17 Million To One, Then "Absolutely Not"
Day 4 of the Frostproof Massacre trial built the State's DNA case against Tony Wiggins on a single number. Three straight cross-examinations spent the rest of the day trying to take it apart.
Three friends went out on a night-fishing trip outside Frostproof, Florida, in July 2020, and only one of them came home. Tony Wiggins is on trial for his life in Bartow, accused of murdering Brandon Rollins, Keven Springfield, and Damion Tillman, and the State of Florida is asking a jury to send him to death. He is presumed innocent, and he stays presumed innocent through every sentence below. His brother, William "Robert" Wiggins, and his girlfriend, Mary Whittemore, were each charged only as accessories after the fact to a capital felony, and both have already resolved those charges by plea. Neither of them is on trial for murder, and neither faces the death penalty. Only Tony Wiggins does.
Day 3 closed on Michelle Miller, a Polk County Sheriff's Office forensic investigation supervisor, mid-testimony, with a second unexplained bag of evidence hanging over the courtroom and the jury sent home before she could answer for it. Day 4 opens by finishing that thread, then spends the rest of a long day on a single number: 17 million to one. That is the odds a DNA analyst gave the jury against anyone but Tony Wiggins matching a critical piece of physical evidence. Everything else in Day 4, three separate cross-examinations across three different videos, is the defense trying to take that number apart piece by piece before the jury ever gets to weigh it.
Finishing Miller: A Line Between "Observed" And "Collected"
Michelle Miller returns to the stand to close out the marathon cross-examination that carried over from Day 3. The defense keeps pressing the same seam it opened the day before: what exactly did Miller do herself, and what did she only watch happen. Miller draws a hard line for the jury between what she merely observed inside the white truck at the center of this case and what she actually collected into evidence herself, and she confirms something that sounds small until you sit with it, that items investigators photographed but chose not to collect were put back inside a vehicle that was processed again later. The defense also gets her through presumptive blood testing, a seat belt strap she cut out of the truck herself, and a gun safe recovered from the property that nobody ever swabbed for DNA at all. The court excuses Miller subject to recall, meaning her testimony is not finished as a legal matter even though the jury has now heard everything she has to say about this evidence.
It's the same pattern Day 3 established across four straight witnesses: real physical evidence, sitting next to an admitted gap in exactly how it was handled. A gun safe that never got swabbed is not proof of anything by itself. It is one more place the defense can point and ask the jury what else might be missing from a chain of custody this case has already shown plenty of cracks in. And it sets the tone for everything that follows: Day 4 keeps testing the same question Day 3 asked all day, whether the paperwork behind a piece of evidence can actually carry the weight the State wants to put on it, except now the evidence in question is not a truck or a chain, it is a laboratory result.
The Analyst, The Casings, And The Number
Once Miller steps down, DNA Labs International analyst Rachel Oefelein takes the stand, and the case shifts from trucks and chains to laboratories and probability. Oefelein walks the jury through how a DNA profile actually gets built, then through whose reference standards her lab actually received to compare against. Investigators recovered seven spent shell casings connected to this case, and Oefelein testifies her lab DNA-tested only two of them. The other five, by her own account, had visible blood on them, and blood on a casing makes further DNA testing pointless, because it swamps out any usable genetic profile underneath it.
The two casings that could be tested are where the number comes from. Oefelein tells the jury the odds against anyone else in the general population matching that DNA profile are 17 million to one, and that number points at Tony Wiggins. She does not stop there. She also walks the jury through where the case's other named men land against the same evidence: Mary Whittemore is excluded outright as a contributor, William "Robert" Wiggins comes back inconclusive, meaning the test could neither include nor rule him out, and Tony Wiggins Sr. is ruled out entirely. Before the day moves on, Oefelein explains one more detail that will matter later: every casing gets an ethanol bath before it ever reaches a ballistics examiner, a routine step in her lab's own process, not anything unusual to this case.
What Seventeen Million To One Actually Means
It is worth pausing on the number itself before the cross-examination starts working it over, because "17 million to one" is doing a specific, narrow job, and it is easy to hear it as something bigger than it is. It is not a statement that Tony Wiggins is 17 million times more likely to be guilty than innocent. It is a statement about a population: the odds that a randomly selected, unrelated person from the general population would happen to share this specific DNA profile are roughly one in 17 million. That distinction is exactly why the defense spends the rest of the day asking who was actually compared against that population, and who was not. Mary Whittemore, tested and excluded. William Wiggins, tested and inconclusive. Tony Wiggins Sr., tested and ruled out. Tony Wiggins himself, the match. Nobody else who shares his blood, as later testimony makes clear, was ever tested at all.
"A Lot Of Silk": Shrinking The Science Down To Scale
Defense counsel Debra B. Tuomey takes over cross-examination and spends the next block methodically walking the reliability chain backward, from collection through packaging through transport through the plain risk of contamination, before she ever attacks a single result. She gets Oefelein to agree that as few as five skin cells left behind can be enough to build a full DNA profile, and that a box of ammunition passes through more than one set of hands before a round from it is ever fired. Then she shrinks the science itself down to a scale a jury can actually picture: roughly 100,000 miles of DNA exist inside a single human cell, and the stretch of it forensic testing actually reads is a sliver of that whole strand, roughly 300 feet.
Tuomey then walks Defense Exhibit 58A's own worksheet line by line, comparing the trace amount of DNA recovered from the casings against the DNA recovered from reference swabs taken directly from a person's cheek. The contrast is not subtle, and Oefelein herself supplies the image that sticks: DNA so abundant on the reference swabs, she calls it "a lot of silk," standing next to DNA on the casings that barely registers at all. Along the way, one exchange goes the other direction. Tuomey makes a claim of her own about how the underlying science works, and Oefelein, still under cross, corrects it live, on the record, before the questioning moves on. It is a small moment, but it is the kind of moment that matters in front of a jury: the witness the defense is trying to undercut also shows she will push back when the premise of a question is simply wrong.
It builds toward a moment neither side gets to walk back from. Pressed one more time on that specific casing result, Oefelein stops hedging and says it in plain words: the sample is degraded. Not possibly. Definitely.
A Vulnerability In The Machine That Read His DNA
Tuomey does not stop at the biology. She pivots the cross toward the software running inside the machine that actually reads a DNA profile once the lab has one, and gets Oefelein to confirm something the jury may not have expected to hear: Thermo Fisher, the company behind the analysis software, has disclosed a known security vulnerability in it. Watch how the lab answers for that admission rather than deflect it. Oefelein describes an on-site server room, three separate key-holders required to access it, and a storage system built to flag the moment any file inside it is altered. It is a defense built on process and access control, not a denial that the vulnerability exists at all.
The Software Itself Becomes The Target
Oefelein stays on the stand for a third straight session of cross-examination, and this block is where the defense stops attacking the DNA sample and starts attacking the machine that interpreted it. STRmix is the specific software the State leaned on to turn a mixed, degraded casing sample into the number the jury already heard. Tuomey walks Oefelein through dye channels and the raw mechanics of how STRmix converts a mixture of genetic material into a probability at all, and in the middle of that walkthrough Oefelein concedes a specific gap in the software's own math, a moment significant enough that it sends both sides to the bench before the cross can continue.
Once testimony resumes, the defense pivots into the federal government's own 2016 report on the limits of DNA mixture science, walking the jury through PCAST, the President's Council of Advisors on Science and Technology, and which federal agencies actually responded to its warnings about exactly this kind of DNA evidence. From there, Tuomey gets Oefelein to admit that STRmix's own software carries a documented history of miscoding, fifteen separate instances, before Oefelein can even finish qualifying her own answer. The defense then gets her to call one of STRmix's own co-developers, Dr. Buckleton, an authoritative voice, and reads his own published words back to her: that it is often impossible to say with certainty which specific genetic markers, or how many different people, are actually present inside a mixed sample.
That sets up the moment the day is now built around. Asked flatly whether the sample tested in this case is robust, Oefelein does not qualify or soften it. "Absolutely not." A sample the State's own case treats as conclusive becomes, in the analyst's own words on cross, low-level. The defense closes this block with what amounts to a sweep of the whole Wiggins family: one by one, they ask whether any of Tony Wiggins' cousins, his own mother, any responding officer, or any bystander at the scene was ever swabbed to rule out a family match. The answer, every single time, is no.
The Exhibits The Jury Will Never See
The day's final cross-examination opens with a ruling on the record before testimony even resumes: two of the defense's own exhibits, the raw data pages they used all through cross to attack the software behind the State's DNA match, will never go to the jury. From there, the defense presses further into STRmix's own validation history, and Oefelein has to place her lab's original 2019 sign-off directly against the exact version of the software actually used to analyze evidence in this case. Tuomey pushes her to admit that a 2024 fix to the software was never accompanied by a full repeated validation, then makes her defend having described it as one anyway.
The cross gets granular from there. Oefelein is walked through the software's own probabilistic modeling, essentially its guesswork engine for resolving a mixed sample, and pressed on validation errors she cannot fully recall on the stand. At one point Oefelein pushes back directly, telling the defense "you can't treat it like a black box," even as she is walked line by line through a 119-page raw-data exhibit built to do exactly that. The defense also gets her to confirm that the scale her own lab uses to describe the strength of a DNA match to a jury changed after 2016, and makes her explain, in her own words, why.
The State briefly brings Oefelein back one more time to revisit STRmix's own history before both sides finally excuse her. The judge closes the day by returning to Defense Exhibits 60A and 60B one final time, the ruling that will keep them out of the jury's hands standing as the day's last word. It is worth being direct about what that means and does not mean. The DNA match to Tony Wiggins remains one of the State's central pieces of evidence in a case where a jury could send him to death row. Every crack the defense found in the software behind that match on Day 4, and every one of its own exhibits the jury never gets to weigh directly, becomes part of the record an appeal could stand on someday, whatever this jury eventually decides.
Two Full Testimonies Also Landed Today
Alongside the day's four Case Story installments, two complete witness testimonies from earlier in the case's own record were released today, each one packaged start to finish, sworn in to excused. Lt. Sam Bunch, the Polk County Sheriff's Office case supervisor with more than 23 years on the job, walks the jury through the earliest hours of the investigation in his own full testimony: the call that first brought his team to the scene, assigning a crime scene detective, and developing the earliest leads. A fresh Dollar General bag found inside a truck at the scene sent detectives straight to the store and its surveillance footage, a trail that led investigators to a heavily wooded family property where Tony Wiggins Sr.'s family lived. On cross, the defense presses Bunch on how that property was actually approached. He confirms he and several colleagues went out in the middle of the night, roughly 1:30 in the morning, and that Tony Wiggins came outside to meet them with no socks and no shirt on. Bunch confirms Wiggins was detained for hours, was read his Miranda rights by another detective, never tried to run or leave, and answered every question he was asked. He is excused under subpoena, told to remain available if either side needs him again.
Michelle Miller's own complete testimony, spanning both Day 3 and Day 4, is also up in full, stitched into one continuous release. It walks both trucks recovered in this case, then the search of the family property across two travel trailers. Miller applies Blue Star to a seatbelt in the white F-150 and testifies it revealed blood invisible to the eye. She photographs a possible fingerprint recovered from a gun found in a bedroom safe, testifying it was etched into the surface and could not be lifted with tape, the same detail Day 3's own coverage already flagged as evidence that could only ever be photographed, never physically recovered. On cross, the defense presses her over the red Chevy S10's chain of custody: who towed it, and who signed for it. Her testimony carries into Day 4, where cross-examination resumes on those same receipt questions before the defense raises the second, unexplained Dollar General bag Miller says she has never heard of, found somewhere else entirely. The court cuts the exchange off as the jury is released for the evening, and Miller is excused, told to remain on subpoena.
What A DNA Match Is Actually Doing In A Case Like This
It's worth stopping here and naming plainly why a day built almost entirely around dye channels, validation reports, and a software company's own disclosed vulnerability matters as much as it does in a capital case. Day 3 was about whether the evidence itself could be trusted, whether a truck sat sealed, whether a chain got mixed up with another chain. Day 4 assumes the evidence arrived intact and asks a different question: can the machine, and the science, that turned that evidence into a number be trusted the same way. A jury that has never heard of STRmix, PCAST, or a dye channel is now being asked to decide how much weight a single probability deserves, a probability produced by software the State's own witness had to admit carries a real vulnerability and a documented history of coding errors.
None of that is proof the number is wrong. Fifteen known miscodes in a piece of software used across an entire industry is not the same thing as a miscode in this specific case's result, and Oefelein's own testimony never concedes that it is. But a capital jury is entitled to hear the full distance between "this software has flaws" and "this specific result is unreliable," and Day 4 is the day the defense spent building every rung of that ladder in public, one cross-examination at a time, rather than saving it for a closing argument the jury would have to take on faith.
Where Day 4 Actually Leaves This
Lay the day out in order and the shape of it is different from Day 3, but the underlying pattern holds. Day 3 was chain-of-custody: an unsealed truck, a wrong-dated diagram, two chains that got confused with each other. Day 4 is the same pressure applied one level up, not to what was collected but to how it was tested and read. Two of seven casings could even be tested at all. The result that came back is one the State's own analyst now calls, in her own words, degraded and low-level, even as she stands behind the 17-million-to-one number it produced. The software that turned that sample into a probability carries a known security flaw and a documented history of miscoding. No one who shares Tony Wiggins' own bloodline was ever swabbed to check whether the match could point somewhere else inside his own family. And the two exhibits the defense used to build that entire case against the software will never reach the jury at all.
None of that is proof of anything by itself, in either direction. It is the second half of the State's physical case, laid out one cross-examination at a time, on a day where the analyst herself supplied the State's strongest number and the defense's strongest doubt in the same breath. Tony Wiggins has not been convicted of anything, and the presumption of innocence he is entitled to does not shrink as the evidence against him grows more technical. He remains the only person on trial for these three murders and the only person facing a death sentence in this courtroom. His brother and his girlfriend already resolved their own, lesser charges by plea and are not on trial here, whatever role the evidence discussed above plays in the case against him alone.
For the whole day, gavel to gavel, with every recess cut out and every witness in order, watch the full edition below.
Watch the system. Question everything.
— Justice
Join the Discussion