Case Study 16.1 — Ray Krone: How a Bite-Mark "Match" Sent an Innocent Man to Death Row

A note on sourcing and tone. The facts below are drawn from the public record of a widely documented U.S. case (Arizona; conviction 1992; exoneration 2002). It is used here to teach how an unvalidated comparison method — bite-mark identification — can become confident, jury-persuading courtroom testimony and contribute to a wrongful conviction, and how a validated method (DNA) later exposed the error. We treat a real murder soberly and confine ourselves to documented, public facts; the human cost on both sides — a guilty person who remained free and an innocent man imprisoned under a death sentence — is the point, not the spectacle.

Background

In December 1991, a woman named Kim Ancona was found murdered in a bar in Phoenix, Arizona, where she worked. The victim had been stabbed, and the body bore bite-mark injuries. Ray Krone, a former U.S. Postal Service worker and an acquaintance who was known to frequent the bar, became the focus of the investigation. He had no prior record of violence and maintained his innocence from the outset.

The case against Krone was substantially built on bite-mark evidence. Investigators obtained an impression of Krone's teeth — by one account using a Styrofoam impression — and a forensic odontologist compared his dentition to the bite injuries on the victim. Krone's teeth were notably crooked, and that distinctiveness became, in the prosecution's hands, the engine of the identification. The press fastened on it: Krone was dubbed "the Snaggletooth Killer."

The forensic evidence

The bite-mark testimony was the pillar of the prosecution's forensic case, and its features map almost exactly onto the warnings of §16.5.

• The comparison itself. A forensic odontologist testified that the bite marks on the victim matched Krone's distinctive dentition. The comparison was the classic bite-mark procedure: the examiner was effectively shown the suspect's dental models and asked whether they accounted for a patterned injury on skin — the very setup §16.5's Cognitive-Bias Watch identifies as an invitation to confirmation. The injury was on skin, the worst imaginable substrate (§16.5): elastic, mobile, and changing over time, so the mark photographed after the fact was not the mark at the moment it was made.

• The certainty projected to the jury. As bite-mark experts of the era routinely did, the testimony conveyed a strong, identifying association — these teeth made these marks — to a jury with no way to know that the method beneath that confidence had never measured its own error rate. The exhibit was exactly the kind §16.5 calls the most dangerous thing a juror may see: a credentialed expert, with images, asserting a match.

• What else the case did and did not have. This was, in forensic terms, a thin case dressed up by a confident comparison. The strongest "scientific" link to Krone was the bite-mark opinion; there was no DNA identification of him as the source of the biological evidence at the scene. That absence would prove decisive a decade later.

A methodological note, threading back to §16.1. Bite-mark identification claimed the top rung of the hierarchy of conclusions — individual identification, "these teeth and no others" — from a foundation that, as the chapter argues, barely supports the second rung (class-level "consistent with"). The Krone trial is what it looks like when a discipline reaches for a rung its science cannot hold.

The two trials

Krone was convicted of murder in 1992 and sentenced to death. On appeal, his conviction was overturned on a procedural ground unrelated to the bite-mark science. He was retried in 1996 — and bite-mark testimony again featured prominently — and again convicted, this time receiving a life sentence. Through both trials and the years between, he maintained his innocence. The repetition matters: the same unvalidated method, presented with the same confidence, persuaded a second jury. The problem was never one stray expert; it was a method that produced conviction-grade certainty wherever it was admitted.

What the evidence did — and didn't — establish

This is the heart of the lesson. The bite-mark testimony established nothing reliable about who killed Kim Ancona. It associated a patterned injury on a difficult substrate with a man whose teeth happened to be distinctive — and it did so through a procedure with no measured error rate, in a discipline that, when tested, cannot reliably do what it claimed (§16.5). What it did establish, tragically, was a conviction: it was persuasive precisely because it was confident, and confident precisely where it had no right to be.

In 2002, DNA testing of biological evidence from the crime told a different story. The DNA excluded Krone and matched another man — someone already in prison for an unrelated offense, with no connection to Krone. After more than a decade of imprisonment, much of it under a sentence of death, Ray Krone was exonerated and released. He is frequently cited among the people freed from death row in the United States by post-conviction DNA testing.

Connect this to Figure 16.2 in the chapter ("Two methods, one defendant, opposite answers"). The DNA did not make the bite-mark method "almost right." It exposed that the method's confident identification had carried no real reliability at all. A discipline at the discredited end of the validity spectrum was overruled by one at the strong end — and the bite "match" turned out not to be weak evidence pointing the right way, but an unvalidated method pointing the wrong way (§16.6).

Outcome

Krone's exoneration became one of the most cited cases in the campaign against bite-mark evidence. It put a name, a face, and a death sentence to the abstract proposition that the method had never been validated — and it arrived as part of a broader wave of DNA exonerations (Chapter 6) that, together with the formal reviews to come, drove bite-mark analysis into the retreat it remains in today (§16.5). The man whose DNA matched the crime, not Krone's teeth, was the actual source of the biological evidence; the real cost of the bad science was therefore double, in the book's recurring phrase — an innocent man imprisoned and the true perpetrator left unaccountable for the years the wrong man served.

The lesson

The Krone case teaches the chapter's argument in its sharpest form: confidence is not validity. The most persuasive exhibit in a courtroom — an expert holding two images and a pointer, asserting a match — can be the most wrong, and it is most dangerous exactly when the underlying method has never measured how often it errs. The right question is not "how sure is the expert?" but "what is the measured error rate of the method the expert used?" For bite-mark identification, as Krone learned across a decade and two trials, the honest answer to that question guts the testimony.

The case also carries the book's four themes at once (developed fully in §16.6): exclusion over proof (the truest forensic act came late, when DNA excluded the convicted man); the validity spectrum (bite marks at the bottom, DNA at the top, applied to the same defendant); cognitive bias (an examiner shown a suspect and asked to confirm a match, finding one in a distorted mark); and the CSI effect cutting both ways (a jury over-trusting a confident expert — the less-discussed direction of the effect, §1.2).

Discussion questions

1. The bite injuries were on skin. Using §16.5, list the specific properties of skin as a recording medium that make it "about the worst imaginable substrate," and explain how each undermines the claim that the mark faithfully recorded Krone's dentition.

2. Krone was convicted, had the conviction overturned, was retried, and was convicted again on bite-mark testimony. What does the second conviction tell you about whether the problem was a single bad expert or the method itself?

3. Distinguish the hierarchy of conclusions (§16.1) the bite-mark testimony claimed (individual identification) from the rung the method's science can actually support. Why is reaching for the top rung the recurring failure mode this chapter warns about?

4. The DNA "did not retroactively make the bite-mark method 'almost right.'" Explain this carefully: why is an unvalidated method that happens to point the wrong way worse than no evidence at all, rather than merely "weak evidence"?

5. Apply the Cognitive-Bias Watch from §16.5. How did the structure of the bite-mark comparison — being shown the suspect's models and asked to confirm a match — invite confirmation bias, and what (if any) safeguard could have reduced it?

6. On cross-examination of the original bite-mark testimony, write the three questions §16.5 says "write themselves." For each, state what the honest answer would have been in 1992 — and how that answer should have affected the weight a jury gave the evidence.