Announcement

Collapse

Announcement

~ gang gang ~
See more
See less

US Covid forecast doesn't look great.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • US Covid forecast doesn't look great.

    Few random numbers.

    You broke 1k dead for 7-day average on 21st. Last time you did it on an upswing was 7th of November.

    You're at 22k for ICU. Last time 1st of December.

    You're hitting 15% for positive tests on 16th. Last time you did that was April last year with about a 1/5 of total tests. The previous peak was 8th of January with 14%.

    For cases you're at 150k. Last time it was in 15th of November when you tested about 1,5x people.

    Basically nothing is indicating that any of those numbers are going down. You don't seem to have huge benefit from vaccinations or ability to treat the sick better. By all accounts you're worse off compared to last year. To a degree that's also true about Europe, but last year you were a month behind and now you're way ahead.

    I don't know if there's any kinda valley or plateau before spring.

    Also everyone's ICU and death numbers are jumbled. Guessing younger people don't die as fast so they keep padding the ICU figure. What's weird is that US doesn't have same kind of disparity with hospitalized to ICU ratio compared to last year.

  • #2
    not even joking or trolling, but it's not really that deadly and those numbers are not scary at all. the biggest danger COVID brings is the people in control freaking out and shutting everything down again. and yes i'm vaccinated

    Comment


    • gimmick
      gimmick commented
      Editing a comment
      Numbers themselves are whatever. It's how fast and early they are growing.

  • #3
    for me the real news is we are going full India on medical resources/response. like we could literally be 10% into the current surge and texas, florida, and a few other states literally dont have beds to treat people and florida is commandeering liquid oxygen from their state water purification systems.

    so another prospective thesis is 'how non-deadly is this virus when people arent recovering in hospitals or able to get regeneron or whatever, but have to roll the dice while laying on their couch for days, shitting and pissing into the cushions every night'. oh also winter is coming so.....

    Comment


    • gimmick
      gimmick commented
      Editing a comment
      Yea it's really early and you haven't had this steep of a rise in death numbers since the beginning, excluding Thanksgiving and Holidays when no one felt like tallying up the dead. Nothing special has happened in the last few weeks far as i know. Deaths are roughly following the rise in hospitalizations from the beginning of the month.

  • #4
    Click image for larger version

Name:	Screen Shot 2021-08-22 at 7.43.19 PM.png
Views:	157
Size:	205.8 KB
ID:	7488

    Comment


    • #5
      Originally posted by sonatine View Post
      for me the real news is we are going full India on medical resources/response. like we could literally be 10% into the current surge and texas, florida, and a few other states literally dont have beds to treat people and florida is commandeering liquid oxygen from their state water purification systems.

      so another prospective thesis is 'how non-deadly is this virus when people arent recovering in hospitals or able to get regeneron or whatever, but have to roll the dice while laying on their couch for days, shitting and pissing into the cushions every night'. oh also winter is coming so.....
      I’d be willing to put a small wager in the fact that Florida and Texas have temporarily peaked. Turns out it’s hot as balls in both places and nobody likes to be outside this time of year. Check the trends July - October last year and the numbers (cases, ICU, Deaths) line up near perfectly. It was right around the end of August last year when Northerners were laughing at DeSantis for gross mismanagement. 6 weeks later the shoe was on the other foot when cases plummeted in the south and started to climb in the north.

      maybe this time is different but I think the new normal in part will have regional Covid season.

      I’m not suggesting Florida will hit heard immunity in a true sense but we are starting to run out of people who haven't either had it or been double jabbed. It’s almost everyone down here at least in my experiance.

      Comment


      • #6
        lmao im not up to wagering money on a infection/death count at this point in my life, that seems like tempting fate.

        but your point is well taken and i do hear that delta is peaking but im also thinking more about like, feb #'s, not october numbers.

        Comment


        • #7
          There's still a whack-a-mole element to it, so i wouldn't be saying that current 5-7 states at the top are going to stay there. It's a big enough country that all states get their time to "shine".

          When looking at the individual states you mostly have to look at the whole year if you were to look for effects of local policies and even then for climate/culture you would have to give more weight to closest neighbors, demographics, mobility, population density and few dozen other things.

          Say Florida is old, decent sized metro, i'm guessing they still have tourists and i don't really know what state is even close in climate. Comparing them to Vermont is pretty impossible. By median age Vermont is older. For population size Florida is 30x Vermont. For tourists Florida is at 30m domestic 1m global and Vermont is at 4m domestic dunno if there's any global interest. Tourists come at completely different times and purposes. Skiing for Vermont and Miami things for 2/3 of all Florida visitors. Miami metro is like 8x Vermont.

          Per capita Covid deaths now are about 200ish for Florida and 50ish for Vermont (per 100k). By the end of the year it's maybe 3-1 for Florida. I have nowhere near enough data or work towards it to say how much of it can be isolated to policy. And with policy i would also mean compliance and resources provided towards it. I'm guessing Vermont is more accepting to mandates and better positioned to shocks to economy.

          Comment


          • #8
            Last year when i predicted 100k deaths before the end of the year with two months to go it was fairly easy. I had seen half of that movie play out in EU. I just saw the first 5 minutes in US and roughly knew what was going to happen.

            At the time Druff and BCB doubted that because all they had seen was few months of steady numbers with a small rise. Before that Tellatard had his haha moment with look how bad EU is now compared to US. Well it was obviously Italy all over again. In the 6-8 months the difference in "lag" had tightened from 2 months to about 1 month. I assumed some climate related thing would keep it there. This thread was about that lag either disappearing, US pulling ahead or disconnect of the two.

            With large population sizes and things like infectious diseases once a trajectory has been set you have a fairly good idea where it goes in the near future. I had seen the first month and then i knew the range of the second month. I think 80-120k was the range i gave and without saying it with 90% confidence.

            After my prediction came true Druff proved himself right by listing some random specific things that caused him to be wrong and how no one could see them coming. His take was that something specific had to happen when the opposite was true. Something very specific had to happen to change the trajectory.

            This time around i don't know what's going to happen. I don't have the cheat code.

            edit: there's similarities to market charting roughly at the industry level with "stable" conditions

            Comment


            • sonatine
              sonatine commented
              Editing a comment
              genuinely empowering to have your perspective on this

          • #9
            Roughly 90 days after update on what went down.

            You hit peak with deaths around month after the OP with 2k deaths for 7-day average on 21st Sep.

            A decent decline that's still going with 21st Oct 1,6k and 10th Nov 1,2k.

            With the current trend you have shot at hitting 1k dead avg. before the winter ascent.

            Last year you racked up around 80k dead between Oct 1st and Nov 1st. This year in the same time period around 130k.

            Last year November resembled previous two months (around 40k). For a total 120k dead for the whole 3 month period. It's the next 3 moths that are really bad.

            Between Dec 1st and Mar 1st was the 240k dead period. Raw extrapolation for this year gives us...

            195k dead between Oct 1st and Dec 1st. 390k dead between Dec 1st and Mar 1st.

            Nov 1st was around 750k cumulative and that gives us around 945k at the end of the year. 1m would break around mid January.

            That's the end of the mildly morbid Saturday morning forecast.

            Comment


            • #10
              Random caveats are plentiful.

              For one holidays always fuck up statistics. So the end of the year number on Jan 1st is likely 20-30k off. They might fix it after a few weeks.

              Better therapeutics getting rolled out.

              Better than expected results from 3rd shot.

              Weaker mutations. The virus "code" could hit a hard wall or significantly reduce fatality in favor of personal survival.

              Our immune system could adapt better/faster.

              All of the above is more likely for Mar 1st than Jan 1st. A lot more likely.

              Comment


              • #11
                Ah and the 3 numbers that tell you what will happen in the shortish term.

                Change in cases, hospitalizations and deaths in the last 14 days. For November 12th those were...

                +9% cases, -10% hospitalizations and -18% in deaths

                ...unadjusted for Veterans Day.

                Start of the week the same numbers were around +4,-12 and -20.

                Cases declined till the 25th Oct. The previous peaks were around Sep 1st for cases, Sep 4th for hospitalizations and Sep 20th for deaths by 7-day average adjusted for an anomaly in cases. 7-day avg. doesn't convert directly to 14 day numbers and there's likely a saturation effect, but i doubt the death numbers will decline more than a week or two from now.

                Comment


                • gimmick
                  gimmick commented
                  Editing a comment
                  19th Nov. +30%, +3% and -6%. 1134 deaths 7-day avg. Lowest it was on 17th at 1088.

              • #12
                Originally posted by sonatine View Post
                so another prospective thesis is 'how non-deadly is this virus when people arent recovering in hospitals or able to get regeneron or whatever, but have to roll the dice while laying on their couch for days, shitting and pissing into the cushions every night'
                sonatine out here just laying the groundwork for a future when he's going to feel a need to explain some shitty couch cushions, y'all

                Comment


                • #13
                  NEW YORK GOVERNOR DECLARES 'DISASTER EMERGENCY' OVER RISING COVID-19 INFECTION RATES

                  i hate this place.

                  Comment


                  • #14
                    Around 60 day bump. You dodged/delayed the worst. It took fairly long to get to same daily death averages as the fall Delta numbers. The current Omicron deaths reflect 400k daily infection numbers. The peak cases look like 800k now. Getting that sorted out could double the deaths for a relatively short period,

                    The early estimates of 70% fatality compared Delta would have been really bad. Those were from a 3rd world country with different climate and limited vaccines/therapeutics.

                    The 10% estimate is optimistic as fuck and your numbers don't support that. 30-40% seems likely. The worst should come around 2-3 weeks.

                    I don't remember if mentioned it here or not, but the fall numbers mostly came from retard country. Weather component hits the non retarded parts of US in the winter. Part that actually does something to stop the spread and limit deaths. Swap from Delta to Omicron was significant. Booster numbers keep going up.

                    Regarding herd immunity and Omicron that's never happening. It took over because it's good at evading antibodies. Reinfection rates are highest of any dominant variant. There's always a small window where you could in theory reduce the R number below 1 before it spreads to more susceptible population and then spins back. That requires fast and consistent action. The bare minimum is enough on that front, but it appears to be too much. Avoid doing anything retarded and wear a mask for a few weeks at peak vaccination/infection rate.

                    Part of the herd immunity is usefulness of memory cells. It's not required, but it makes it easy. Covid appears to be too fast for it to make a difference. So you need "fresh" antibodies. Specific antibodies almost always drop off. There's blueprints to make more with memory cells, but mechanism is slow and mutations occasionally evade them completely. Memory cells don't trigger, because they don't recognize the virus.

                    Omicron replicates fast in the upper respiratory tract. That causes infected to be infectious faster and shot gunning larger viral loads. It also increases mutation opportunities since it's partially tied to replication and reproduction (infecting others). Mutation itself is neutralish effect. Sometimes you get Delta and sometimes Omicron. For evolutionary reasons it aims for Omicrons, but the cost of rarer Deltas is steep.

                    14-day numbers that i don't care enough to check for anomalies were, 8% cases, 24% hospitalizations and 41% deaths (Jan 22.). They're roughly in that ballpark. Hospitalizations to ICU ratio has been around 1/6 vs previous 1/4 to 1/5 with earlier strains.

                    Comment


                    • #15
                      Oh and i tried to look up how weather anomalies correlate with the virus, but i didn't get too far. The metric that had the strongest early correlation is called "Surface Atmosphere Temperature Anomaly" and who ever came up with that decided that SATAn is reasonable abbreviation for it.

                      I got thousands of google hits with "SATAn Covid". None of them had anything to do with weather.

                      Comment


                      • sonatine
                        sonatine commented
                        Editing a comment
                        so good
                    Working...
                    X