observation about models versus reality:

This Isn’t the Flattened Curve We Were Promised

See other columns I wrote here.

Another skeptical column from me today:

The Covid-19 Tracking App Won’t Work

See other columns I wrote here.

Hi all,

I’m back at Bloomberg, writing about reasons to doubt the daily data we keep seeing. I’ve added a few reasons since my post last week. Also, I’m preparing myself for bad data today and tomorrow delayed from Easter weekend:

10 Reasons to Doubt the Covid-19 Data

See other columns I wrote here.

This is a guest post by Gary Cornell. Gary holds a Ph.D. in mathematics from Brown University and was the co-founder of the the major technical publisher Apress. He has written or co-written numerous best selling programming books and has been a Mathematics professor, a visiting scientist at IBM’s Watson Labs and a program director at the National Science Foundation.

I’m not a doctor nor do I play one at daily press briefings. But like most mathematicians, I do know something about basic statistics. And, like most academics, I read everything that comes out about drugs I am taking. Obviously, I concentrate on the statistical sections and the list of side effects and drug interactions in these research papers. And so, this being 2020, I have a google alert for the drugs I am taking.

Anyway, one drug I am taking is the maximum dose of metformin. It is the fourth most prescribed drug in the United States and is used by more than 150,000,000 people world wide. It is the usual first drug prescribed for Type 2 diabetes. It is a good drug, the side effects are usually mild and it is even being explored (the TAME trial) as a possible “longevity” drug. A good drug…

So I was shocked to see this link popping up in my in box:

read it here: https://www.biorxiv.org/content/10.1101/2020.03.31.018556v1

This note is from researchers  at Johns Hopkins – which I hasten to point out is currently rated the #2 school in the United States for “medical research”. People there are not general considered practitioners of psycho ceramics, in other words.

Holy ^&%$#, I thought. And then I tuned in to the Sunday “briefing” where Trump doubled down on his pushing claiming that “what do you have to lose” – and prevented Fauci from tempering his response. Though mice results aren’t conclusive and perhaps fatalities are off by a factor of 10 or a 100. Who the &^%$ knows if taking this will kill me? Maybe it is much worse for people on the maximum dose of metformin like me. But, absent proof it is safe for people taking metformin, pushing it if you are taking metformin except in life or death situations would seem to me malpractice for a doctor and practically criminal for an economist or politician.

Oh, in case you are thinking that maybe except for people taking metformin, hydroxychloroquine is a “good” drug, Drugs.com says there are 332 drug interactions (59 of them being major). And here is the list of side effects

So, answer is, you have a lot to lose.

You’ll need:

• Dishcloths which some people call tea towels (could use any cotton cloth but slightly thicker is better)
• 20 gauge metal wire cut into 2.5″ pieces (could use pieces of a thin wire hanger instead)
• Elastic cord (could use rubber bands or strips of cloth instead)
• A sewing machine
• Good scissors
• Crochet needle and sharpies
• Cardboard

I’m following the pattern for “Mask 2 (large)” on this webpage. But to be honest I found it hard to follow which is why I’m going to tell you quite plainly how to do this relatively quickly.

First, download and print this pdf: mask+2+large+pattern

Or simply eyeball the following picture with the ruler as a guide:

You’ll want to cut out the printed version and then outline it onto cardboard, then cut out the cardboard so you’ll have a form you can reuse a bunch of times with sharpies:

This has been used a bunch and is kind of a mess. That’s ok.

Then you outline with a sharpie on your dishcloth:

I got 12 cardboard outlines on! That means 3 masks. Be sure to avoid the hemmed edges.

Next you cut them all out:

Next, pair up the cloth pieces to match:

Next, sew along the foot of those matched boots for both pairs with a 1/4″ seam:

And now put those two pieces together, with the seams on the outside for both pieces:

Next, sew all around the above piece (so sew the two pieces together) with a 1/4″ seam except for about two inches at the bottom seam:

It’s time to turn this whole thing inside out by squeezing it through that two inch slit!

Poke your fingers into all four corners plus the nose part at the top to make sure it’s all the way inside out.

Here’s the other side:

Next, sew a three inch line along the nose top (a 1″ seam) and stick the metal wire into that channel:

Do you see the channel? The wire has to go in this area except, of course, you need to have it on the inside.

Next, you want to sew along the entire edge (very close to the edge, maybe 1/4″), starting at one end of the nose wire channel. Halfway along you’ll carefully close the hole at the bottom:

Here it is at the end:

this picture is overexposed but the idea is you close up the nose wire channel on both sides.

Next, you fold back 1.5″ of the ear flaps and sew down:

This is the back of the mask after sewing down both flaps.

Next, measure out 1 yard of elastic cord and tie together the ends:

Next, use a crochet needle to pull through flaps and then tie it together:

Finally, yank the elastic cord until it’s hidden inside a flap and tidy everything up by snipping off the stray threads.

It’s ready for a cute model!

Now’s the time to squeeze the metal wire to make it fit your nose.

I am, like you, restless and having trouble coping with the tragedy going on. It’s especially hard to think through the logical details of issues that only two weeks ago seemed urgently important. So instead, like you, I find myself with an internal dialogue of how the publicized statistics are consistently biased or wrong. At the risk of simply supporting your own internal thoughts, here are a few of mine:

1. We still aren’t testing people, even in New York, which is the most tested population in the current mostly highly infected country according to the crap data we have.
2. What that means to me is that we can ballpark how many actual cases we have if we know what the condition is for actually getting tested. In New York, it’s something close to “needs hospitalization.” Considering that only about the worst 10% of cases in countries that do widespread testing actually need hospitalization, that means we can multiply our confirmed case count by 10 to get an estimated total case count.
3. That means that, instead of 60K cases in New York state, which is what this webpage says this morning, we can assume it’s actually more like 600K.
4. Similarly as a nation, we should multiply the confirmed case count of 143K by ten to get an estimated 1.43 million cases in the US.
5. Is that an overestimate? Perhaps. It’s possible that enough testing is happening in those car wash type setups, where people are at least capable of driving a car, to make it pessimistic.
6. On the other hand, we’ve seen plenty of examples in the NYC area of people calling their doctor with intensely bad symptoms who are told not to overburden the hospital system and to take care of themselves at home.
7. Also, it’s worth pointing out that multiplying by 10 assumes that more than half, and perhaps up to 75% of all actual cases are entirely asymptomatic. This is something we’ve been seeing in places that have done randomized or comprehensive testing.
8. All the above are ballpark reckoning, but honestly I trust my numbers more than any official ones.
9. Especially because we’ve been hearing stories told in Spain and Italy that their death counts are not including horrible fucking things that have been happening in nursing homes. That means those terrible numbers are heavily underestimating actual deaths.
10. Also, we should not trust China’s death count numbers, which some say are underestimating actual death counts by a factor around 15.
11. And if we don’t trust their death counts, we should also not count their confirmed case count, which has been tiny recently.
12. Why this matters a lot to us right now is that China closed Wuhan on January 23rd, which means they are/were under quarantine stricter than ours for more than two months, and we’d REALLY like to know what the actual situation is right now, but we don’t.
13. Long story short, being a skeptical data scientist means not trusting the data whatsoever. The best we can do is use the data and our real world knowledge to ballpark what might actually be happening. We will never know the true numbers.
14. One exception might be the Netherlands, which I’m keeping my eyes on. I don’t think they lie as much as most other countries.
15. I could be wrong about that too.
16. I hope tomorrow’s post will be more optimistic.
17. One last comments: Sunday reported deaths are lower than other days because of the way reporting happens.  Doctors and others are taking a well-deserved rest. So don’t get excited about flattening curves based on Sunday data: