# Is the late-July flattening of the US curve real?

**URL:** <https://boards.straightdope.com/t/is-the-late-july-flattening-of-the-us-curve-real/916059>\
**Category:** The Quarantine Zone\
**Tags:** covid-quarantine\
**Created:** [July 23, 2020, 1:33am UTC](https://boards.straightdope.com/t/is-the-late-july-flattening-of-the-us-curve-real/916059 "2020-07-23T01:33:14Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![illoe](https://avatars.discourse-cdn.com/v4/letter/i/a3d4f5/32.png) [@illoe](https://boards.straightdope.com/u/illoe)\
**Post date:** [July 23, 2020, 1:33am UTC](https://boards.straightdope.com/t/is-the-late-july-flattening-of-the-us-curve-real/916059/1 "2020-07-23T01:33:15Z")

</div>

I’ve been following the US case trackers pretty closely, and have been a little surprised at the speed with which the US curve has flattened since the reporting protocols were changed to bypass the CDC. Florida in particular has turned right around.

Aside from waiting a few weeks to see if Florida turns into a nice dubious bell curve\* or rolls off gently like all other places have done, does anyone have any insight into whether this flattening is real?

_one of my favorite apocryphal statistical stories is the origin of Benford’s law arising from a close examination of accounting records, where too many leading 9s revealed a crook._\*

\*\*Not to imply in any way that we’re dealing with crooks in this case.

---

<div class="post-metadata">

**Author:** ![elfkin477](https://avatars.discourse-cdn.com/v4/letter/e/a9a28c/32.png) [@elfkin477](https://boards.straightdope.com/u/elfkin477)\
**Post date:** [July 23, 2020, 1:51am UTC](https://boards.straightdope.com/t/is-the-late-july-flattening-of-the-us-curve-real/916059/2 "2020-07-23T01:51:54Z")

</div>

What? The US recorded the most cases ever three days ago. Are you talking about cases going down for two days and asking if that’s statistically meaningful?

---

<div class="post-metadata">

**Author:** ![Si\_Amigo](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/si_amigo/32/2877_2.png) [@Si\_Amigo](https://boards.straightdope.com/u/Si_Amigo)\
**Post date:** [July 23, 2020, 1:57am UTC](https://boards.straightdope.com/t/is-the-late-july-flattening-of-the-us-curve-real/916059/3 "2020-07-23T01:57:24Z")

</div>

Even Trump admitted yesterday it’s going to get worst before it gets better. Shady.

---

<div class="post-metadata">

**Author:** ![illoe](https://avatars.discourse-cdn.com/v4/letter/i/a3d4f5/32.png) [@illoe](https://boards.straightdope.com/u/illoe)\
**Post date:** [July 23, 2020, 2:05am UTC](https://boards.straightdope.com/t/is-the-late-july-flattening-of-the-us-curve-real/916059/4 "2020-07-23T02:05:13Z")

</div>

Maybe it’s just the NYTimes?

As of today their running averages have been flattening out for 5 or 6 days, now — long enough to see a qualitative change from what had appeared to be pretty strong growth. Is there a better data source?

---

<div class="post-metadata">

**Author:** ![Snowboarder\_Bo](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/snowboarder_bo/32/229_2.png) [@Snowboarder\_Bo](https://boards.straightdope.com/u/Snowboarder_Bo)\
**Post date:** [July 23, 2020, 3:03pm UTC](https://boards.straightdope.com/t/is-the-late-july-flattening-of-the-us-curve-real/916059/5 "2020-07-23T15:03:12Z")

</div>

> **[COVID - Coronavirus Statistics - Worldometer](https://www.worldometers.info/coronavirus/)**
>
> Daily and weekly updated statistics tracking the number of COVID-19 cases, recovered, and deaths. Historical data with cumulative charts, graphs, and updates.

> **[United States COVID - Coronavirus Statistics - Worldometer](https://www.worldometers.info/coronavirus/country/us/)**
>
> United States Coronavirus update with statistics and graphs: total and new cases, deaths per day, mortality and recovery rates, current active cases, recoveries, trends and timeline.

---

<div class="post-metadata">

**Author:** ![Machine\_Elf](https://avatars.discourse-cdn.com/v4/letter/m/82dd89/32.png) [@Machine\_Elf](https://boards.straightdope.com/u/Machine_Elf)\
**Post date:** [July 23, 2020, 3:50pm UTC](https://boards.straightdope.com/t/is-the-late-july-flattening-of-the-us-curve-real/916059/6 "2020-07-23T15:50:41Z")

</div>

> [@illoe](#):
>
> _one of my favorite apocryphal statistical stories is the origin of Benford’s law arising from a close examination of accounting records, where too many leading 9s revealed a crook._ \*

**[That’s not where it originated](https://en.wikipedia.org/wiki/Benford%27s_law)** - but it is indeed a useful tool in the IRS tax fraud detection arsenal.

---

<div class="post-metadata">

**Author:** ![Monocracy](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/monocracy/32/4493_2.png) [@Monocracy](https://boards.straightdope.com/u/Monocracy)\
**Post date:** [July 23, 2020, 6:12pm UTC](https://boards.straightdope.com/t/is-the-late-july-flattening-of-the-us-curve-real/916059/7 "2020-07-23T18:12:51Z")

</div>

> [@elfkin477](#):
>
> What? The US recorded the most cases ever three days ago. Are you talking about cases going down for two days and asking if that’s statistically meaningful?

The 7 day moving average had been climbing an average of 1500 cases per day for just over a month up until that day you mentioned as the most cases ever. In the five days since, it’s risen less than 1000 total.

I don’t think 5 days is enough to declare a trend, but it is a little suspicious that it happened at the same time as the new reporting protocols. We’ll have to wait and see.
