The issues
Forensic error
The lab coat isn't proof.
The data
52%
of Innocence Project DNA-exoneration client cases — Innocence Project
- Population
- People exonerated by DNA evidence and represented by the Innocence Project
- Time period
- Cumulative total, current as of April 2026
- Cases analyzed
- 257 cases
What the number means
In just over half of the Innocence Project's 257 DNA-exoneration cases, misapplied or misleading forensic science was a factor. Because these are, by definition, cases where DNA later proved innocence, they are unusually likely to have involved another, less reliable forensic method at trial — that's part of why this figure runs higher than the Registry's full-population figure below, not a sign that one source is wrong.
29%
of Full National Registry of Exonerations database — National Registry of Exonerations
- Population
- Every exoneration in the Registry, all crime types
- Time period
- Cumulative since 1989, current registry-wide figure (3,861 exonerations)
- Cases analyzed
- 3,861 exonerations
What the number means
Across the Registry's full population — including many cases with no biological evidence to retest at all — false or misleading forensic evidence is a documented factor in about 29% of exonerations. The gap from 52% largely reflects that the Innocence Project's cases are pre-selected for having DNA to test, which correlates with the kinds of violent, forensic-evidence-heavy cases where flawed pattern-matching testimony is most likely to appear in the first place.
These figures come from different populations, not a disagreement about the facts — see each statistic's own population and time period above for why they diverge.
Wrongful-conviction statistics vary depending on the population studied, the time period examined, and how researchers define and classify contributing factors. Percentages shown here are attributed to their original sources and should not be interpreted as representing all criminal cases, or all wrongful convictions.
What is it?
Forensic disciplines that sound scientific — bite-mark analysis, hair microscopy, arson pattern reading, and some fingerprint, ballistics, and tool-mark comparisons — have been used in court for decades without the rigorous, statistically validated testing the word "science" implies. Most of these are pattern-matching disciplines: an examiner looks at two things (a bite mark and a dental cast, a bullet and a test-fired casing) and reaches a subjective judgment about whether they're consistent. A 2009 National Academy of Sciences report and a 2016 White House PCAST report both found that, with the partial exception of DNA analysis, most of these disciplines lack the error-rate studies needed to support the near-certainty many examiners historically claimed in court. "Forensic error," as an exoneration category, covers both an examiner reaching a wrong conclusion and an examiner overstating how certain a correct-sounding conclusion actually is — and both have sent people to prison for crimes the underlying evidence never actually proved.
How does it happen?
- 01
Evidence is collected at the scene
Physical evidence — a bite mark, hair, a shell casing, a tool mark, burn patterns — is collected and sent for analysis, often to a lab affiliated with law enforcement rather than an independent institution.
- 02
A pattern-matching method is applied
Many forensic disciplines rely on an examiner's subjective visual judgment about whether two things are "consistent" — a judgment shaped by training, but without the large-scale, blind error-rate studies that back up disciplines like DNA analysis.
- 03
A match is declared
The examiner concludes a match, exclusion, or "consistent with" finding — a conclusion that can be affected, research shows, by knowing which result investigators are hoping for, even without any intent to mislead.
- 04
Certainty is overstated in testimony
Analysts have historically testified to a level of certainty — "to the exclusion of all other individuals in the world," for instance — that the underlying method has never been shown able to support.
- 05
The defense often can't afford a rebuttal expert
Retaining an independent forensic expert to challenge a state analyst's conclusion is expensive; many defendants, especially those with overworked public defenders, go to trial with the state's forensic conclusion effectively unchallenged.
- 06
The jury treats "forensic science" as inherently reliable
Jurors, reasonably, tend to give scientific-sounding testimony significant weight — without visibility into which disciplines have been independently validated and which haven't.
- 07
Later re-examination overturns the finding
Years or decades later, DNA testing, a field-wide methodological reassessment (as happened with bite-mark analysis and some fire-science indicators), or an independent re-review reveals the original conclusion could not actually be supported by the method used.
Why can it matter?
Forensic error typically enters a case at the investigation stage, when a lab result or expert opinion either confirms suspicion of a particular person or helps build probable cause for an arrest. From there, it shapes the entire prosecution: a confident-sounding forensic "match" is often the single piece of evidence a jury finds hardest to doubt, because it's presented as objective science rather than one person's subjective judgment. It can also affect the defense's own strategy — an attorney facing a state expert's confident testimony, without funding for an independent expert, may have little choice but to attack credibility rather than substance. And on appeal, overturning a forensic-error conviction typically requires either new DNA testing or a broader field-wide reassessment of the method itself, both of which can take years to materialize.
Important distinction
A forensic error doesn't always mean the analyst was being dishonest. Many now-discredited techniques — bite-mark analysis chief among them — were treated as legitimate, court-accepted science for decades before rigorous validation studies caught up with them.
Important distinction
A conviction being overturned on forensic grounds doesn't mean the analyst "lied" on the stand. Often the technique itself was never capable of the certainty that was claimed, regardless of the analyst's own good-faith belief in their conclusion.
Illustrative scenario
Fictional example for educational purposes
This is not a real Xonorate case. It illustrates how this issue can develop in a criminal case, generally.
The incident
The initial investigation
Where the problem begins
How the investigation develops
How it snowballs
A caliber match and a prior weapons charge make one person the focus
A tool-mark comparison — a subjective, unvalidated pattern-matching method — declares a "match"
The match is described in the report with a certainty the method cannot scientifically support
Investigators treat the case as solved and stop pursuing an unrelated tip
The defense, without funding for an independent tool-mark expert, cannot meaningfully rebut the state's finding
The jury hears the match described as conclusive, scientific proof
The examiner likely believed the conclusion. Tool-mark comparison training, for decades, taught exactly this kind of confident, absolute-sounding language as the field's professional standard — it wasn't a fabrication so much as an entire discipline overstating what its own method could actually prove. The deeper failure was structural: no independent proficiency testing required, no blind verification by a second examiner, and no funding on the defense side to challenge a conclusion dressed in scientific authority.
The courtroom
The state's firearms examiner testified as an expert witness, describing years of training and thousands of prior comparisons.
What the jury hears
“"Based on the unique striations, I can say to a reasonable degree of scientific certainty that these casings were fired from this weapon, to the exclusion of every other firearm."”
What the underlying record may show
“No national or state validation study has established a reliable, measured error rate for tool-mark comparison at the level of certainty claimed; a defense reanalysis years later, using a blind second-examiner protocol, found the original comparison "inconclusive" rather than a confirmed match.”
The conviction
Years later
The breakdown
- What happened
- A subjective tool-mark comparison was reported and testified to with a certainty the method has never been validated to support, and it became the case's central evidence.
- Why it mattered
- It ended pursuit of another viable lead and became the deciding evidence at trial for a discipline that cannot actually support the certainty claimed.
- What should have been examined
- Whether tool-mark comparison, as a discipline, could scientifically support an "exclusion of every other firearm" conclusion at all; and whether a second, blind examiner reviewed the finding before it was reported.
- What safeguard could have helped
- Mandatory blind verification by a second examiner; testimony limited to language the discipline's actual validation research can support; and defense funding for an independent forensic expert in any case where the state's forensic evidence is central.
The bigger lesson
Forensic testimony's power in a courtroom comes from sounding like objective science — but several widely used pattern-matching disciplines were accepted in court for decades before anyone rigorously tested whether they could support the certainty examiners claimed. The failure here isn't one dishonest analyst; it's an entire evidentiary category whose court-accepted confidence outran its actual scientific validation, for far longer than it should have.
See it in documented cases
These documented cases are presented as real-world examples of how this issue can appear in an actual case. They are not used to calculate the prevalence statistics above.
Kevin Baker
New JerseyExoneratedShortly before 6 a.m. on January 28, 1995, an unidentified caller reported finding two bodies in a pool of blood in Camden's Roosevelt Manor housing complex. Rodney Turner, 35, and Margaret Wilson, 40, had both been shot in the head. Five days later, based on street rumor, police arrested Kevin Baker and Sean Washington, then both 23, largely on the word of a single eyewitness, Denise Rand, who was high on crack cocaine at the time of the shooting and gave inconsistent, wavering testimony at trial. Baker and Washington were convicted after a two-day trial in August 1996 and sentenced to 60 years to life. Baker maintained an alibi — that he was with his girlfriend, Michelle Redden, at her mother's home and then her apartment that night — but the lawyer never called Redden as a witness. In 2011, the Last Resort Exoneration Project at Seton Hall Law School began reinvestigating, eventually uncovering a 911 recording (made by Washington, not identified for 18 years) and new ballistics and forensic analysis showing both victims were shot at close range while on the ground, directly contradicting the eyewitness's account of them being shot while running. In December 2019, the New Jersey Appellate Division unanimously vacated both convictions, finding the new evidence "powerfully undermines" that testimony. Camden County prosecutors declined to retry the case, and Baker and Washington were released on February 11, 2020, after nearly 25 years in prison.
Read the full case →Sean Washington
New JerseyExoneratedShortly before 6 a.m. on January 28, 1995, an unidentified caller reported finding two bodies in a pool of blood in Camden's Roosevelt Manor housing complex. Rodney Turner, 35, and Margaret Wilson, 40, had both been shot in the head. Five days later, based on street rumor, police arrested Sean Washington and Kevin Baker, then both 23, largely on the word of a single eyewitness, Denise Rand, who was high on crack cocaine at the time of the shooting and gave inconsistent, wavering testimony at trial. Washington and Baker were convicted after a two-day trial in August 1996 and sentenced to 60 years to life. Washington had in fact called 911 himself that morning after discovering the bodies while walking to a nearby pay phone, fearing one victim was his nephew — but that call went unidentified for 18 years. Beginning in 2011, the Last Resort Exoneration Project at Seton Hall Law School uncovered the 911 recording and new ballistics and forensic analysis showing both victims were shot at close range while on the ground, directly contradicting Rand's account of them being shot while running. In December 2019, the New Jersey Appellate Division unanimously vacated both convictions, finding the new evidence "powerfully undermines" Rand's testimony. Camden County prosecutors declined to retry the case, and Washington and Baker were released on February 11, 2020, after nearly 25 years in prison.
Read the full case →Anthony Ray Hinton
AlabamaExoneratedState forensic examiners testified that all six bullets from three related crimes were fired from one .38-caliber revolver found under a relative's mattress — ballistics testimony that was the only physical evidence tying Hinton to the murders. Independent experts later disputed that conclusion.
Read the full case →Ronnie Long
North CarolinaExoneratedJust after 9:30 p.m. on April 25, 1976, a man broke into a house on South Union Street in Concord, North Carolina, and raped Sarah Bost, a 54-year-old white widow who was home alone. The attacker covered her face and told her not to look at him. Bost did not identify anyone from a photo array at the hospital, and her first description was of a Black man in a dark leather jacket with no mention of facial hair. Later she described him as "light-skinned" or "yellow." Ronnie Long, a 20-year-old Black cement mason who lived with his parents, was arrested on a trespassing charge on April 29. Instead of showing Bost his photo, police brought her to his May 10 court date, where she sat disguised in a red wig and sunglasses and told officers he was her attacker before he said a word. Police then arrested him for rape and burglary. An all-white jury, seated from a 49-person pool that included only two Black residents after law enforcement screened the jury list, convicted him on October 1, 1976. He was sentenced to life. No physical evidence tied him to the crime, and several alibi witnesses had placed him elsewhere that night. Thirty years later, records finally released under a 2005 court order showed that the State Bureau of Investigation had tested 14 pieces of evidence in 1976 and that none of it implicated Long. The lead detective had told the jury the opposite, and a later judge wrote that he "lied. Repeatedly." Forty-three fingerprints lifted from the scene, never disclosed, excluded Long. In August 2020 the full U.S. Court of Appeals for the Fourth Circuit, by a 9-6 vote, reversed the dismissal of his habeas petition. He walked out on August 27, 2020, after 44 years, 3 months, and 17 days. Governor Roy Cooper pardoned him on grounds of innocence on December 17, 2020. In January 2024 the City of Concord agreed to pay $22 million and the State of North Carolina $3 million to settle his civil-rights lawsuit.
Read the full case →Byron Halsey
New JerseyExoneratedByron Halsey spent 19 years in New Jersey prisons for the 1985 sexual assault and murders of the two young children of his girlfriend, Margaret Urquhart, in the Plainfield rooming house where they all lived. Police interrogated Halsey — a man with a sixth-grade education and severe learning disabilities — for roughly 30 hours over a 40-hour span until he signed a statement that prosecutors used as a confession, and they stopped investigating neighbor Clifton ("Cliff") Hall. DNA testing obtained by the Innocence Project in 2006 excluded Halsey and identified Hall as the source of crime-scene evidence; Halsey's convictions were vacated on May 15, 2007, and all charges were dismissed on July 9, 2007.
Read the full case →Kennedy Brewer
MississippiExoneratedBrewer was convicted largely on bite-mark testimony from a forensic dentist, who testified that marks on the victim's body were human bite marks matching Brewer's teeth. The marks were later determined to be insect bites, and DNA testing excluded Brewer entirely.
Read the full case →Dayonte Resiles
Florida (Broward County)Active caseJill Halliburton Su was found dead in her Davie, Florida home on September 8, 2014, bound and stabbed in a bathtub. Nine days later, a DNA database search returned the name of Dayonte Resiles, then twenty years old, with no known connection to the victim or her family. He was arrested the following day. The State's case rested on DNA. Prosecutors told jurors that Mr. Resiles' DNA was recovered from items at the scene and that this established he was the killer. The defense argued throughout that a DNA association does not establish who committed a homicide, when the DNA was deposited, or how it got there. His first trial ended in a mistrial in December 2021 under extraordinary circumstances: the jury returned a manslaughter verdict, and when jurors were polled, one said she did not agree with it. Following the mistrial, a juror stated that some jurors had been unwilling to convict on the greater charge because of his race. A retrial three months later produced a first-degree murder conviction. The Fourth District Court of Appeal affirmed in a one-word per curiam opinion in January 2025. Mr. Resiles has maintained his innocence for twelve years, including in a letter to the court and in open court at his own sentencing. Xonorate is supporting his request for independent post-conviction review.
Read the full case →Xonorate Investigates
Read Xonorate's ongoing investigations and case developments.
Resources
Legal resources, innocence organizations, and support services.
The bigger picture
Both figures here — 52% of Innocence Project DNA cases, 29% registry-wide — point to the same underlying problem: pattern-matching forensic disciplines were used in American courtrooms for decades with far more confidence than their actual scientific validation could support. Xonorate's documented cases show two different versions of that failure: Anthony Ray Hinton's case rested on disputed ballistics conclusions, and Kennedy Brewer's rested on bite-mark testimony from a method later shown to be unreliable — in Brewer's case, marks that turned out to be insect bites, not a human bite at all.
Have questions about this issue?
Ask Xonorate can help you explore the research, legal framework, and related Xonorate material.
Ask Xonorate →Explore related issues
Sources / research
- DNA Exonerations in the United States ↗
Innocence Project — n=257 client cases; current as of April 2026
- Exonerations by Contributing Factor ↗
National Registry of Exonerations — Full registry, 3,861 exonerations since 1989