The Corner

Don’t Let FiveThirtyEight’s Misleading Statistics Fan Ferguson Outrage

Last night Ben Casselman, FiveThirtyEight’s chief economics writer, published a post showing how “incredibly rare” it is for grand juries to refuse to indict. The post was soon dominating the left side of my Twitter feed, presented as evidence of prosecutorial bias. And, yes, unquestionably it is rare for grand juries not to indict. 

So this is evidence the process was corrupt, right? Not so fast. FiveThirtyEight did not measure the number of cases where an individual is accused of a crime yet the prosecutor chooses not to take the case to a grand jury because the evidence is so weak, contradictory, or nonexistent that the prosecutor doesn’t believe there’s even probable cause for an arrest, much less evidence remotely sufficient to prove guilt beyond a reasonable doubt. This form of prosecutorial discretion is exercised all the time, and it’s entirely proper. As Andrew McCarthy noted yesterday, “In most cases where prosecutors decide not to charge a crime, they do not present the case to the grand jury in the first place.” Do we want zealous prosecution every time there’s a criminal accusation? Or do we want prosecutors to prosecute the crimes when there’s probable cause? 

Let’s break this down:

1. Yes, a prosecutor — if he chooses — can secure indictments through grand juries. 

2. If a prosecutor aggressively pursues an indictment when the evidence is ambiguous, weak, highly contradictory, or otherwise raises serious questions about even the existence of probable cause, we call that action “prosecutorial abuse.”

3. When evidence is weak — as noted in point 2 — good, ethical prosecutors will tend to drop cases rather than take them to grand juries or preliminary hearings (thus not showing up on FiveThirtyEight’s statistics).

4. The exception to point 3 often lies in publicly controversial cases, such as police shootings, where the prosecutor doesn’t want to be seen as the sole decider of the case. In those circumstances, prosecutors will often send a case to the grand jury, but not as an advocate for the indictment but rather as a conduit for all known evidence in the case — essentially punting their own prosecutorial discretion to the grand jurors. FiveThirtyEight acknowledges this happens:

Ordinarily, prosecutors only bring a case if they think they can get an indictment. But in high-profile cases such as police shootings, they may feel public pressure to bring charges even if they think they have a weak case.

“The prosecutor in this case didn’t really have a choice about whether he would bring this to a grand jury,” Ben Trachtenberg, a University of Missouri law professor, said of the Brown case. “It’s almost impossible to imagine a prosecutor saying the evidence is so scanty that I’m not even going to bring this before a grand jury.”

5. Consequently, it’s rare for grand juries to refuse to indict, except in the kinds of cases outlined in point 4.

I’m a big fan of FiveThirtyEight and enjoy reading it daily, but — and they should and do know this — statistical analysis has its profound limits, especially when the statistics are not even measuring the precise question at issue. Here, a more valid question would be: “How often are defendants indicted when the prosecutor does not believe there is evidence sufficient for a conviction on any potential count?” In a functioning criminal-justice system, the answer should be as close to “never” as possible.

I’ve been reviewing the evidence in the case (over at The Atlantic, Conor Friedersdorf has an interesting summary of conflicting witness testimony). I’ve also been reviewing the outraged commentary — including from one Salon writer who declares, “But white folks only really pay attention if they fear they have something to lose. Smoke flares in their nostrils, because then they are confronted with the possibility of charred, burning, white flesh. No more water. The fire next time” — and I’ve yet to see a detailed analysis arguing convincingly that the grand jury made the wrong decision based on the evidence before it.

When buildings are burning, and pundits are calling on protesters to confront Americans with the “possibility of charred, burning, white flesh,” now is not the time to further inflame passions with quick, misleading statistical posts, especially from otherwise-respected outlets. Sometimes, the clicks just aren’t worth it.


The Latest