Beware The Flood Of Trump Sentencing Disinformation

I just turned 54 and I’ve grown used to things hurting — lots of things hurting — first thing in the morning. However, I am unaccustomed to awakening only to be kicked right in the junk by the Washington Post before my morning coffee.

I refer to an appallingly bad Washington Post column about the hypothetical sentence Donald Trump faces based on his (to date) three pending criminal cases. When I awoke on Friday, about two dozen gleeful people had sent it to me in some sort of cruel exercise of modern bearbaiting.

The Washington Post column is inexcusable. Here’s why.

How Federal Sentencing Actually Works

In our recent Very Special Episode of Serious Trouble about how to be an informed and critical consumer of legal news, Josh Barro and I spent a long time discussing one of the most important things the media gets badly wrong: it reports the statutory maximum sentences defendants face as if those numbers had anything to do with the actual sentences the defendants face. They almost never do. Adding up all the statutory maximum sentences for all the counts charged in an indictment is an exercise in clickbait disinformation. I’ve been ranting about this for more than a decade but we are still here.

Here’s how federal sentencing works. The United States Probation Office, with the input of the parties, proposes a recommended sentence calculated using the United States Sentencing Guidelines. The Guidelines are extremely complex; calculating a sentence under them can be comparable to filling out a business’ tax return. Using the Guidelines yields two key numbers: a Criminal History Category (a number 1 through 6 representing the seriousness of the defendant’s criminal record) and an Offense Level (a number 1 through 43 representing the seriousness of the crimes of conviction). The parties can then object to the Probation Office’s calculations and make arguments to the court about whether and how the Probation Office got it wrong. Once the judge determines the Criminal History Category and Offense Level, he or she applies them to the Sentencing Table to generate a sentencing range in months:

In the old days, the resulting sentencing range was mandatory and bound the court; in 2005 the Supreme Court ruled that the Guidelines could only be treated as a recommendation, not a mandate. So now federal judges treat the Guidelines sentence as a recommendation — a starting point — and sentencing defendants based on mandatory factors set forth in federal statutes. 

What’s the result? A sentence — especially in white collar cases — is almost always below the statutory maximum. In cases where a defendant has been convicted of multiple counts and the statutory maximum is high, the sentence is almost always far below the statutory maximum. White collar criminals only get sentenced close to the statutory maximum when (1) they are only convicted of one or two counts with very low statutory maximum (say, one count of a crime with a five-year statutory maximum, as part of a plea bargain), or (2) cases involving truly extraordinary amounts of money. That’s why you repeatedly see white collar criminals being sentences to just a small fraction of the statutory maximum sentence they face.

How The Washington Post Blew It

The Washington Post’s deeply irritating column is very nominally about telling you that actual sentences and maximum sentences are different, but it does it in a way almost certain to mislead and misinform, and provides no useful information of substance.

First, the core of the Post’s article is a splashy infographic meant to act as Resistance porn. It adds up, on a Candyland-style path, the maximum sentence on each count in the New York state indictment against Trump, the Florida federal indictment for wrongfully retaining documents, and the new Washington D.C. indictment for his January 6 fraud. The infographic is plainly the beating clickbait heart of the piece. It’s also completely useless. The combined maximum sentences from the three cases is a meaningless, masturbatory number. It has nothing to do with what federal sentences he faces, as I note above, and New York criminal law experts (and I am not one) say that the maximum sentences in that case have nothing to do with the plausible sentence he faces there. 

Second, when the Post gets around to contradicting its own splashy infographic, its effort is desultory. The column says:

Well, no. That’s not the first reason the number is wrong at all. The number of counts has very little impact on federal sentences (or, in this context, on New York sentences). Sentences are typically driven by what the Guidelines call relevant conduct — the universe of facts underlying the conviction. Most of the time in white collar cases that universe is governed by the amount of money involved in the crime. A defendant who is convicted of two counts of wire fraud and a defendant convicted of ten counts of wire fraud will get very similar sentences if both cases involve a million dollars in loss, for example. Most of the time the number of counts is puffery by the government. That’s certainly the case in New York — where the prosecution’s 34 counts of doing the same thing is performative and non-substantive — and in Florida, where Jack Smith’s charging 32 counts of wrongfully retaining documents is mostly performative and about making sure he can present all of those documents at trial, not about driving a sentence. The Post’s “first” point therefore misses the point.

But the Post isn’t done yet:

This is hyper-technically true but incredibly misleading and uninformative. First, it’s “unlikely” that Trump will get the maximum statutory sentence in the same sense that it’s “unlikely” that I will win a billion dollars in the lottery this week. It’s absolutely not going to happen and nobody who knows the system thinks there is any chance it will. People convicted of crimes don’t “often” receive lighter sentences than the maximum, they almost inevitably receive lighter sentences, especially in white collar cases involving multiple counts. Moreover, this utterly useless paragraph provides the reader with no information whatsoever about why defendants receive lower sentences, or the mechanism by which they are sentenced. It doesn’t even mention the Guidelines or any other term or concept a curious reader could use to puzzle out what’s going on. The implication most readers would take from the article is that judges are just lenient.

The column’s reference to concurrent and consecutive sentences is also extremely misleading. Federal judges choose a sentence under the process I described above and then make the sentence consecutive only when it’s necessary to accommodate that sentence. So, for instance, if a judge imposed a seven-year sentence in a case where the defendant was convicted of five counts each with a five-year maximum sentence, the judge would impose a five-year sentence on one count and a two-year sentence on another to run consecutively, thus accommodating the sentence. As the Guidelines put it:

The Post also fails to give us any useful information about how judges decide whether to make sentences consecutive or concurrent with sentences in other cases — as Trump could face here — even though there are guidelines on that too.

What’s particularly appalling here is that the Post’s column purports to be about dispelling the notion that Trump faces hundreds of years in jail, and accusing him of touting that idea to fundraise. But it offers no useful information about how or why Trump’s sentence will be lower. It makes that point, as Bluesky user George Harris pointed out to me, only after “privileging the lie” — describing the false story first, for multiple paragraphs, with an eye-catching infographic, only to half-ass the rebuttal later in the story.

The Post also ignores that you can make informed, fact-based estimates of what Trump’s sentence will be. My friend and colleague Mitch Epner did a careful Guidelines analysis of the D.C. case. The Guideline range he calculates would only be a recommendation to the judge, and in white collar cases judges often go below the Guidelines, but they very rarely go higher, to Mitch’s calculations are a sensible estimate of the highest sentence we’re likely to see. There are a number of good, fact-based, expertise-based Guidelines estimates of the D.C. and Florida cases out there, and a number of good expertise-based estimates for the New York case. All of them help inform and educate a reader. The Post doesn’t discuss or link to any of them.

As Mitch pointed out to me, this is all especially galling because the author of this column, Philip Bump, has a newsletter called "How To Read This Chart" billed as “a weekly dive into the data behind the news.” Mr. Bump knows data, and charts. He knows the hundreds-of-years figure for Trump’s sentence is nonsense but builds the entire column around an infographic pushing it, even though he also knows the likely impact on readers. He knows what an average reader will take away from this column that’s written around an eye-catching infographic, a column that offers only bland and uninformative generalities and no data whatsoever about how sentences actually work.

Resist Disinformation

Some people, like Trump and his supporters, will lie to you about what sentence he faces because that helps Trump fundraise and helps his narrative about being a victim of an abusive vendetta. But the Washington Post isn’t in Trump’s pocket. Why would it indulge in this hackery? The answer is simple, and lies at the heart of the media’s culpability for our broken, carceral criminal justice system: it’s attention-getting. This is the nerd-clickbait version of the ancient “if it bleeds, it leads.” This is part of the complex process of media malpractice that leads to widespread perceptions that crime is up even when it's down, making society worse for everyone. 

The Post does a lot of good, accurate work on criminal justice, including on the cases against Trump. This ain’t it. Editors, please gatekeep.