The Media Is Lying To You About Florida (Again)
The ridiculous, desperate attempts to disparage Ron DeSantis
BREAKING: Media reports misrepresent basic statistical principles about COVID data in a desperate attempt to damage a political opponent.
Are you surprised? You shouldn’t be.
This has been a recurring problem, particularly and most importantly in Florida. It began in large part due to CNN and other mainstream outlets giving fired geographer and alleged felon Rebekah Jones a platform to spread *actual* misinformation about a supposed cover up in Florida’s death data. Desperate to hurt an ideological adversary and willing to ignore the many glaring inaccuracies in her story, the media promoted her lies and spawned an indefinite belief in a disproven conspiracy theory among many in their audience.
Florida’s results were simply not disastrous enough to make the media happy. To confirm their biases, they had to believe in a cover-up. There must be something hidden; some grand conspiratorial explanation for why the state wasn’t leading the world in COVID mortality. Governor Ron DeSantis opened beaches, he opposed mask mandates, he kept schools open, he allowed bars and restaurants to continue operating — in short, he did everything the national experts, worshipped by the media, said not to do. And Florida’s results were no worse than the national average, and in some ways, significantly better.
That undeniable fact has been a thorn in the side of many media outlets who prioritize political opposition above accurate reporting. So when Florida’s seasonal surge arrived in mid-Summer 2021, many in the press went right back to searching for inaccurate conspiracies they could use justify their doubts and appeal to their ideological base.
Which brings us to the most recent example, the Miami Herald, where just the other day reporters invented a controversy they could sell to their readership and political allies: Florida is misleading the public on deaths.
So let’s go through this and highlight how their assertions uncover nothing except their own incompetence, desperation and ignorance.
It’s important to first provide some context around Florida’s increase in metrics. We can do this by showing, once again, that during the fall and winter, when most of the country was experiencing a significant increase in cases, hospitalizations and deaths, Florida remained open and outperformed states that closed businesses, mandated masks with severe punishments for non-compliance, limited capacity or closed dining, forced curfews, kept schools shuttered, and generally followed expert and media consensus.
For example, here’s reported deaths in California and Florida from September 1, 2020-April 1, 2021:
And cases in Florida and New York over the same time frame:
These are important reminders, because it reinforces that Florida has operated with few state mandated restrictions for nearly a year now. Most U.S. states “reopened” months later; for example California didn’t remove its statewide mask mandate and “tier system” until nearly 9 months after Florida largely removed most statewide restrictions. Yet cases in Florida remained comparatively low throughout 2021 and the large increase seen over the past month and a half began within the same time frame as 2020:
Cases started to increase just a few weeks after they did last year, with no major change in statewide policy, no significant behavior change or dramatic, ribbon cutting “reopening.” Similarly, with no major statewide restrictions introduced, cases have declined on their own, a few weeks after they did in 2020.
Which brings us to this most recent “controversy,” drummed up by The Miami Herald:
Florida’s Department of Health started reporting deaths to the CDC by date of death. That’s it. That’s the grand conspiracy. Florida is counting deaths by when individuals actually died, instead of when local county health departments get around to reporting it to the state.
Now, this headline has already been debunked in several places; Phil Kerpen’s excellent opinion piece in the Wall Street Journal, a Twitter thread by DeSantis Press Secretary Christina Pushaw and from a quote in the actual article itself:
“Deaths by date of death curve is the most accurate you can get,” Salemi said. “You know exactly when people died, you know how to construct the curve and exactly when we were experiencing surges in terms of deaths.”
Yes, that’s correct, the Herald is so desperate to criticize Ron DeSantis, they published an article that contains a quote from an epidemiologist debunking the criticism in their own headline.
In any case, it’s important to present just how unremarkable this transition is, as well as how common and most importantly, accurate it is to report deaths by date of death.
One of the main conceits of this latest hit piece is that the change to report by date of death seemingly came out of nowhere. It was a complete shock; an unprecedented move, with the implication it was done to make Florida’s situation look “better”:
As cases ballooned in August, however, the Florida Department of Health changed the way it reported death data to the CDC, giving the appearance of a pandemic in decline, an analysis of Florida data by the Miami Herald and el Nuevo Herald found.
Bizarrely though, some writers and/or editors at the Herald apparently don’t read their own paper. A story published *weeks ago* covered this exact change with a quote from the Florida Department of Health announcing it and the reason behind it:
On Aug. 10, the CDC changed the way it reported new cases and deaths in Florida. Cases and deaths used to be logged as total new cases reported on a single day. Now, Florida is reporting cases by the “case date,” according to the CDC rather than the date the case was logged into the system. The result of this change is a lag in cases by date and a significant number of cases backfilling over time.
DOH spokesperson Weesam Khoury said Florida’s new reporting system “will ensure that continuous epidemiological analyses provide the most updated data to the public.”
The critique that this was some sort of out of the blue change meant to hide the “real” data can easily be proven wrong, by their own paper. The transition in reporting to the CDC was done weeks ago, with an on the record statement by a Department of Health spokesperson explaining it. How is this remotely controversial?
We can also examine a possible reason why the change was made.
As one of, if not the only Governor in command of the actual (lowercase) science (no trademark) — expert Dr. Jay Bhattacharya described him as “extraordinary” in Newsweek — it would make sense that Florida’s COVID data reporting strategy is updated to, in the words of a local epidemiologist, “the most accurate you can get.”
Reporting cases and deaths by date of event instead of date of report generally provides a more accurate picture of the true timeframe of the pandemic. Date of report is frequently subject to data dumps and corrections, and often creates bizarre trends that can be remarkably misleading.
For example, based on the New York Times COVID data, Florida reported -40,527 cases on June 4, 2021. Apparently the Miami Herald believes it’s more accurate to say that 40,000 people had their COVID cases removed in one day. Even using seven day averages can’t fix irregularities like that:
Notice how the curve in early June 2021 disappears under zero for a few days? That’s the effect of reporting -40,527 cases. Imagine if a major policy change had been implemented two weeks prior; trying to parse out potential impacts would be impossible because the data is hopelessly skewed by a one-day aberration.
That’s why reporting by date of event is considered more “accurate.” Sweden famously uses the same standard, and while there is undoubtedly a reporting lag for the most recent days, it creates a more consistent picture over time:
Notice how the curve never drops below zero? That’s because you don’t have validation efforts or reporting dumps skewing the data. And just for the record, I always cut off the most recent week with charts on deaths in Sweden to account for a significant portion of the delay.
Equally bizarre is the Herald’s insistence on complaining about the shift to date of death in Florida, when both options are still available.
Here’s the data by date of report, easily accessible from the New York Times GitHub:
And here’s the CDC version, using date of death:
The shaded bar indicates the past ~10 days, where most additions will be made.
The point is, these are both easily accessible. It took me a few minutes to create these charts. Anyone with a basic knowledge of Excel and the ability to navigate a website could have done it. This “controversy” is apparently due entirely to Florida reporting date of death specifically to the CDC, when the CDC itself told the Herald weeks ago that this was not remotely uncommon:
According to a statement from CDC spokesperson Jasmine Reed on Aug. 18: “Florida’s aggregate case and death data includes case date for cases and date of death for deaths. The method applies to data shared by Florida and to data displayed on COVID Data Tracker. Other States also use this reporting method and states can vary in the reporting method. For example, data as of the date that states submit may be the date that a state received its data from its reporting entities, or it might be another dating method that the state prefers.”
It beggars belief that a newspaper could have multiple quotes on the record explaining the change, the reason for the change, and confirmation from the public health agency responsible for aggregating the data that this is not a unique or inaccurate practice, and STILL publish a story claiming it was misleading. Not to mention the bizarre fascination with the CDC’s presentation specifically, when again, the numbers by date of report are still widely available. The New York Times has it. Johns Hopkins has it. It’s insanity.
It’s also necessary to highlight how common it is throughout the country to report deaths by date of death. For example, California, does the exact same thing on their own tracking site, yet media darling Gavin Newsom has never faced this level of scrutiny over his state’s data reporting. Even though, as you’ll see, the differences in California could easily lend themselves to media outrage.
Here’s their curve by report date:
See how the seven day average drops below zero in mid-August? That’s because California removed 352 deaths from the overall totals on August 10th. We’ll come back to that, but first, here’s the curve by actual date of death:
Boy, looks a little different doesn’t it? That’s because it’s dramatically different. Using the more “accurate” date of death method shows that 352 people in California with COVID didn’t suddenly spring back to life on August 10th; in fact the state has counted 57 deaths so far on that date.
More importantly, the peak in California during the fall and winter surge is significantly higher when using date of death. By reporting date, the state’s 7-day average peak was on January 31, 2021, at 1.37 daily deaths per 100k. By date of death, the 7-day average peak was on January 11th, at 1.74 daily deaths per 100k. Using date of report in California artificially lowers the peak.
This startling difference could be due to nefarious and misleading executive overreach, or it could be something as simple as county data delays, holiday reporting, backlogs, data validation, or any combination of those factors.
So does that difference mean California was misleading the public and “undercounting” COVID deaths to artificially lower the peak during their winter surge? If you’re the Miami Herald and you don’t like the Governor, that’s exactly what it means. In reality, it’s almost certainly one of the many quirks of COVID data analyzation.
And if you were paying close attention above, a few things might have been immediately noticeable: in Florida to this point, the peak of deaths by date of death is actually lower than it is by date of report. It’s the opposite of California. By report date, the peak is right now, at 1.22 daily new deaths per 100k. By date of death, it was back on August 5, 2020 at 1.05 new deaths per 100k. Recent numbers might adjust up to reach that peak or exceed it, but it reinforces how wildly variable COVID data reporting is.
If you use report date to compare deaths in California and Florida, as I did above, it understates just how significantly Florida out performed California during the winter surge. When using the more “accurate” date of death data, California’s winter peak reached population adjusted levels 66% higher than Florida’s EVER reached.
And even more hilariously, California also uses date of death in their reporting to the CDC:
That’s right. This entire controversy, this whole nonsensical article, was based on Florida reporting their data to the CDC in exactly the same method as California.
I guess Newsom also cooked up a dastardly plot to make Californians think deaths were declining recently when they weren’t! Bizarrely, the Los Angeles Times hasn’t yet dramatically revealed their exclusive investigation and analysis uncovering the state’s efforts to mislead and misrepresent. Wonder why!
The media obsession with misinformation about Florida’s COVID data is staggering in its totality and commitment to inaccuracy. Just look at this absurdity from MSNBC with regards to a complete non-story:
This is utter nonsense. There is no “cover-up.” There is no “explosive new report.” There is only politically motivated drivel; a desperate attempt to hurt Ron DeSantis by what once could charitably be described as a “newspaper.”
They simply cannot accept that Florida’s numbers aren't exceptionally bad. Through the end of 2020, per the CDC, Florida had the 10th lowest age-adjusted COVID mortality rate in the United States. California was higher. New Jersey was 149% higher.
Florida allowed children to live relatively normal lives with schools open, youth sports, and social contact. And with all of that normalcy, only 27 out of 3,524 total pediatric deaths in the state from January 1, 2020 to August 28, 2021 were “with” COVID. New York State had 32 “with” COVID. In California, 0.79% of all pediatric deaths in that same time frame were “with” COVID. Florida’s rate is 0.77%.
The repeated attempts to frame Florida’s numbers in the worst possible light are getting progressively worse. This latest effort from the Miami Herald is an atrocious, malicious, purposeful misrepresentation. It ignores their own previous reporting pointing out the change. It implies that Florida is a dangerous outlier compared to other states, when California reports data *the exact same way*. Quotes from an epidemiologist, from the Department of Health and from the literal CDC point out when and why this change was made.
It’s completely absurd, it’s pointless, it’s vapid, incompetent and ignorant. More simply, it’s a media “report” about COVID and Florida. Of course it’s a lie.