Home› Tug Tavern

Howdy, Stranger!

It looks like you're new here. Sign in or register to get started.

Sign In Register
Welcome to the Hardcore Husky Forums. Folks who are well-known in Cyberland and not that dumb.

Captain Obvious Here - The problem with Inslee's Metrics for Phases...

DoogieMcDoogersonDoogieMcDoogerson Member Posts: 2,488
in Tug Tavern
I am beginning to think this is by design. People who are scientists and mathematicians wouldn't choose these metrics.

Metric 1: Rate per 100,000 population of newly diagnosed cases in past 2 weeks
========================================================
Gee, I wonder if the # of people we're testing is an important variable to consider if you're watching trends. I guess not since it's never mentioned. My math and comp sci background says it should.

Metric 2: Number of individuals tested for each new case in prior week
========================================================
Let's have a metric based on flawed metric #1 above.

Metric 3: Percentage of individuals testing positive during the past week
========================================================
Gee, I wonder if the percentage of cases will go UP when it's warm and flu season is not running concurrently?

Metric 4: Percentage of licensed beds occupied by patients
========================================================
Shit metric. Meaningless. Kick out elective procedures if you don't have room.

Metric 5: Percentage of licensed beds occupied by COVID-19 patients
========================================================
Reasonable metric. Only good one in the bunch.

What the metrics should be?

Probably just need 1 metric - # hospital admissions / 100K population. Knowing how that trends is likely all that matters. Deaths likely correlate with this metric.

Pierce County used to show that on the website but now they've switch to State-provided charts.

Obviously, this is very Grim:



Since the #s don't work, the press talks about "Highest # of cases since march" "Cases explode in Pierce County" to keep the Karens all worked up.



· Share on FacebookShare on Twitter
«1

Comments

  • RaceBannonRaceBannon Member, Swaye's Wigwam Posts: 104,543 Founders Club
    DoogieMcDoogerson said:

    I am beginning to think this is by design. People who are scientists and mathematicians wouldn't choose these metrics.

    Metric 1: Rate per 100,000 population of newly diagnosed cases in past 2 weeks
    ========================================================
    Gee, I wonder if the # of people we're testing is an important variable to consider if you're watching trends. I guess not since it's never mentioned. My math and comp sci background says it should.

    Metric 2: Number of individuals tested for each new case in prior week
    ========================================================
    Let's have a metric based on flawed metric #1 above.

    Metric 3: Percentage of individuals testing positive during the past week
    ========================================================
    Gee, I wonder if the percentage of cases will go UP when it's warm and flu season is not running concurrently?

    Metric 4: Percentage of licensed beds occupied by patients
    ========================================================
    Shit metric. Meaningless. Kick out elective procedures if you don't have room.

    Metric 5: Percentage of licensed beds occupied by COVID-19 patients
    ========================================================
    Reasonable metric. Only good one in the bunch.

    What the metrics should be?

    Probably just need 1 metric - # hospital admissions / 100K population. Knowing how that trends is likely all that matters. Deaths likely correlate with this metric.

    Pierce County used to show that on the website but now they've switch to State-provided charts.

    Obviously, this is very Grim:



    Since the #s don't work, the press talks about "Highest # of cases since march" "Cases explode in Pierce County" to keep the Karens all worked up.



    @BearsWiin ? True?
    · Share on FacebookShare on Twitter
  • RaceBannonRaceBannon Member, Swaye's Wigwam Posts: 104,543 Founders Club
    GrundleStiltzkin said:

    How long will it take for everyone other than Daomoné & me to realize positive case <> illness.


    Wash you’re hand, Washuhntonians

    Uhhh
    · Share on FacebookShare on Twitter
  • PurpleThrobberPurpleThrobber Member Posts: 43,596 Standard Supporter
    edited June 2020
    Probably just need 1 metric - # hospital admissions / 100K population. Knowing how that trends is likely all that matters. Deaths likely correlate with this metric.

    The Throbber has been banging this drum since the beginning. Hospital occupancy rate is all that matters.

    Don't even get me started on the absurdity of death with COVID stats being sold as death from COVID.

    Bunch of Vernon Adams clones fucking with the maff here.



    · Share on FacebookShare on Twitter
  • DerekJohnsonDerekJohnson Administrator, Swaye's Wigwam Posts: 62,422 Founders Club
    Today's headline in The Seattle Times was written last week I'm sure:

    "EXPERTS EXPECT VIRUS DEATH TOLL TO RISE"

    "GRIM DATA"

    But of course... nothing has happened! It's all speculation. It's not just fake news, it's speculation that's turned into news.
    · Share on FacebookShare on Twitter
  • theknowledgetheknowledge Member, Swaye's Wigwam Posts: 4,799 Founders Club
    I like that the graph the Seattle Times uses to show death are always a 0-current nimbler graph when a graph showing daily deaths statewide or nationally are striking in their slope downward. One makes it ;look like the corona death will never stop and the other? Not so much.
    · Share on FacebookShare on Twitter
  • PostGameOrangeSlicesPostGameOrangeSlices Member, Swaye's Wigwam Posts: 25,500 Swaye's Wigwam
    Liberals have never been known for their math or stats prowess
    · Share on FacebookShare on Twitter
  • GreenRiverGatorzGreenRiverGatorz Member Posts: 10,163
    Q for the stats gurus - where's the best source for hospitalization stats? Worldometer does a good job of tracking and visualizing death trends, but hospitalization is tougher to find.
    · Share on FacebookShare on Twitter
  • FireCohenFireCohen Member Posts: 21,823
    PostGameOrangeSlices said:

    Liberals have never been known for their math or stats prowess

    I think it applies to US in general now
    · Share on FacebookShare on Twitter
  • RaceBannonRaceBannon Member, Swaye's Wigwam Posts: 104,543 Founders Club
    GreenRiverGatorz said:

    Q for the stats gurus - where's the best source for hospitalization stats? Worldometer does a good job of tracking and visualizing death trends, but hospitalization is tougher to find.

    I tried to find out how many beds in Riverside County and failed miserably
    · Share on FacebookShare on Twitter
  • PostGameOrangeSlicesPostGameOrangeSlices Member, Swaye's Wigwam Posts: 25,500 Swaye's Wigwam
    FireCohen said:

    PostGameOrangeSlices said:

    Liberals have never been known for their math or stats prowess

    I think it applies to US in general now
    You're not wrong
    · Share on FacebookShare on Twitter
  • Bob_CBob_C Member, Swaye's Wigwam Posts: 10,240 Swaye's Wigwam
    Jay’s a data guy. He took Stat 311 back in the 70’s to earn that Econ degree. Probably a lot of power BI and Tableau was used.

    I’ve seen these people in business a lot. There are a lot of posers that claim to be about data and remind you over and over that they are data guys. They use it for confirmation bias occasionally. Anyone who repeatedly tells you they are about the data, isn’t about the data. Leading with data shows you are about the data.

    Jay’s a poser.
    · Share on FacebookShare on Twitter
  • dncdnc Member Posts: 56,614
    Bob_C said:

    Jay’s a data guy. He took Stat 311 back in the 70’s to earn that Econ degree. Probably a lot of power BI and Tableau was used.

    I’ve seen these people in business a lot. There are a lot of posers that claim to be about data and remind you over and over that they are data guys. They use it for confirmation bias occasionally. Anyone who repeatedly tells you they are about the data, isn’t about the data. Leading with data shows you are about the data.

    Jay’s a poser.

    Silent Bob_C > Jay
    · Share on FacebookShare on Twitter
  • creepycougcreepycoug Member Posts: 22,976
    GrundleStiltzkin said:

    How long will it take for everyone other than Daomoné & me to realize positive case <> illness.


    Wash you’re hand, Washuhntonians

    Which hand?
    · Share on FacebookShare on Twitter
  • alumni94alumni94 Member Posts: 4,858
    Remember, cases is not equal to people that tested positive. One person can test numerous times, each counting as a case.
    · Share on FacebookShare on Twitter
  • GreenRiverGatorzGreenRiverGatorz Member Posts: 10,163
    alumni94 said:

    Remember, cases is not equal to people that tested positive. One person can test numerous times, each counting as a case.

    I would guess that's an insignificant number. Moot point anyways since cases shouldn't be used to establish anything involving the trends/progression of the virus. Hospitalizations and deaths are all that matter.
    · Share on FacebookShare on Twitter
  • alumni94alumni94 Member Posts: 4,858
    GreenRiverGatorz said:

    alumni94 said:

    Remember, cases is not equal to people that tested positive. One person can test numerous times, each counting as a case.

    I would guess that's an insignificant number. Moot point anyways since cases shouldn't be used to establish anything involving the trends/progression of the virus. Hospitalizations and deaths are all that matter.
    I agree, but all we hear about is the cases. The cases are driving the decisions on policy and fear. If there is a way to show that cases are a poor, inaccurate way to calculate, then maybe...just maybe, people will look at data that is more realistic.
    · Share on FacebookShare on Twitter
Sign In or Register to comment.