I think we’re talking past each other. Let’s start where we agree: The aggregate CFR measured at any point in time using cumulative cases and cumulative deaths will be understated for exactly the reason you pointed out: not everyone who tested positive that will die (and will ultimately be included in the CFR) has died yet.
But I’m talking about something else: Trying to estimate the CFR for the people who are currently testing positive. I’m interested in this number because I’m interested in trying to predict likely deaths in the next few weeks. I think the best simple formula for that is: Expected future deaths = Current cases * estimated CFR for current cases
The reason that cumulative CFR is likely to be overstated for that purpose (despite the bias you mentioned) is that the population of people who are testing positive now is very different from the population of people who tested positive much earlier in the pandemic. If you assume that there are “serious” cases and “benign” cases, each of which has a very different CFR, then the ratio of those cases will significantly affect the CFRs for any given cohort.
When we observe a decline in measured CFR over time (based on current deaths relative to lagged positive tests), I think what we’re seeing is a greater proportion of benign positive tests.
On the prediction side I think we have a bounded range at 2,000-3,000 per day for 7DRA about August 12.
Given uncertainties around cfr: reporting of cases, reporting of deaths, positivity, demographics, etc I think that’s actually pretty tight. I’d be surprised if we miss.
Just for round numbers I’m going with a slope increase of 50/day unabated for the next four weeks getting us to about 2200.
So what you want is to know cfr for every x days after positive test. Unfortunately the data doesn’t tell us that because we don’t tie people to an identifier to know a death on July 16th had tested positive on June 2nd.
So I’m just assuming roughly 4-5% of 7dma 5 weeks after a date as a rough guide but obviously that will be affected by if cases are rising, falling, steady on either side of a week, so it’s a very rough estimate.
So we hit 60k cases 7dma about 3 days ago so that’s 3k deaths in mid-late August. And case average is rising so over 3k later in August.
At some point I may build a neural network to predict deaths on a date based on the data but until then I’m going on this as a rough guide.
By gut check, our cases are much higher than April with similar positive rates so our deaths should be higher than the mid2k we had back then.
So here’s a way to visualize it, I think:
What happens when we calculate a CFR based on current cumulative deaths compared to identified cases as of 4-5 weeks ago. If we do that, I think the evidence points to a likely CFR for current cases closer to 3%.
And what’s interesting (to me) is that the numbers look extremely similar regardless of whether you use a 28-day lag or a 35-day lag. That suggests to me that most of the deaths from a given cohort are likely happening within 28 days.
So I think it’s going to be much closer to 2k than 3k based on current cases.
Also it should be noted that most of the places spiking aren’t doing a ton to get r down, which is part of projecting higher deaths.
If everyone locked down today, then we may only get to 1500-2000 deaths/day from what’s already out there, but since we aren’t locking down, it’ll go higher.
If nothing was being done anywhere then I’d be saying over 4k in 8 weeks but lots of store chains are requiring masks and many states are at least closing bars and some dine-in so that will help a good amount.
Need to do it state by state because Florida and Georgia will be worse than California. Georgia isn’t even closing bars yet; even Florida did some bar closing.
Marshmallow experiment was a mess. Poor kids eat the marshmallow. Rich kids don’t. Rich kids do better in life.
Amoung poor kids who waited and rich kids who didn’t. There’s no difference. I.e. the marshmallow means nothing and it’s another way to blame poor people for their poverty.
I think I’m just going to post this on all my whiny local FB groups when they start bitching.
I keep seeing people posting and tweeting about how they see fewer trump flags, etc…but in my neck of the woods, I see more, and the ones who had them before are doubling down…they have a blue AND a red trump flag now!
These are the figures used by the IPCC in their latest report.
World spending on energy infrastructure is about 1.8 trillion per year. That probably needs to go up to about 2.5 to 3.5 trillion per year to keep warming below 2 degrees.
Yeah I’m counting on Trump getting at least 62,984,828 votes and if thats 100% of 2016 voters or 90% of 2016 and 10% of first time voters, people who sat out last time, or people who don’t like lawn signs it doesn’t matter what matters is finding at least 65 million people to cast their vote for Joe Biden and all these polls are just guesses. I don’t know anyone personally who admits to voting for Trump in 2016 who says they will vote for Joe Biden so I’m putting them all down for Trump again.
@Melkerson to be tucker carlson for a year and then have to go back to being me no amount of money but i’d be tucker carlson for life for like 50k~ a year. your avg professional troll makes like 2k usd a year and only a few lucky ones hit it big and make millions but i just do it out of love of the game and wouldn’t be greedy.
The thing that is hardest for me mentally is that we don’t even know for sure that we’re getting a second marshmallow. Like, maybe I stay locked down for the next week/month/year and I end up:
missing out on a chance to see my Dad for the last time before he dies
irreparably eroding friendships with people that will never be fully rebuilt because it’s just too hard to overcome both the inertia of not seeing someone for a long period of time and the realization of how much we did not prioritize staying in touch during this time
closing my fertility window
I mean, I’m still staying quarantined because I believe it is the right thing to do, but I’m not under any illusions that I’m necessarily going to receive some big cosmic cookie at the end of this ordeal.
I actually know a few people who voted for Trump that aren’t going to this time (whether that means Biden or just leaving it blank, who knows). Of course, this is in CA where it doesn’t matter.
