Wisconsin Covid-19 Trends

This graph is for those who believe reports of negative outcomes Wisconsin are “fake news”.

Figure 1: Wisconsin current Covid-19 hospitalizations (blue, left log scale), fatalities as of date reported (red, right scale). Source: Covid Tracking Project, accessed 10/14/2020.

From Wisconsin State Journal (10/13/2020):

Citing inaction by the state Legislature, a St. Croix County judge on Monday rejected a request by a conservative legal group for a temporary injunction against Gov. Tony Evers’ statewide mask mandate.

The GOP-led Legislature has met once since the pandemic began, in April, to take up legislation in response to COVID-19. Lawmakers and Evers ultimately signed what officials on both sides of the aisle described as “imperfect” legislation.

The package aimed to complement federal aid allocated to the state, included additional Medicaid funding, allowed for increased unemployment benefits from the federal government and waived a one-week waiting period for benefits until February. The bill also allowed the state budget committee to spend up to $75 million during the public health emergency on coronavirus-related needs, though the committee has not exercised that authority.

Vos and Fitzgerald did not respond to requests for comment on Monday.

Johnson, who tested positive for COVID-19 earlier this month, said he is opposed to the governor’s mask mandate.


30 thoughts on “Wisconsin Covid-19 Trends

  1. macroduck

    Three months ago, 35 states had transmission rates below 1. Now, only ten states do. A transmission rate below 1 would, over time, end the pandemic. A transmission rate above one widens the pandemic.

    Wisconsin, at 1.04, is still worsening, though at a slower rate than late August and early September. And Wisconsin has a lower transmission rate than over half of U.S. states: https://rt.live/.

      1. macroduck

        Yeah. If you dial back 3 months, the transmission rate is lower and the ranges of estimates is narrower. Troubling, but not surprising.

        1. pgl

          Yea – I’m getting tired of our religious zealots demanding the alleged 1st Amendment right not to wear a damn mask.

  2. Ivan

    A tragedy and a natural outcome when you have GOP in charge of any branch of government. I hope the voters in your state will learn from their mistakes and vote blue no matter who. Unfortunately a lot of them will probably just adopt the next GOP con – herd immunity anybody (you know like the Swedish disaster) ?

    1. Ivan

      If this continues to get much worse, an interesting question is how many voters will want to risk their lives to go vote for Trump on election day. He has been betting on a win at the in-person voting sites, and then fighting the mail-in votes in his stacked supreme court. But what if the loses also on the counts of in-person voting.

  3. Willie

    It’s not looking great in Washington State, either. Inslee has done a reasonably job of managing the pandemic here, and he doesn’t have a bunch of ideologues on the courts screwing it up. But, it’s still looking shaky. This pandemic is not going to magically go away.

  4. pgl

    Covid-19 hospitalizations and deaths have both jumped. But wait for it – I sense Bruce Hall’s keyboard typing right now. He will tell us once again he sees no correlation!

  5. The Rage

    Going to fall/winter was always going to accelerate cases, hospitalization and increase deaths. At 65000, that isn’t that bad. If it doubles, then it gets bad. UV has collapsed, weather is getting colder and viral loads are stronger.

    Don’t be alarmed by rising cases ie I don’t see Washington states rates as overly bad. Now the plains….

  6. Ooe

    paging Dr. Death aka Dr Redfield of the Center for Lack of Disease Control. He claimed that thing would be better. not.

      1. oee

        Your adminit attacks are not worthy of an educated person. You may disagree that CDC should do its job, but to F@ck myself is not an argument.

  7. 2slugbaits

    FWIW, I calculated the Wisconsin positivity rate using daily data from 16 Mar thru Sep 15 and then estimated a SARIMA model (0,1,1 1,0,1) w/o constant. The SARIMA model was selected based on minimizing the information criterion. All criteria selected the same model. The out-of-sample forecast showed a MAPE of 13.6%. The Theil’s U value is 0.88, so the forecast model is better than guessing. The forecast results are not encouraging if you’re a Cheese Head.

  8. Bruce Hall

    NPR shows this: Total Cases And Deaths, State By State
    Data as of 7:24 a.m. on Oct. 15


    Lot’s of cases; more than Michigan.
    Not so many deaths; a little more than 1/5 of Michigan’s with a per capita death toll of 1/3 Michigan’s. Michigan… isn’t that the state where the Gov. destroyed the small business economy to keep everyone safe?

    What’s in those cases anyway? Are they largely empty?

    1. pgl

      Gee Bruce – NPR drew the 7-day daily death count and you forgot to mention the fact that this statistic stands at 985 as of Oct. 14 but was only 913 3 weeks ago. Since you continue to tell us that deaths are declining I guess it figures that a Trumpian liar like you would not mention this inconvenient fact.

      I compared per capita new cases in NY to Michigan and it seems your state is 2X as compared to my state. Continue hiding in your basement. I’m sure your mom does not mind risking her life so you can have food.

  9. Moses Herzog

    Nerd numbers and a Cheese Head reference. This is a 5-star comment and one David Lee Roth reference away from a historical moment.

  10. sammy

    This graph is an example of “fake news.”

    The unit measurement for cases goes from 0-1000, while the deaths goes from 0-20. This makes it look like cases = deaths. If the measures were proportional you would see a really high cases line and a really small, almost imperceptible, barely moving, death line.

    1. Menzie Chinn Post author

      sammy: Let me get this straight. You are against all cross-plots? Graph legend clearly indicates *in the graph* and the notes to the graph indicate the different scales. You are afraid people are idiots/innumerate/illiterate? Or are you projecting?

      1. Dr. Dysmalist


        Always the latter. With this bunch, every accusation is a confession.

        I’ll bet he doesn’t understand the utility of the log scale with both series in one graph, either. IIRC he complained about that once, too.

        Personally, I find it very useful. Thank you for this.

      2. pgl

        You have to endure Bruce Hall whining about not seeing any correlation (even as everyone else does) so when you make spotting correlations easier for these Mr. Magoo wannabes his good buddy Sammy goes off. I guess you can’t win.

      3. Macroduck

        Neither. sammy (once again) typed up a word salad with the sole motive of tossing dust in the air. Any excuse to question facts that make his masters look bad. sammy has neither the depth to project nor the vocabulary fear innumeracy.

    2. baffling

      sammy, let me explain something to your little mind. the point of the graph is to illustrate how hospitalizations and deaths are very closely related. not long after hospitalizations (and not simply cases, as you indicate) rise, so do deaths. from your approach, this would never be seen.

      “If the measures were proportional you would see a really high cases line and a really small, almost imperceptible, barely moving, death line.”
      that barely moving line represents the deaths of approximately 2% of those hospitalized. you may think that is trivial, sammy, but it is not.

      1. pgl

        I thought most people knew that not all cases translate into deaths. I thought most people have been following the fact that getting this virus still gives one a probability of dying that is considerably below 100%.

        But I guess Sammy is so incredibly stupid – he has not figured this out by now.

      2. pgl

        I just checked with BEA to see what share of GDP comes from our motor vehicle sector and it appears to be less than 3%.

        Let’s give Sammy a small change. Show a graph that tells us the correlation between GDP and motor vehicle value added (I bet there is one). His Michigan buddy Bruce Hall might enjoy this.

        But one caveat – Sammy is not allowed to scale these two series in any other way than what he demands our host to do.

        This should be fun!

      3. Macroduck

        Putting a series with a range in the hundreds or thousands on a chart scaled for tens of thousand is a classic way of lying with statistics. sammy is demanding that Menzie do sammy’s lying for him

    3. pgl

      “This makes it look like cases = deaths.”

      No it does not. Are you really this damn stupid or are just being your usual angry argumentative self.

      The graph was labeled clearly but I do know MAGA hat wearing fools like you have trouble reading even well presented graphs.

    4. Macroduck

      The 1954 classic “How to Lie with Statistics” (jttps://en.m.wikipedia.org/wiki/How_to_Lie_with_Statistics) warns that using a chart with an inappropriately large scale will dishonestly mask variation in a time series. sammy is here demanding that Menzie lie on sammy’s behalf by putting Covid deaths on a scale that is an order of magnitude too large. By employing a left and a right scale, Menzie avoids the lie that sammy’s masters want told.

      1. pgl

        Darrell Huff’s 1954 How to Lie with Statistics was a gem but note that was not the only thing he wrote:


        ‘Darrell Huff, author of the wildly popular (and aptly named) How to Lie With Statistics, was paid to testify before Congress in the 1950s and then again in the 1960s, with the assigned task of ridiculing any notion of a cigarette-disease link. On March 22, 1965, Huff testified at hearings on cigarette labeling and advertising, accusing the recent Surgeon General’s report of myriad failures and “fallacies.” Huff peppered his attack with with amusing asides and anecdotes, lampooning spurious correlations like that between the size of Dutch families and the number of storks nesting on rooftops–which proves not that storks bring babies but rather that people with large families tend to have larger houses (which therefore attract more storks).’

        Huff also wrote How to Lie with Smoking Statistics which was funded by the tobacco industry. This blog post details the back and forths on whether it should be published, which it was not, The blog post concludes:

        ‘On the other hand, this document makes me think that Huff may have seen his role as producing talking points in support of a predetermined conclusion. I guess we’ll never know if he really wanted to publish How to Lie with Smoking Statistics. Maybe he intentionally sabotaged it because he sensed it would ruin his reputation, whereas it was possible for him to keep the consulting and testimony under the radar.’

        All this makes me wonder whether our Usual Suspects are taking lessons from Huff in terms of How to actually lie on behalf of Trump’s anti-science campaign. The book should be entitled How to Lie with COVID-19 Statistics.

  11. samir sardana

    The Economist doubts the COVID success of Pakistan !


    Indians cannot bear the pain of the success of Pakistan and Imran Khan.These Indian Parasites have proliferated in the Global Media – as Indian Corporates and the Indian State, sprinkle their ads,in the Global Media.

    The Right approach of the Economist,should have been to start from the Number of Tests per Million – which is just 16000 in Pakistan,vs around 65000 in India.That has some basis – but is also specious,as it is 30% higher than that of Indon-ass-eeah,which has a population, 20% HIGHER than Pakistan,and is next to PRC – with millions of Chinese in
    Jakarta and islands of that nation.dindooohindoo

    The Pakistani tests are also 30R higher (on a per million basis) w.r.t Nippon – which has a large population of Japanese in PRC,and also,has a very high frequency of air travel with the PRC.Several flights from Shang and Beijing,transit via Nippon.So the “Low Aggregate Tests” Theory of Pakistan,is busted.The Marginal Cost of each RTPCR test,is USD 10 and Pakistan cannot waste its funds on that

    Let us ASSUME,still, that the Pakistania are NOT coming for testing,AND that the Pakistani state is NOT testing.Therefore,the people are NOT showing symptoms – as else,by now,there would be a sharp SPIKE in deaths (young and old).There is 1 brilliant innovation by the Pakistani state – in that,they are testing the dead.AT the EOD,even if patients are not tested,AND THEY DIE – the virus would be in the dead body.

    The death rates in Pakistan,in general,are normal, and COVID rates in the dead,are also normal (w.r.t the age and morbidities of the dead) – so the STRATEGY of NOT doing random testing,TO AVOID scare mongering,has worked,as the bet is on the immunity of the Pakistanis, and a close monitoring of the generic death rates – with their COVID content.If it lasts the winter – then this could be a Harvard Case Study.

    India on the other hand,has a TEST COUNT of 60000 per Million.For the size of the Indian Pop,and the Intl flights transiting via India,in the pre Covid tenor,and the Tourists in India,in the pre-COVID tenor,this statistic is pathetic.Brazilw,ith a sixth of India’s Population,has 85000 tests per million ,and Chuna has 100000 plus

    The worst HIT nations are France and UK,with a DAILY CASE COUNT OF 18000,and a population of 65 million – which is 5% of the Indian population – and so ON A PER CAPITA BASIS, IT IS TYHE HIGHEST IN THE WORLD – far ahead of USA and Brazil.

    France sold the Rafael to the Indians,and UK has millions of Indians (who are also,the worst hit group)

Comments are closed.