Yes, Karen Read Did It

A comprehensive explanation of how the evidence in this case, when properly understood and contextualized, leads to only one reasonable conclusion

The trial’s over. Karen Read was acquitted, and double jeopardy has attached. She will never again face any criminal charges related to the death of her boyfriend, John O’Keefe. In light of that reality, this will likely be my last post on this subject. And because this is no longer a criminal matter, I’m not going to talk about reasonable doubt or burdens of proof or rules of evidence in this post. Those are concepts that apply within a criminal courtroom, concepts intended to guide and constrain how judges and jurors evaluate evidence. They are designed (for very good reason) to make the prosecution’s job more difficult. As the saying goes, our criminal justice system is designed around the idea that it’s better that 10 guilty people go free than one innocent person be convicted.

What I want to do in this last post is explain, comprehensively and for posterity, why Karen Read is one of those lucky 10 guilty people who gets to go free.

For those of you who follow this blog and are already very familiar with the evidence and arguments in the case, I apologize ahead of time for belaboring some of these points, but I’m really writing this for a different audience. My hope is that people who are less familiar with the case – but who want to know the truth – will stumble across this post and come away with a better understanding of what actually happened to John O’Keefe.

The Timeline of Events is Undisputed and Conclusive

To truly understand this case, you have to start with the timeline of events established by the undisputed digital evidence. This evidence is clear, undisputed, incapable of fabrication or tampering, and once you understand it, it becomes very clear that just about everything Karen’s defenders focus on is just noise.

There are three primary sources of digital evidence in this case: John’s iPhone, Karen’s iPhone, and the computer system within Karen’s Lexus SUV. From these three sources, none of which were challenged in any meaningful way by the defense at trial, we can put together an incredibly clear and precise timeline of events from the night of January 29, 2022.

We know from the testimony of numerous witnesses and from surveillance footage that Karen met up with John at C.F. McCarthy’s, a pub in Canton center, at around 9:00 p.m. that night. Based on the surveillance footage and receipts from the bar, it’s clear that Karen had the equivalent of at least 6-7 vodka drinks at McCarthy’s. At around 11:00 p.m., the two left that bar and went across the street to the Waterfall Bar & Grille, where they met up with the people who would eventually go to 34 Fairview Road (Brian and Nicole Albert, Jen and Matt McCabe, and Brian Higgins). At the Waterfall, Karen appears (based on surveillance footage and receipts) to have had at least another 1-2 drinks.

Just after midnight, the Waterfall closed and everyone headed out. It had just started to snow, and John and Karen had been invited to Brian and Nicole Albert’s home (34 Fairview Road) to continue drinking. As they left the bar, John can be seen on surveillance footage holding a cocktail class containing Karen’s vodka soda. They got into Karen’s Lexus SUV, with Karen behind the wheel. We know from the Lexus computer data that the vehicle was powered on at 12:12 a.m.

John and Karen had never been to the Albert home before, so as they drove away from Canton center, they realized they needed directions. So John, riding in the passenger seat, called his friend Jen McCabe. We know this call occurred from the data from both John and Jen’s iPhones. Based on the testimony of Jen and Matt McCabe – corroborated by statements Karen herself made in various interviews – we know that Jen informed John that the Albert house was just up the street from “Bella’s mom’s house,” a reference to a woman John had dated just prior to Karen.

While speculative, there is very good reason to believe that Jen’s reference to “Bella’s mom” triggered a fight between John and Karen. While John and Karen were, by all accounts, getting along well during their stint at C.F. McCarthy’s and the Waterfall, their phone records indicate that they had been fighting much of that day by text message and that their relationship was in a rocky state. Just a month prior, they had gotten into a huge blowout fight during a trip to Aruba, triggered by a (very likely mistaken) belief by Karen that John had been kissing a family friend. The fallout of that incident was still evident in their heated text message exchanges on January 28. We also know — via an extensive text message record — that Karen had come very close to initiating an affair with Brian Higgins (a friend of John’s) in the weeks prior and had vented to him about the rocky state of her relationship with John.

Back to the timeline. After getting directions from Jen, John entered the address (34 Fairview) into the Waze app on his iPhone, which triggered the phone to start recording very precise, high-frequency GPS location data. From this point onward, we have very precise time and location data for the SUV. We know that Karen drove down Cedarcrest Road and, at first, missed the left hand turn onto Fairview Road. Realizing she had missed the turn, Karen turned left into a driveway and initiated a three-point turn. While executing this three-point turn, Karen put the SUV in reverse and depressed the gas pedal to greater than 30% of its maximum, which caused the first of two “trigger events” that were recorded that night in the Lexus’ “techstream” data system. That system is designed to detect potentially unusual acceleration events (particularly in reverse) and to record relevant vehicle data for diagnostic purposes. For timeline purposes, this first trigger event is really important because it provides a common time point we can use to align the clock in John’s iPhone with the clock in the Lexus’ infotainment system.

Based on the Lexus’ clock, this trigger event (reversing out of the driveway) began at 12:23:33. Based on the GPS data in John’s iPhone, this event likely occurred at 12:24:02, plus or minus a second or two. That means that, at this point, the two clocks were off by about 29 seconds (or, put another way, the Lexus’ clock was running roughly 29 seconds behind the actual time).

Karen’s SUV arrived back at the intersection with Fairview Road at 12:24:18, at which point a pickup truck was coming down Cedarcrest in the opposite direction and also preparing to turn onto Fairview. That truck was driven by Ricky D’Antuono and it had two passengers, Ryan Nagel and his girlfriend Heather Maxon. All three of these witnesses have testified under oath that they saw Karen and John in the SUV and watched as they turned onto Fairview just ahead of them.

We know from the GPS data in John’s phone that Karen’s SUV pulled past the driveway at 34 Fairview, without stopping, and eventually came to a stop by the flagpole on the far left side of the front yard at 12:24:38. This is consistent with the testimony of all three people in the pickup truck, who said that they pulled up right behind Karen’s SUV as they waited to pick up Julie Nagel (Ryan’s sister).

It’s worth noting at this point that the Apple Health data in John’s iPhone recorded him taking some steps and climbing three flights of stairs during the time period immediately preceding their arrival at 34 Fairview. When the conspiracy theories about this case first started proliferating (fueled by selective defense team leaks), this Apple Health data was cited as evidence that John got out of the SUV and entered the house. But we now know, without question, that was not the case. The Apple Health data comes from the same source (John’s iPhone) as the GPS data, which indisputably shows that John was inside the moving vehicle throughout this period. Moreover, the prosecution’s expert, Ian Whiffin, established through testing that an iPhone that is held in the hand of a passenger in a moving vehicle can record steps as it moves and flights climbed as it ascends elevation. Whiffin even showed where specifically on the route to 34 Fairview the vehicle would have ascended elevation and how those locations and times correspond with the flights recorded in John’s Apple Health data. The defense offered no contrary evidence or testimony.

So we’ve established that John and Karen arrived by the flagpole at 12:24:38. From that point, John’s phone shows no movement at all until 12:31:56, over seven minutes later. This is consistent with the eyewitness testimony of both the people in the truck and the witnesses from the house. The three people in the truck have all testified that they never saw anyone get out of Karen’s SUV or walk towards the house. During this period, Julie Nagel (Ryan’s sister) walked out of the house and spoke with her brother in the truck, telling him she was going to stay longer and no longer needed a ride home. Julie also testified (in the first trial) that she never saw anyone leave the SUV or walk toward or inside the house.

Both Jen and Matt McCabe have also testified that they looked outside and saw Karen's SUV idling by the flagpole during this period but never saw anyone get out of the vehicle or come inside. Further corroborating this testimony are Jen’s text messages to John. At 12:27, she texted John “here?”, and at 12:31 she texted “pull behind me” – both of which are consistent with seeing Karen’s SUV outside but not with John having already left the vehicle or having already come inside.

As noted earlier, while speculative, there is a strong basis for believing that John and Karen were fighting with each other as they sat together in the vehicle during those seven minutes. During her ride to Good Samaritan Hospital the next morning, Karen told the paramedic that her last conversation with John had been an argument.

After what they estimated to be roughly five minutes, the three people in the D’Antuono truck pulled away, passing the driver’s side of Karen’s SUV as they did so. Notably, these witnesses remember seeing Karen behind the wheel, but do not remember seeing John. Karen’s defenders have long tried to argue that this means John was already inside the house at this point. But this is almost surely not true. First, as noted above, the GPS and Apple Health data from John’s phone give no indication that he moved away from the flagpole area. Second, the witnesses in the truck specifically remember seeing the SUV’s dome light on as they passed. That fact strongly suggests that one of the vehicle’s doors had just recently been opened. Given the height of the SUV and the darkness of that corner of the yard, the most likely inference is that John had, just moments before, stepped outside of the vehicle and was standing on the passenger side of the SUV as these witnesses passed, blocked from their view.

We’ve now reached the most important part of the story. John’s Apple Health data indicates that he began to move at 12:31:56 and that he ceased moving, forever, 20 seconds later, at 12:32:16. During those 20 fateful seconds, John recorded 36 steps (equivalent to roughly 82 feet), though we can’t know whether they were all in one direction or (more likely) in one direction and then another. We do know, from the defense team’s own measurements, that 83 feet is not enough to get from the flagpole area to the house or garage (much less back again).

We also know that during those 20 seconds, John opened his phone with FaceID (at 12:32:04), saw Jen McCabe’s 12:31 text message, and then closed his phone by pressing the side button (at 12:32:09).

But here’s the most important fact. According to the Lexus’ techsteam system, a second trigger event occurred at 12:31:38, per the Lexus’ internal clock. When a trigger occurs, the system records 10 seconds worth of data: five seconds before the event and five seconds after. Per the Lexus’ clock, this means that data was recorded from 12:31:33 to 12:31:43. But as established by the prior trigger event, we know that the Lexus’ clock was running roughly 29 seconds behind the more accurate clock in John’s iPhone. This means that the trigger event data recorded from the Lexus actually began at around 12:32:02 (six seconds after John began moving) and ended at around 12:32:12 (four seconds before John recorded his last movement).

And what does the trigger event data tell us about the movement of Karen’s SUV? Well, at the beginning of the period (approximately 12:32:02), the SUV was moving forward at 13 mph. Then, at approximately 12:32:07 (as John was looking at Jen’s text message), the SUV was put in reverse and the accelerator was depressed to 70% of maximum. The SUV rapidly accelerated and, by the end of the data period (approximately 12:32:12), it was moving at the astounding speed of 24 mph, in reverse, and still accelerating.

Notably, the techstream data does not (and cannot) provide any evidence of the timing of a collision. That’s not what it’s designed to do. All it can tell us is that Karen put the vehicle in reverse, gunned the accelerator, and continued to accelerate for at least five seconds. And when combined with the data from John’s iPhone, we know that this reckless reversal happened just outside of 34 Fairview Road and began approximately two seconds before John’s last interaction with his iPhone and nine seconds before John’s last recorded movement.

That sequence of events is incredibly damning, and that’s before we get to the rest of the timeline or any of the physical evidence. And most importantly, the defense did not dispute any of this data. The only argument offered by the defense in response to any of this digital data was a half-assed attempt to argue that the Lexus’ internal clock may have been correct, i.e., that Karen reversed the vehicle at an insane speed for over five seconds, but that this occurred 20+ seconds before John last interacted with his iPhone.

Before explaining why this argument is incorrect as a factual matter, it’s worth pausing for a moment to bask in its absurdity. The defense did not argue that there was anything incorrect or fabricated about the techstream data. They didn’t dispute that Karen gunned her vehicle recklessly in reverse for over five seconds just outside 34 Fairview. They instead contended that John managed to survive this close encounter with Karen’s SUV, decided to stop and check his text messages moments later, and was then fatally injured in some totally unrelated event just seven seconds after that. That was their argument. Seriously.

But absurdity aside, this argument is also clearly factually incorrect. The logic the prosecution’s expert applied in determining the time variance between the iPhone and the Lexus – by comparing the times associated with the three-point turn on Cedarcrest – is very straightforward and unassailable. In an attempt to muddy the waters, the defense’s expert noted that some of the timestamps recorded in the Lexus’ infotainment system relating to phone calls Karen made to John’s phone later than night closely correlate (within 1-2 seconds) with the timestamps from John’s phone, indicating that the time variance might be much smaller. But as the prosecution’s expert pointed out in response (clearly correctly), those calls occurred while the Lexus was turned off, meaning the timestamps didn’t come from the Lexus’ clock but instead from Karen’s iPhone. Those calls also didn’t occur nearly as close in time to the relevant trigger event as the three point turn.

In short, there is every reason to believe that the prosecution’s estimation of the time variance is the correct one. Indeed, the prosecution’s estimation of a 21-29 second time variance is almost surely on the conservative side. Comparing the first trigger event with the GPS data related to the three-point turn strongly suggests that the variance was almost exactly 29 seconds.

So let’s return to the timeline. We know that John’s iPhone recorded its last movement at 12:32:16. But that’s not all the digital data tells us. Following their arrival at their destination, 34 Fairview, the Waze app stopped requesting high-frequency GPS data, so the GPS data recorded thereafter was more sporadic and less precise. For the rest of the night, the error radii of the GPS pings were wider, occasionally expanding to include the house. Karen’s defenders suggest that this means the phone could in fact have been inside the house. But that’s a terrible interpretation of the data.

Throughout the night, more precise GPS pings were also recorded, all indicating that the phone was located in the front yard, not the house. And every single ping recorded was consistent with the phone being located by the flagpole, in the very spot it was found the next morning. Remember, during this entire time period, the phone also recorded no movement whatsoever in the Apple Health data.

But that’s not all. In addition to the GPS and Apple Health data, the phone also recorded two other highly relevant types of data. First, as the prosecution’s expert Ian Whiffin explained, during the entire period from 12:32 to 6:04 a.m. the phone continuously recorded that it was in a “pocket state,” meaning that its camera was obscured (consistent with being inside a pocket).

Second, the phone recorded battery temperature data throughout the night. That data showed that the battery temperature began to drop after John got out of the SUV at 12:31 and continued to drop during the course of the night, eventually reaching an equilibrium somewhere just below 50 degrees. It then began to dip again after 6:04 a.m., when John’s body was discovered and the phone was found underneath him. The temperature dropped to a low of 37 degrees before ultimately rising again after Kerry Roberts picked it up and put it in her jacket pocket around 6:15 a.m.

This data is entirely consistent with the phone being stationary underneath John’s body all night long. We know from the bar surveillance footage that John kept the phone in his back pocket when it wasn’t in use. If the phone was in his back pocket when he was struck by the SUV, it would explain how it ended up underneath him. This would also explain why it was in a “pocket state” and why the battery temperature never reached the temperature of the outside air. It was pressed between John’s warm body and the cold ground, eventually reaching an equilibrium temperature between the two. It then began to dip again when exposed to the elements following the discovery of John’s body.

Indeed, the battery temperature data is so compelling that it forced the defense to entirely change their theory. Instead of arguing that John went into the house and/or basement (as they had for over two years), they argued in the second trial that he went into the garage, which was not temperature controlled.

But it’s not hard to spot the many irreconcilable flaws in this theory. First, the Apple Health data did not record enough movement for the phone to have reached the garage, much less made its way back to the flagpole area where it was found the next morning. Second, while the error range of the GPS data could perhaps allow for the phone to have been in the garage for a matter of minutes, it would have to have been promptly returned to the flagpole area. But we know from the battery temperature data that the phone could not have just been left in the yard, exposed to the elements. It had to have been shielded from the elements by something warm.

But beyond that, the garage at 34 Fairview is not even connected to the interior of the house. The only way into it is through a separate exterior door on the front of the house. The notion that people (including the family dog) left the warmth of the house and went into the dark, cold garage to wait for John’s arrival is facially absurd, as is the idea that John sprinted from the SUV toward the house (while checking his phone for messages) only to be lured into the garage and immediately ambushed. Even if it were remotely compatible with the digital data – which it clearly is not – none of this comes close to passing a basic plausibility test.

So let’s get back to the timeline. After John’s phone recorded its last movement at 12:32:16, the next data point we have is Karen’s iPhone logging into the Wifi at 1 Meadows (John’s home) at 12:36:39. Read’s defenders often assert that it would be impossible for Karen to have made it home in just four and a half minutes. To support this claim, they cite Google Maps data suggesting that this is a six minute trip. But this argument can be easily dismissed. There are multiple ways you can travel from 34 Fairview to 1 Meadows and all of them can be easily completed in less than four and half minutes, especially at night when there is no traffic. If you have any doubt about this, people have recreated this trip and filmed themselves doing so (just Google it). Moreover, Google Maps data assumes both normal traffic delays and that drivers are going roughly the speed limit. But Karen had just, indisputably, driven her vehicle at 24 mph in reverse for over five seconds, and (as we’ll get to in a minute) was both highly intoxicated and incredibly angry as she drove back to 1 Meadows. It’s absurd to assume that she would have been concerned about speeding.

Which brings us to the rest of the data from Karen’s phone. We know that, beginning at 12:33 (a minute after John stopped moving), and throughout her entire drive home, Karen called John continuously. At 12:37, just after pulling up to John’s house, she left her first voicemail, screaming “John, I fucking hate you.” She continued to call him incessantly (33 times) over the next hour, leaving him several other angry, intoxicated voicemails (“You fucking pervert!”; “You’re a fucking loser!”; “Fuck yourself!”)

We also know, from the Lexus data, that the SUV’s engine was running for six minutes following Karen’s arrival at 1 Meadows, a span bookended by her first two voicemails to John. We don’t know what she was doing during that time, other than calling John, but the length of time is notable, as is the fact that Ring camera footage from that timeframe appears to have been deleted. I strongly suspect that at least one of the things she did during those six minutes was inspect the damage to her vehicle and its taillight.

Before moving on to all of the other evidence, it’s worth noting one other important thing about the undisputed digital timeline data. It is absolutely inconsistent with the manufactured story Karen has been telling the world since mid 2023. At no point in the days following John’s death did Karen ever claim that he had gone inside the house. Nor did she act in a way that was consistent with her believing he could have gone inside the house. She never asked Jen McCabe or Brian Higgins or anyone else whether they saw John at the house or what happened to him. It wasn’t until August 2023, over a year and half after John’s death – and after Aidan “Turtleboy” Kearney had dutifully unveiled the defense’s conspiracy theory – that Karen first debuted her lawyer-crafted version of what happened that night.

It began with a well-rehearsed interview with ABC’s Nightline in which she claimed that, as soon as they arrived at 34 Fairview, John jumped out of the car and she watched him walk up and into the house. She claimed that he was going to check and make sure that they were at the right place and were welcome. After a while, when he didn’t call or come back out to get her, she said that she grew frustrated and tried to call him. He didn’t answer. Eventually, after 10 minutes of waiting, she got fed up and left.

The undisputed digital evidence proves that almost all material elements of this story are false. It is beyond dispute that Karen was only at 34 Fairview for eight minutes and that, for the vast majority of that time, John was in the car with her. It is beyond dispute that she never texted or called him until after she had left 34 Fairview. And the digital data from John’s phone is simply incompatible with him having walked up to and inside the house (at least while carrying his phone).

As I’ve explained at length previously, it is very likely that, in the process of constructing this story, Karen and her defense team were unaware of or overlooked the fact that her phone logged into the WiFi at 1 Meadows at 12:36:39. They likely believed that the first evidence of her presence at 1 Meadows was the 12:42 voicemail she left John, in which you can hear her getting out of the vehicle inside the garage. In other words, they mistakenly believed that the digital evidence allowed for her to have remained at 34 Fairview until around 12:37. So they thought they could credibly claim that she was outside of 34 Fairview for 12-13 minutes in total and that she called John while she was waiting.

This false timeframe also tracks with another claim the defense was pushing at the time, that footage from the Canton Library security camera was mysteriously missing from 12:37-12:39 and that this missing footage would have shown her driving back to 1 Meadows with an intact taillight. The implication was that the conspirators who were framing Karen had deleted this exculpatory footage. As we know now, however, the library camera is motion activated (nothing was deleted) and Karen was already back at 1 Meadows before 12:37. Indeed, there’s no reason at all to think she even passed the library on her way back (there are multiple different routes she could have taken).

So before we even get to the physical evidence, the undisputed digital evidence alone establishes that John was only outside of Karen’s vehicle for 20 seconds before he stopped moving, that he never went anywhere near the house, and that during those 20 seconds, Karen gunned her vehicle in reverse for over five seconds, achieving a speed of at least 24 mph. The digital evidence also establishes that, from the moment it stopped moving, John’s phone was outside all night, by the flagpole, under his body.

The Physical Evidence Is Equally Damning

One of the most frequently repeated (and insanity inducing) claims made by Karen’s defenders is that “there’s no evidence of a collision.” The people who say this have no idea what the word “evidence” means or what kinds of evidence are typically used to convict people in hit-and-run cases.

First, and most obviously, John O’Keefe’s body was found amidst a debris field of broken taillight pieces that came from Karen’s SUV. Tiny shards of Karen’s taillight were entangled in his clothing. His missing shoe and hat were also found by the curb, amidst the taillight debris. His DNA was found on the taillight fixture, and one of his hairs was attached to the back of the SUV. This is exactly the kind of evidence that is typically used to convict defendants in hit-and-run cases. Indeed, it’s such strong and conclusive evidence of guilt – in and of itself – that the defense’s only option was to make the incredibly bold claim that all of this evidence was planted by law enforcement.

But even with the firing of Michael Proctor (the state trooper assigned to manage the case), there has never been any credible indication of evidence planting or tampering in this case, and the timeline of events simply does not allow for it.

Let’s start with some very elementary logic. There was no damage to Karen’s taillight on January 28, 2022 (it was recorded on video multiple times that day). As of 5:07 a.m. on the morning of January 29, 2022, significant taillight damage can clearly be seen to the right taillight of the vehicle as it pulled out of the driveway at 1 Meadows. A huge challenge for Karen’s defenders has always been finding a non-incriminating explanation for how that damage occurred.

The story they eventually settled on (and maintained throughout the second trial) is that Karen damaged the taillight at 5:07 a.m., when her SUV lightly tapped John’s Chevy Traverse as she backed out of the garage, an incident that was captured on Ring camera footage. That claim was always pretty far-fetched, as the nudge was barely perceptible, occurred at negligible speed, and no taillight pieces can be seen falling into the pristine snow on the driveway.

But in the second trial, the prosecution completely obliterated this claim through expert testing and testimony. Dr. Welcher of Aperture performed laser scans of both vehicles and used sophisticated computer modeling to show that the parts of the vehicles that came into contact were nowhere near the right taillight and, in fact, the shape of the vehicles would not have allowed the right taillight to have made contact with the Traverse in any scenario. In addition (and supported by the prior testimony of the defense’s own ARCCA experts), the negligible speed of the impact (< 1 mph) would not have been enough to break the taillight even if the shape of the vehicles had allowed for contact.

In response to this conclusive debunking, the defense offered zero counter-evidence or analysis. Their crash reconstruction witnesses didn’t even address it.

So we know, with absolute moral certainty, that the taillight was not damaged in the driveway at 1 Meadows. It was already broken. We also know that the vehicle was powered off the entire night and inside the garage. And we know that the three people who spent five minutes parked directly behind the SUV at 34 Fairview the night before saw no damage to its taillight. That means the taillight had to have been damaged after the D’Antuono truck pulled away (around 12:31 a.m.) and before Karen got back to 1 Meadows (at 12:36 a.m.).

So we can be incredibly confident that the taillight was in fact broken at 34 Fairview, right where all of the taillight pieces were subsequently found.

And once you’ve established that some of the taillight pieces had to have been at the scene the entire time, the claim that law enforcement planted any taillight pieces completely falls apart. As should be obvious to anyone with a pulse, the planting of taillight pieces is an all or nothing proposition. There is no universe in which anyone – regardless of their motives – would ever think that breaking off more pieces of taillight and planting them alongside pieces that were already at the scene would somehow make a suspect look more guilty or strengthen the case against them. That is a ludicrous proposition.

But the rest of the taillight planting story makes just as little sense and is just as irreconcilable with the timeline. By the time Michael Proctor and his boss, Sgt. Bukhenik, left Good Samaritan Hospital, Karen and her father had already driven the SUV and whatever remained of its taillight down to Dighton. And, by the way, we don’t have to guess at the state of the taillight at that point. Its status was captured on dashcam video by the Canton PD (unknowingly) when they stopped by 1 Meadows that morning at 8:23 a.m. to do a wellness check on John’s niece and nephew. And you can clearly see in the picture below (which Karen’s supporters have gone through Herculean mental gymnastics over the years to avoid acknowledging) that a massive piece of the taillight was missing at that point and that the damage appears to align exactly with the damage documented after the vehicle was taken into custody. Karen’s online supporters often claim that this image was photoshopped, but it was not. It’s a still shot from a longer video, and Karen’s defense team has never even alleged that it was tampered with.

Proctor and Bukhenik did not arrive in Dighton until after 3:00 p.m. that day and did not take the vehicle into custody until 4:12 p.m., when it was loaded onto a tow truck. This event was also captured on video, and while the angle and distance make it tougher to tell the precise extent of the damage, it’s clear that a big chunk is missing from the taillight.

It is at this point that Officer Nicholas Barros of the Dighton Police Department enters the story. While a minor, forgettable character in the first trial (called briefly by the prosecution to corroborate the testimony of Proctor and Bukhenik), in the second trial, Officer Barros was called by the defense and offered seemingly explosive testimony. He claimed that the damage to the taillight that he observed in Dighton was much less extensive than the damage later documented by the forensics team in the Canton PD sallyport. To this day, Karen’s supporters claim that Barros’ testimony is hugely significant and somehow proves that her taillight was tampered with by law enforcement.

This is complete nonsense.

I’m not going to question Officer Barros’ motives or suggest that he lied or intentionally tried to mislead the jury in the second trial, but there’s no question that his testimony on direct examination was inaccurate. Using the laser pointer, he specifically testified that the rear-facing portion of the taillight was completely intact when he observed it in Dighton, that the only damage was to the lower portion of the passenger-side-facing part of the taillight. But the portion he described as being fully intact at around 3:30 p.m. that day is clearly and indisputably gone in both the 5:07 a.m. Ring footage and the 8:23 a.m. dashcam footage (see image above). And sure enough, when confronted with this on cross-examination, Barros completely crumbled, looked confused, and ultimately agreed that those images were consistent with what he saw that day. He also seemed to shrink back into the stand when confronted with the fact that he completely misremembered and misrepresented his testimony from the first trial, which he bafflingly seemed not to have watched in preparation for the second trial. In short, Barros’ testimony was an incoherent and self-contradictory mess and should not be interpreted by any rational person as evidence that Karen’s taillight was tampered with.

But here’s the most important part of the taillight saga. By the time the SUV was towed back to Canton PD Headquarters (arriving 5:36 p.m. that night), the state SERT team was already at 34 Fairview and had begun their grid search of the area by the flagpole, which was covered in over 3 feet snow piled up by the plows. By 5:41 p.m. that team had found the first pieces of taillight, under all of the snow, flush against the curb. They subsequently found John’s missing shoe in the same area.

At trial and in the media, Karen’s defenders have spent an inordinate amount of time talking about surveillance footage of the SUV from the Canton PD sallyport (more on that below). But it’s important to understand the timeline here. Both Proctor and Bukhenik were at Canton PD when the first taillight pieces were found across town at 34 Fairview. There is a vanishingly small window of time, if any, in which any taillight pieces could have been taken from the vehicle, driven to 34 Fairview, and somehow buried under the snow for the SERT team to find. And it would have required an unidentified (and very skilled) accomplice to execute, almost surely someone actually on the SERT team. And this mystery accomplice would have had to bury the pieces (and shoe presumably) right under the noses of the rest of the SERT team, under the glare of spotlights and a bunch of onlookers, including local media. We’re talking Mission Impossible level degree of difficulty.

For obvious reasons, Karen’s defenders want to ignore the timing and sequence of these events. They want to focus on taillight pieces found on subsequent days and ignore the fact that the first taillight light pieces were found by the SERT team on the very first day, at street level, under 3 feet of undisturbed snow. Indeed, from the very first Turtleboy article in April 2023, there’s been a constant assertion – which never really made any sense – that it is somehow inherently suspicious that the taillight pieces weren’t found at the time John’s body was discovered and were instead found, a few pieces at a time, over a multi week period.

This argument would make a lot more sense if John’s body was found on a clear summer day instead of at the onset of an historic blizzard, one that produced the biggest snowfall ever recorded from a January storm. At the time Canton police were attempting to recover evidence from the area where John’s body was found, it was snowing incredibly hard and the wind was swirling. Everything was covered in several inches of snow, with even more along the curbs due to the plowing of the streets. The only area searched at that point was the area immediately around the body, and even that area was not searched thoroughly due to the conditions.

Remember, in Massachusetts, the local PD doesn’t have jurisdiction over these types of situations (unattended deaths), nor training on how to deal with them. The Canton PD was forced to take some initial investigative steps that morning because the weather conditions prevented the state police from responding to the scene in a timely way. But the rapidly deteriorating weather made any kind of systematic search impossible. Any taillight pieces would necessarily have been under the snow, and the Canton PD didn’t even know to be on the lookout for taillight pieces at that point, because no one from law enforcement had yet observed or were aware of Karen’s taillight damage.

So the first actual attempt to look for taillight debris occurred later that afternoon, by the state SERT team. And due to the conditions, they limited their search to a few car lengths of area by the curb. The other pieces of taillight were found on subsequent days, as the state police focused on clearing out the snow in different areas. Finally, when the snow eventually began to melt, additional taillight fragments were revealed over time, as the snow receded. It is likely that some of these last pieces were either overlooked in the early searches or had been moved from their original resting spots by interaction with plows or other vehicles.

One thing that is certain is that no one attempting to frame someone for a crime would decide to sprinkle out the taillight pieces over days and weeks. Spacing out the planting of evidence makes no sense at all. There would be no point to it, and each attempt would bring additional risk of being observed. And it’s not as if each incremental piece of taillight discovered somehow made the case against Karen stronger. Turtleboy and others have often tried to make a big deal out of the fact that one of the last taillight pieces was found by former Canton Police Chief Ken Berkowitz as he drove by the scene (he alerted the state police team, who recovered it). They assert, without explanation, that this is somehow incredibly suspicious. But what is the actual allegation here? That Proctor saved a piece of taillight to give to the police chief to plant days later? How would doing so possibly further the goal of framing Karen Read? What possible rationale could Proctor have had for doing that? None of it makes any sense at all.

And this gets us to the question of why: why would Proctor or any other member of law enforcement go to such extraordinary lengths to frame an innocent woman for the death of John O’Keefe, a fellow cop who was, by all accounts, universally beloved? This is a question Karen’s defenders tend to ignore entirely or handwave away with vague references to “corruption” or the “blue wall.” Some even invoke, as a comparison, the case of Sandra Birchmore, a local woman who was killed by a cop in a neighboring town (Stoughton), who then staged the death to look like a suicide. There have been legitimate questions raised about the quality and thoroughness of the state police investigation into that matter, which involved some of the same troopers as the Read case and initially concluded that Birchmore’s death was a suicide (the Feds ultimately intervened and charged the Stoughton cop with murder).

But the Birchmore case actually illustrates the type of motive that needs to exist if you’re going to posit extraordinary police conduct. In that case, the officer who staged the murder to look like a suicide was the very same officer who committed the murder. He had long been having an affair with Birchmore (dating back to when she was a minor) and she was pregnant and threatening to expose him at the time she was killed. So he was covering up his own heinous crimes. The state police may have been slow to see that crime for what it was, but I’ve seen no credible allegations that they planted or tampered with evidence or engaged in any kind of corrupt cover up in that case (and the Feds definitely would have looked into that).

In the Read case, the allegations of corruption center around former state trooper Michael Proctor, who lives in Canton and happened to be the trooper on call the morning John O’Keefe’s body was found. When the conspiracy theories first took off in 2023, the defense team (aided by Turtleboy and his online followers) attempted to portray Proctor as being a close family friend of the Albert family. The evidence behind this claim is super weak. Proctor’s sister, Courtney, is a good friend of Julie Albert, the wife of Chris Albert and mother of Colin Albert, none of whom were at 34 Fairview during the relevant time that night. Chris is the brother of Brian Albert, the homeowner. There are pictures of a young Colin Albert as part of the wedding party at Courtney Proctor’s wedding, pictures which also include Michael Proctor.

Based on this attenuated connection, the defense has asserted that Michael Proctor went to extraordinary lengths – from the very first second he was on the case – to plant incriminating evidence against Karen and destroy exculpatory evidence. But why? Even if you assume the laws of space and time would have allowed him to do this, why would Michael Proctor risk his livelihood and freedom to protect the brother of his sister’s friend’s husband? The obvious answer is that he wouldn’t, that no one in the history of crime and corruption has ever gone to such lengths based on such insufficient motive. This is why Karen and her defense team spent years arguing that Colin Albert – who was a 17 year old high school student at the time – killed John O’Keefe. It never made any sense, and Colin never had any plausible motive, but if Colin was the culprit, it at least made Proctor’s alleged behavior marginally less inexplicable. Rather than engaging in extraordinary criminal misconduct to protect his sister’s friend’s husband’s brother (who it was unclear he’d ever met), he was instead doing it to protect his sister’s friend’s son (whom he’d at least been photographed with years earlier).

But even if you find that a remotely compelling motive, the reality is that Colin wasn’t at 34 Fairview at the same time as John O’Keefe and had no conceivable motive to attack John O’Keefe. Karen and her supporters tried for years to come up with something, anything, that might help them in their efforts to cast suspicion on Colin and came up completely empty, so much so that they were barred from even making this argument at trial. Simply put, Colin had nothing to do with anything.

But Karen and her defenders still insist that Michael Proctor planted all of the incriminating physical evidence against Karen. Their only evidence of this is the fact that Proctor was fired from the state police based on his conduct in this case. That sounds superficially compelling until you look at what that conduct was and how it surfaced. In an incredibly unusual move (which I’ll discuss in more detail below), the Boston U.S. Attorney’s Office opened their own investigation into this matter in late 2022, at the urging of Karen’s defense team. Using grand jury subpoena powers, they were able to obtain and review years’ worth of Proctor’s private communications, which were then turned over to Karen’s defense team and the Commonwealth just prior to the first trial in 2024.

Many of Proctor’s private text messages were deeply embarrassing. He shared details of an active investigation with his high school buddies and, most notably, he made a number of highly disparaging and unprofessional comments about the primary suspect in that investigation, Karen Read. Proctor was ultimately fired because of these transgressions.

But with respect to the question of Karen’s innocence or quilt, the key takeaway from Proctor’s private communications is that he clearly thought she was guilty. Those messages are not remotely consistent with a reality in which Proctor knew Karen was innocent and was trying to frame her for the crime. And it’s not as if he was being circumspect in his communications. He was oversharing and clearly saying things he never thought would ever see the light of day.

It is not possible to read those texts and rationally come away convinced that Proctor was orchestrating some sort of cover up or frame job. Undoubtedly, the Feds reached the same conclusion, as they ultimately closed their investigation without charging anyone with anything.

Finally, with respect to the Proctor and the notion of a law enforcement cover up, it’s important to emphasize that most of the key evidence in this case, in particular the digital evidence from the iPhones and the data from the Lexus, could not have been fabricated and would not even have been known at the time Proctor and his accomplices would have to have formed their conspiracy and attempted to frame Karen. They couldn’t possibly have known about the techstream trigger events or the GPS or Apple Health or battery temperature data from John’s phone. But they were experienced detectives, and they knew that these data sources existed and would eventually be analyzed. And they would have known what types of data can be extracted from modern cars and from iPhones.

In other words, they would have known that if they tried to frame Karen Read for a hit-and-run collision that never occurred, that the data eventually extracted and analyzed from the SUV and the phones would very likely tell a different story. For example, perhaps John’s phone data would show that he had gone in the house and/or interacted with his phone after Karen had already left the scene. And perhaps the vehicle data (or other sources) would show that it was in a different part of town when John died or that the taillight had been broken in some other way at some other time. It certainly wouldn’t show any evidence of a collision at 34 Fairview.

But that’s not what happened. To the contrary, every bit of new data that has been obtained and analyzed has perfectly aligned with the narrative that Proctor and his accomplices supposedly made up that first day of the investigation. The phone data showed that John had never gone inside the house. The vehicle data showed that Karen had gunned her SUV in reverse, right in front of 34 Fairview, mere seconds before John recorded his last movement. No conspirators are that lucky.

Karen’s Behavior is Irreconcilable with Innocence

There’s a reason I haven’t discussed Karen’s behavior much until this point. That’s because I want everyone to understand just how strong the case is against her before you even begin to consider her statements and behavior. But make no mistake, Karen’s statements and behavior from day one are those of a guilty person, not an innocent one.

Remember, as I noted above, that from the moment John recorded his last movement at 12:32:16, Karen did nothing but call his phone incessantly. She called him 33 times over the next 45 minutes, before apparently passing out. During this time period, she also called her parents, twice (they didn’t answer). When was the last time you called your parents after 1:00 a.m.?

She then resumed calling John incessantly when she woke up just after 4:30 a.m. This is not the behavior of someone who (as she would later claim) watched John go inside the house at 34 Fairview. This is the behavior of someone who had reason to fear, from the moment she drove away from the scene, that something bad might have happened to John.

I am not suggesting that Karen necessarily intended to hit John or even knew with any certainty that she had hit John. What I’m suggesting is that she almost surely knew that she had hit something and at least feared that it might have been him. This fear would have been compounded when she got back to 1 Meadows and noticed the state of her taillight.

After she woke at 4:30, in addition to calling John incessantly, Karen did a number of other very incriminating things. First, as confirmed by John’s niece and the various people she called that morning, she was in a state of panic that was entirely disproportionate to the situation she later claimed to have been in. Had she really seen her intoxicated boyfriend go inside someone’s house mere hours earlier and then left him without a ride home, the far more logical assumption to make would have been that he was asleep on a couch. But Karen clearly assumed that he was not in a house at all.

One of her first calls that morning was to Jen McCabe (she had John’s niece call Jen because Karen didn’t have her number). Rather than asking Jen what happened to John after he went in the house, she told Jen that she’d last seen John at the Waterfall. Once Jen corrected her and pointed out that she’d seen Karen’s SUV pull up outside 34 Fairview, Karen hung up.

She next called John’s long time friend Kerry Roberts. She told Kerry that John was dead, that he may have been hit with a plow. Not only was this a very different story than she’d just told Jen McCabe minutes earlier, but it was an incredibly bizarre claim to make if she’d seen John go into the house after she dropped him off.

Minutes later, Karen backed out of her driveway and lightly bumped the back for her SUV into John’s Traverse. It’s impossible to know whether that was intentional, but certainly her internal censors and warning sounds would have been blaring at her as she slowly approached the other vehicle in reverse. Karen’s whereabouts over the next 20 minutes are also unaccounted for, and it certainly seems like she at least attempted to retrace her route from the Waterfall to 34 Fairview before eventually heading over to Jen McCabe’s house.

As we know from uncontested testimony (as well as Ring footage), Karen then met up with Jen McCabe and Kerry Roberts before heading back to 1 Meadows. At this point, we know from the testimony of all three women (Karen’s testimony in the form of her interview with Gretchen Voss) that Karen showed the other two women her broken taillight and that a lot of it was missing, with (in Karen’s words) wires exposed.

What happened next is by far the most incriminating sequence of events. With Kerry driving and Karen in the backseat, the three women drove together to 34 Fairview. As they approached the house from the South side, in the pitch black, with snow swirling, Karen shouted from the backseat for Kerry to stop the car. She said that she saw John. Kerry and Jen couldn’t see a thing, and both testified that they initially thought Karen was crazy. Karen, per her own words in the I.D. documentary, saw a “weird shaped hump” of snow which she somehow knew was John. In that same interview, she said “I was looking to find him, I was expecting to find him.” And, sure enough, there he was, covered in snow, by the very flagpole where they had been parked the night before.

In the minutes that followed, as various paramedics and police officers arrived, Karen made a number of very incriminating statements. At trial, two different paramedics (in addition to Jen McCabe) testified that, when asked what happened to John, Karen said “I hit him” multiple times. Others on the scene heard similar statements at different times (sometimes in the form of a question; other times as a declaration). Karen herself, in the I.D. documentary, seemingly admits to having said she hit him and, regardless, has never disputed that she at least was asking if she hit him at this point.

Just as incriminating are the things she didn’t say and the questions she didn’t ask. At no point did she ever suggest, to anyone, that John went inside the house. At no point did she ask Jen McCabe, or anyone else, what happened to John after he went into the house. Later that day, she texted Brian Higgins (who was also at the house the night before) and wrote “John’s dead.” She didn’t ask Higgins what happened to John after she dropped him off.

Later that morning, as John’s friends and family (including his adopted niece and nephew, whom Karen was very close with) were gathering at John’s house to grieve and mourn him, Karen arrived and went immediately upstairs to gather some belongings. She then immediately left without socializing at all, without trying to console John’s niece or nephew, and she and her father decided to drive Karen’s damaged SUV all the way down to Dighton (over an hour away in good conditions) in the height of the blizzard.

When the state police (Proctor and Bukhenik) eventually made it down to Dighton to interview her that afternoon, she shut down the interview pretty quickly, having already reached out to criminal defense counsel before they arrived. At no point did she state that John had gone inside the house or suggest that the police investigate the people who were there. When they confiscated her iPhone, she did not provide them (that day or ever) with a password, which kept them from being able to access its contents for many months (until a forensic software update eventually allowed them to).

When Karen was arrested a couple of days later, her counsel (David Yannetti) emphasized during her arraignment, and afterward to the media, that his client had no criminal intent. He never suggested that the state had the wrong person. The clear and intended implication was that John’s death had been a tragic accident.

Indeed, it wasn’t until a full year and half later, in August 2023, that Karen debuted to the world her new story, that she was innocent, that she’d seen John go into the house, and that she was framed by law enforcement. And as detailed above, we know from the undisputed digital evidence that the story Karen rolled out in August 2023 was a manufactured lie. None of it is remotely reconcilable with the digital data.

No Possible Alternative Theory is Remotely Plausible or Reconcilable with the Evidence

Turtleboy’s first “article” on this case – published back in April 2023 – offered the first of many attempts to construct an alternative theory of what might have happened to John. In light of the role the defense team itself played in constructing and rolling out that story, it’s remarkable to look back on it now and see how ludicrous it is. Putting aside the fact that almost all of the key pillars of that story have since been proven false, the story itself is one of the more incoherent and implausible tales I’ve ever seen anyone put forward with a straight face.

In that story, Turtleboy claimed that John went inside the house where he was – for reasons unstated – beat up by Brian Albert and his nephew Colin, with the family dog Chloe joining the fracas at some point. The Alberts, along with Jen and Matt McCabe, then decided to cover everything up and somehow, someway, pin it on Karen. The plan that veteran police detective Brian Albert eventually came up with was to drag John’s body out onto his own front lawn and just hope for the best. Thanks to Karen’s incredibly fortuitous breaking of her taillight later that morning and the quick thinking of a cadre of corrupt law enforcement officers at both the state and local level, they were then able to somehow pin everything on Read. And as the coup de grace, criminal mastermind/soccer mom Jen McCabe was somehow able to jedi mind trick Karen that morning into believing that she had killed John.

If this sounds like the plot of the worst Lifetime movie ever written, I think you’re being a little too hard on the Lifetime script writers. They’ve never written anything this bad, something where the motivations and decisions of the main characters make less sense, where the plot has bigger holes or relies on more absurd coincidences.

And that was back in 2023, before most of the evidence in the case was even known. Faced with the actual evidence, the defense’s story has shifted dramatically since then. By trial number two, Colin Albert had been written out of the plot, and the story — to the extent there was one — had come to focus on someone who was not even featured in Turtleboy’s original theory, ATF Agent Brian Higgins. And instead of the basement, the supposed altercation was theorized to have taken place in the garage. And because of the lack of any evidence that John was “beaten up,” the suggestion was instead that he had been pushed back and hit his head on something.

But as I’ve noted in detail above, all possible permutations of this story suffer from the same fatal defects. First, there was simply no time for any of these things to have happened. John was only outside of the vehicle for 20 seconds before he stopped moving forever. And the digital evidence from the phone makes clear that he didn’t go anywhere near the house, that he was lying on top of his phone, by the flagpole, from 12:32:16 onward. There’s just no time for a dog attack or a heated altercation with persons in the house. The only remotely plausible alternative scenario that ever existed was one involving an altercation that took place outside, after John was dropped off, but the more precise digital timeline rules that out too.

But where the alternative explanations have always fallen down is that they all require a series of impossibly coincidental circumstances for the crime ever to have been pinned on Karen in the first place, as well as the creation and maintenance of an unprecedentedly large conspiracy in the aftermath.

Karen’s defenders, understandably, like to focus their attention on Michael Proctor. But there’s no universe in which Michael Proctor – even if he had a plausible motive for doing so – could ever have framed Karen Read by himself. Any fair reading of the evidence (and how and when it was discovered) makes clear that Proctor would, at minimum, have needed the help of multiple fellow members of the state police and likely members of the Canton PD as well.

Moreover, there were nine people in the house at 34 Fairview at the relevant time that night. Any remotely plausible alternative scenario requires all nine of those people to have lied about what happened that night. And not just once, but under oath, in front of state and federal grand juries, across 3+ years and two trials, and in the face of unprecedented media scrutiny and relentless personal harassment.

And keep in mind, there was a federal investigation in which all of these people were hauled in front of a federal grand jury and grilled about their stories. Two of the people who were in the house that night – Sarah Levinson and Julie Nagle – are not family members; they were just friends of Brian Albert Jr. who happened to be there. At the time they were being questioned by the FBI and hauled before a federal grand jury, don’t you think their parents sat them down and pressed them to come clean if there was anything they were holding back? I assure you that these witnesses would have been the first people approached by federal prosecutors and offered immunity if they had anything incriminating to offer against the likes of Brian Albert or Brian Higgins (or anyone else who was at the house). And their parents and any legal counsel they consulted at the time would have browbeat them into taking such a deal if they in fact had anything like that to offer.

Anyone who has any familiarity with actual criminal conspiracies knows how absurd it is to believe that a conspiracy of this size and scope could withstand three and half years of the most intense scrutiny imaginable, particularly when the supposed conspirators are not a part of some criminal syndicate but are just random suburbanites who happened to be drinking at a house one night.

Anyone who was paying close attention at the second trial likely noticed that Karen’s defense team was careful not to articulate any specific or well fleshed out alternative theory of what happened. There’s an obvious reason for that. It is not possible to construct an alternative explanation of what happened that night that is remotely consistent with the undisputed evidence in this case. If you have any doubt about this (and you’re a masochist), I challenge you to spend as much time as you can stomach reading or listening to the countless pro-Karen Read content creators who have been grifting off of this case for the last three years. Try to find even a single one of them who has ever managed to come up with a coherent alternate theory of what happened that night. You will not find one. Most of them have never bothered to try, and the ones who have simply ignore all of the evidence they find inconvenient.

So How Did the Defense Win?

If you’ve stuck with me this far and you weren’t particularly familiar with the details of this case before now, you’re probably wondering how on earth this case ended in an acquittal. I’m still struggling with this question myself, but I believe it comes down to two things. First, the relentless attention this case received in traditional and social media (particularly the latter) created a jury pool full of people with preconceived ideas about this case. And because the defense and its surrogates have been enormously successful in steering that media conversation, those preconceived ideas tended to be favorable to the defense.

As someone who lives in Norfolk County, it’s still amazing to me how little my fellow residents actually understand the facts and evidence in this case. A huge percentage of people around here formed their opinions about the case back in 2023, when the conspiracy theorizing was out of control and most of the actual evidence was still not public. And most have failed to meaningfully re-examine their beliefs (or the evidence) since then.

But the other explanation is one of trial strategy. The defense did a really good job, especially in the second trial, in one particular area. They were successfully able to frame this case in a way that put an impossibly high burden on the prosecution. The defense was able to convince the jury (and much of the public) that in order to convict someone in a hit-and-run scenario like this – where there were no eyewitnesses or video footage of the incident – the prosecution needs to commission scientific testing sufficient to prove, to a scientific certainty, the precise mechanics of the alleged collision. That proposition seems at least superficially reasonable until you consider what it would actually entail.

In the first trial, the prosecution fell short largely because it failed to invest in outside expert witnesses (who are expensive) and instead relied on state troopers and the local medical examiner, who were not very effective witnesses. Thanks to some help from the Feds (who retained external crash reconstruction experts from ARCCA) and the media attention the case garnered (which attracted publicity seeking experts like Dr. Russell, the dog bite lady), the defense was able to “out-expert” the prosecution in that trial. While the defense’s wild conspiracy story largely fell flat, their experts were sufficient to get them to a hung jury.

In the second trial, the defense didn’t even bother trying to tell a story (because it was impossible to tell a coherent one), but they wisely doubled down on their expert strategy, bringing in credentialed and articulate medical and scientific experts who were willing to state authoritative conclusions that went far beyond what their testing or the science warranted. In particular, these experts were successful in planting the idea with the jury that unless the prosecution can prove, through scientific testing, the exact mechanics of the alleged collision, nothing else matters.

The problem with that argument is that it creates an impossible standard, one that, if accepted, would essentially make it impossible to ever convict anyone of a hit-and-run absent video footage or direct eyewitness testimony. What’s ironic is that the prosecution in this case actually took the highly unusual (and expensive) step of hiring well-respected external crash reconstruction and medical experts (and were repeatedly criticized by the defense and in the media for the cost involved in doing so). But as Dr. Welcher (the prosecution’s crash reconstruction and biomechanical expert) correctly explained, while the digital and physical evidence can tell us with a high degree of logical and scientific certainty that John O’Keefe was hit by Karen Read’s SUV, there is no practical way of determining the exact mechanics of that collision.

The reason for that is pretty simple to understand. There are far too many unknown variables here to allow for meaningful testing. We can’t possibly know precisely where John was standing when he was hit, what position his body was in, how fast the SUV was moving, what direction the vehicle was moving relative to his body, or where precisely the vehicle made contact with his body. There are almost infinite permutations possible and slightly tweaking any of the variables can result in wildly different outcomes. It is not remotely feasible for the DA’s office to purchase a fleet of $90,000 Lexus SUVs and to keep running tests until they get it right. The defense, which had far more resources to spend on these things, ended up destroying their exemplar Lexus in one of their poorly designed tests (in which the crash dummy was struck in a way that no one has ever alleged John was). If this kind of trial-and-error testing is needed in order to convict someone of a hit-and-run, no one would ever be convicted of that crime. Even doing one such test would be prohibitively expensive in almost all normal situations, much less the hundreds or thousands of tests it would take to control for all the variables here.

By contrast, the defense experts ran a handful of tests – while subtly tweaking the variables in ways that favored their preferred outcome – and simply opined that the results were inconsistent with the damage and injuries actually observed in this case. That’s essentially the equivalent of dropping two plates and noting that each one broke differently. But it can be an effective strategy, particularly when you have experts who are willing to confidently state conclusions that go well beyond what their testing actually proves while simultaneously hand waving away all compelling evidence to the contrary (such as the techstream data, taillight debris, etc.).

The key things to understand about the defense’s experts are 1) that these are professional expert witnesses who are known (particularly in Dr. Laposata’s case) for their willingness to opine authoritatively in whatever way their client wants them to, and 2) the defense went out of its way to make sure that these experts were unburdened by any meaningful understanding of the full evidence in this case. In other words, it is much easier to express skepticism that a collision occurred here when your analysis is limited to a subset of the evidence and you are not exposed to the evidence that logically rules out all other possibilities.

The trick here is to pretend (as the ARCCA witnesses did) that the digital and physical evidence is somehow not actual evidence and doesn’t need to be factored into to your “scientific” conclusions about what happened. I can’t tell you how many times I’ve heard Karen’s supporters say things like “I trust the science here, and the science says there was no collision.” Putting aside how incredibly naive it is to simply treat the testimony of paid defense experts as gospel truth, the reality is that no legitimate scientific approach to problem-solving would ever simply ignore or handwave away so much critical data and evidence. Good science relies just as much on data and observations and logic as it does on experimental testing.

The defense’s crash reconstruction and medical experts intentionally chose not to grapple in any meaningful way with the logical implications of the digital or physical evidence in this case. They instead focused myopically on the parts of the evidentiary record that made it easier for them to express the opinions the defense team wanted them to express. That is not science. That is advocacy.

---

But What About…?

Lest I be accused of ignoring any of the key evidence or arguments in support of Karen’s innocence, I want to use the remainder of this piece to go through the various the arguments I hear the most often. Again, I want this post to be a comprehensive resource, so if you can think of any other arguments that I haven’t addressed, let me know in the comments. I’ll be happy to update this post over time.

How do you explain John’s arm injuries (or lack thereof)?

This is easily the most compelling argument offered by the defense, in large part because there’s just never going to be enough information to answer it definitively. What we know is that John had superficial abrasions on his arm, above and below the elbow, that he incurred prior to his death. We also know that the broken taillight on the SUV was at roughly the height of his elbow and that the scope of the abrasions approximately corresponds with the size of the taillight. And thanks to the defense’s own testing, we know that contact with a human elbow – even at speeds considerably below the speed Karen was traveling during the trigger event – is more than capable of causing damage to the taillight comparable to what was observed on Karen’s SUV.

But were the abrasions caused by uneven parts of the surface of the light? By broken edges of the light after it began to crack? By shards of taillight that somehow got inside John’s sleeve? Or is it possible (as I and others have theorized) that the light was already broken prior to contact with John’s arm (perhaps after John threw the cocktail glass he was holding at it) and it was the jagged edges of an already partially broken taillight that caused the abrasions. We’ll never know the answer to this question, but given the very precise and well-established timeline of evidence, I think we can have a high degree of confidence it was one of the possibilities above.

I also think we can very quickly rule out the possibility that the abrasions were due to dog bites or scratches. The only “evidence” for this theory is the testimony of Dr. Marie Russell, a kooky woman who latched onto this case as a way of publicizing her availability as an expert witness, whose opinion is based solely on looking at a single photograph, and whose testimony has significantly changed each time she’s given it (e.g., giving inconsistent explanations as to whether various scratches were the results of teeth or claws). Being as generous as possible to Dr. Russell, I think she saw a photograph and thought “that could be a dog bite” and, in order to secure the gig, that opinion eventually morphed into “I’m confident that’s a dog bite.”

But the reality here is that these are just abrasions. There’s no real methodology behind Dr. Russell’s opinion, nothing distinctive she can identify that points to a dog as opposed to a million other potential causes. There was no canine DNA found on the shirt. And the timeline here simply doesn’t provide any opportunity for a dog to have played any role in what happened.

The most compelling argument the defense made at trial, however, was not about the abrasions, but about the lack of more serious injuries. They pointed to the lack of broken bones observable in the autopsy x-rays and the lack of bruising. Again, though, these might be more compelling arguments if the digital and physical evidence didn’t so decisively prove that a collision occurred.

As to broken bones, while it’s certainly possible (maybe even probable) that an arm would break in most cases when colliding with a moving SUV, even the defense’s experts didn’t testify that it’s a foregone conclusion. A lot depends on the speed of the impact (which is not known) and the position of the arm/elbow at impact. The impact in this case certainly could have occurred at far less than 24 mph (particularly in a scenario where the taillight was already partially broken via prior impact with the cocktail glass). And a closed elbow is capable of withstanding quite a bit of force without breaking (just look at any martial demonstration). The x-rays also don’t tell us anything meaningful about what other damage may have been incurred to the arm (muscle tears, ligament damage, etc.). So the lack of broken bones does not come anywhere close, in my opinion, to creating any reasonable doubt about what happened here.

Similarly, while this issue was not particularly well explained by any prosecution witnesses, I think the lack of bruising is also not as big a deal as Karen’s defenders seem to think it is. All of the same considerations noted above apply here (i.e., we don’t know how fast the SUV was going or where the point of impact was on the arm). I think it’s also worth considering that John was lying on frozen ground in sub-freezing temperatures, which could slow bruising, and that he may not have been alive for all that long (dead bodies don’t bruise). Which brings me to the next issue.

Why didn’t he show signs of hypothermia or frostbite?

The defense’s medical expert testified that John didn’t show signs of hypothermia or frostbite. The defense used this testimony to argue that John was killed somewhere warmer and that his body was only later moved outside. It’s worth noting first that the state medical examiner disagreed with this finding, testifying that she did see early signs of hypothermia in her examination of John’s organs.

But beyond that, and contrary to the defense’s contention, the lack of these symptoms, even if true, in no way logically requires concluding that John died somewhere warmer. All it would mean is that he died before these symptoms developed. And the defense’s own expert testified that he could have succumbed to his injuries in as little as half an hour. If true, that could also help explain why there was no significant bruising visible on John’s arm. The prosecution seemed reluctant, at trial, to embrace the possibility that John could have died quickly. But I think that was because they wanted to argue that Karen could have saved him had she called for help, not because a relatively quick death is somehow incompatible with John having died by the flagpole from a collision.

Doesn’t the medical evidence prove that John was beat up?

Absolutely not. It’s amazing how this talking point has managed to persist in the face of overwhelming evidence to the contrary and the defense having long abandoned it. It dates back to the original Turtleboy article in April 2023 in which he used autopsy photos leaked to him by the defense to argue that Brian and Colin Albert beat up John. The claim was based on the fact that John had so-called “raccoon eyes” in the autopsy photos. But as every single medical expert in this case – including the defense’s own medical experts – have made clear, John’s black eyes were the result of the skull fracture that started at the back of his head and radiated through (a so-called “coup-contrecoup” injury), which resulted in blood leaking into his ocular cavities.

That injury, which ultimately caused John’s death, was the result of falling backward onto a hard surface. Karen’s defense team did not challenge this; their argument was that the hard surface in question could have been in the garage or basement, as opposed to the frozen ground outside.

Indeed, all of the medical experts – prosecution and defense – agreed that there were no obvious signs of an altercation. John did have a small laceration above his right eye that the defense argued could have been the result of a punch. But even the defense’s hired gun medical expert readily conceded that the laceration could have been caused by just about anything. The prosecution, for its part, noted that the laceration was almost exactly at the height of the small spoiler that sticks out of the rear of Karen’s SUV.

And there were no other signs that John was in a fight. No defensive wounds, no broken or bruised knuckles, nothing under his fingernails. Indeed, the only other notable wound was a small laceration on John’s nose, which Karen herself (in the I.D. documentary) claimed was caused by a shard of glass, which she personally plucked from his nose when she found his body.

Needless to say, that fact seems far more consistent with the prosecution’s theory of the case than the defense’s. Indeed, it’s not remotely clear at this point what the defense’s theory of the case even is. But whatever they’re suggesting, it’s not that John was “beat up.”

But wasn’t the investigation just too sloppy for us ever to know the truth?

This is one of the most common sentiments I hear from people who have only casually followed this case and is a testament to the success of the defense team’s efforts to influence the media narrative. When you press people and ask what was so bad about the investigation, there are typically two things that get mentioned: the solo cups and the failure to search the house.

Yes, it is true that the Canton PD used red solo cups in an effort to collect drops of blood they observed in the snow where John’s body had been found. The defense and its surrogates have long mocked this decision, but here’s the reality. Town police are not jurisdictionally responsible for conducting investigations like this. But on that particular morning, they found themselves in a highly usual situation. The state police, who are responsible for conducting investigations like this, were not in a position to respond due to the weather conditions. And those same weather conditions – an active blizzard – left the Canton Police with a tough choice. They could either try to collect some of the blood through whatever means they had at their immediate disposal, or in a matter of minutes, that evidence would likely be gone forever, lost in the mounting snow. So they did the best they could with what they had. It’s easy to criticize, but it’s not clear what other options they had.

Moreover, there is no reason to believe that this decision compromised the investigation. That blood was clearly John’s, and nothing we’ve learned about this case since then even remotely suggests that the blood collected in those cups was important evidence.

Others have criticized the police for failure to secure the crime scene, but again, it was an active blizzard. Their attempts to put up crime scene tape all failed due to the conditions, and before long the entire crime scene was covered in almost three feet of snow, effectively preserving whatever evidence existed at ground level. As noted above, there is no indication whatsoever that the failure to cordon off the area had any impact on the evidence ultimately collected. The timeline here simply does not allow for any of the key evidence to have been planted.

The criticism that investigators failed to search the house is also misguided. First, Karen’s defenders often falsely suggest that police never went into the house. This is not at all true. Sergeant Lank and Officer Goode went inside the house at around 6:30, shortly after John’s body was found, and spoke to Brian and Nicole Albert, as well as Matt and Jen McCabe. They saw nothing out of the ordinary and obtained statements from all of these people making clear that John had never come inside the house.

And at that point, none of the facts pointed to anything having happened to John inside the house. Karen Read had just told multiple police and paramedics that she “hit him.” Simply put, there was no probable cause to search the house at that point. It’s true that the Canton PD could have asked the homeowners for consent to search it, but as explained above, this was not their investigation. They were merely trying to do what they could while they waited for the state police to arrive and take over. A decision to search the house (or seek a warrant) would have been a major investigative step, and the Canton PD would have been massively overreaching if they’d initiated something like that on their own.

And by the time the state police did arrive and take over, the evidence was already pointing all in one direction: at Karen Read. The state police learned immediately that Karen had told witnesses she hit him, that her taillight was broken, that John was missing a shoe, and that they’d last been seen together in her SUV right at the spot where his body was found the night before. It would have been absurd if they’d chosen to prioritize attempting to search the house at that point instead of interviewing Karen and seizing her vehicle. And by later that same day, they’d secured the SUV with the broken taillight and the SERT team had found the taillight pieces and missing shoe by the curb. Karen, for her part, was interviewed that afternoon and never suggested that John had gone inside the house. And then she ended the interview and lawyered up.

So in very short order, all signs pointed to Karen Read as the culprit. There is no way, with those facts, that the state police could have secured a warrant to search the inside of 34 Fairview, and there was no rationale for attempting to search by consent. Literally not one shred of evidence or any witness statement suggested that John had ever gone inside that house. Even Karen and her lawyers didn’t suggest that John had gone inside the house until over a year later.

If John was outside all night, why didn’t Lucky the plow driver see him?

For understandable reasons, Karen’s defenders attribute enormous (and totally unwarranted) significance to the testimony of plow driver Brian “Lucky” Loughran. Among other things, Lucky claims that he drove his plow by 34 Fairview at a time when, under the prosecution’s theory, John O’Keefe’s body should have been lying in the snow but that he didn’t see anything.

Anyone who is familiar with criminal investigations or trials knows, however, that eyewitness testimony is very unreliable. And that’s especially true in situations like this, where the witness claims not to have seen something. Obviously, if you see something unusual or out of the ordinary, it is much more likely to register as noteworthy to you, and you have a much better chance of remembering the details later on (e.g., the specific time you noticed the unusual thing). But can you remember the details of when and where you didn’t see anything significant yesterday? How about 30 minutes ago? Even if you believe Lucky has great eyesight and a photographic memory (neither of which are true), this is a situation where Lucky was driving his plow in an active blizzard and had zero reason to be paying close attention to a specific area in the far corner of one particular yard. He had no possible way of knowing, at the time, that that particular spot had any significance whatsoever.

Moreover, by the time enough snow had fallen for Lucky to be out plowing it, John’s body – like everything else – would have been covered in snow. So we’re not even talking about failure to notice a body in the dark corner of a random yard; we’re talking about failure to notice a mound of snow amidst a landscape of snow covered objects, in the dark corner of a random yard. And even if, for some strange reason, Lucky was actively scanning the terrain for mounds of snow, all it would have taken for him not to notice this particular mound is for his gaze to have been momentarily focused elsewhere as he passed by that particular spot. Maybe he was looking ahead. Maybe he was scanning around to the other side of the road. Maybe he glanced at his phone or at his dashboard. Maybe he spaced out for a moment because he was tired or bored. Or maybe a mound of snow just wasn’t enough to attract his attention as he passed by it in a moving vehicle, in the dark, in the driving snow. This is the same guy, after all, who admitted to crashing into a basketball hoop with his plow that same night. To contend that his failure to have noticed something, under these conditions, is strong evidence of its non-existence is ludicrous. Tell that to the basketball hoop.

The only other notable part of Lucky’s testimony is his claim to have seen a Ford Edge parked in front of 34 Fairview at some point in the middle of the night. This claim is at least marginally more credible as a car is much easier to see and Lucky would have some reason to take note of a vehicle blocking his ability to plow (which is potentially subject to being towed). I’m more skeptical of his memory of when and where he saw this vehicle, though. While Lucky is, by all accounts, a nice guy who means well, he also strikes me as a highly suggestible witness. And the first time he was questioned about any of these things was by a defense investigator many weeks after the night in question. He was then publicly threatened repeatedly by Turtleboy until he eventually agreed to sign an affidavit.

Long story short, I think Lucky’s claim not to have seen John’s body should be given zero weight and that his claim to have seen a Ford Edge in that area is only marginally more credible. And as to the latter, I don’t see how that claim meaningfully changes the over mix of evidence given the strong evidence that John and his phone were in the same location all night and the complete lack of evidence supporting any conspiracy.

What about the 2:27 google search?

I’m not sure there’s any aspect of this case that makes my blood boil more than dealing with 2:27 dead-enders, who still maintain – despite conclusive evidence to the contrary – that Jen McCabe googled “hos long to die in the cold” at 2:27 a.m. on January 29, 2022. The truth – for reasons I’ll explain in a moment – is that Jen attempted to google that question at 6:24 a.m., after she and Karen had found John’s body in the cold and while paramedics were trying to resuscitate him, which makes complete sense.

There’s an interesting story behind the rise and fall of this particular data point, and it leads to one of the all time “what if” questions in this case: specifically, had Cellebrite (the company that makes the leading forensic data extraction tool for iPhones) not mischaracterized this timestamp in its 2023 software update, would the Free Karen Read movement ever have gained any traction? The supposed 2:27 google search was easily the most intriguing and seemingly exculpatory piece of evidence disclosed in Turtleboy’s original April 2023 article, and just about every person I’ve ever talked to who believes Karen is innocent was originally pushed in that direction by learning about this google search.

So what are the actual facts? In the days after John’s death, Jen McCabe voluntarily provided her phone to the state police for forensic extraction. Using the Cellebrite tool, the MSP forensic team ran an analysis of Jen’s phone, one that found nothing of particular relevance other than her calls and texts with John that evening. Through the standard discovery process, that phone extraction was later produced to Karen’s defense team. Then, sometime in early 2023, Cellebrite released a software update that, for the first time, included a report of the data in an obscure database with the iPhone that had not previously been part of a standard Cellebrite report. That database contained several artifacts related to an attempted google search of the phrase “hos long to die in the cold.” One had a timestamp of 6:24 a.m. (with a slight variation of that query having a timestamp of 6:23 a.m.). But there was a third entry related to this search with a timestamp of 2:27 a.m., and the Cellebrite report characterized this artifact as “deleted.”

Based on that report (which was leaked to Turtleboy by the defense), Karen’s defenders have maintained ever since that Jen googled “hos long to die in the cold” at 2:27 a.m. and then deleted that search to cover her tracks. Obviously, if that were true, it would be a big deal. Indeed, it’s really the only piece of evidence ever offered that, if true, genuinely points toward the existence of a conspiracy.

That said, for anyone paying close attention to the facts, there were some enormous red flags about this supposed search from the beginning. First, there were other searches associated with the same timestamp indicating that Jen was looking at high school basketball standings and information about a club basketball team for her daughter. It would be very odd if Jen, while lying in bed, had suddenly pivoted from reading about her daughter’s basketball teams to researching how long it would take her good friend John to die of hypothermia. This is particularly true given — per the defense’s own version of events — John was already dead and his body still inside the house or garage at this point.

Second, it was clear from the data that Jen did in fact attempt to search “hos long to die in the cold” at 6:24 a.m. Turtleboy and Karen’s defense team tried to explain this fact by suggesting that Jen was “recreating” her incriminating search from the night before (while somehow remembering the exact same misspelling) in order to try to cover up the first search.

But that’s not how anything works. No one who has any familiarity with computers or technology – or who has ever googled anything – would ever think that entering a search a second time, many hours later, would somehow overwrite evidence that the first search occurred. That’s insultingly dumb.

When they learned of this new Cellebrite report and the supposed 2:27 search, the Norfolk DA’s office went directly to Cellebrite to try to understand what it meant. The Cellebrite forensic analyst who ended up looking into the issue, Ian Whiffin, would eventually become a key prosecution witness. After extensive research and testing, Whiffin discovered that the timestamps in this particular database do not represent the time a google search occurred. They represent the time when the Safari tab in which the google search was made was first opened. So in Jen McCabe’s case, she opened a new Safari tab when she was looking at the Hockomock standings the night before (at 2:27) and then subsequently searched “hos long to die in the cold” in that same open tab the next morning. It’s as simple as that.

Lest there be any doubt about this conclusion, during the second trial, Ian Whiffin performed a live demonstration for the jury, using the exact model of Jen’s iPhone and the same IOS version she was using on January 29, 2022. In the demonstration, he showed how the timestamp recorded in the database represents when the Safari tab is opened, not when the google search is subsequently made. Whiffin’s analysis of this issue has been published and peer-reviewed and is not genuinely in dispute within the IT world. Indeed, his conclusions were so clear that his employer, Cellebrite, decided to update its software to remove the timestamps from its reports of this particular database, reasoning that there was too much risk of the timestamps being mischaracterized or misunderstood.

And Whiffin is not the only expert who has looked deeply into this issue. The prosecution called a second expert, a well-renowned IT expert named Jessica Hyde, who had researched this same issue extensively and backed up Whiffin’s conclusion. She further concluded that Jen McCabe never attempted to delete anything related to the search; indeed Hyde pointed out that it is not even possible for a user to delete anything from that particular database.

In response to this testimony, the defense offered…absolutely nothing. The defense essentially conceded this point at trial, offering no expert testimony or other evidence related to the search in their own case-in-chief. The defense had a hired gun expert, Richard Green, who had previously tried to argue that the search did in fact take place at 2:27, but he likely was unwilling to testify to this in the second trial, given the overwhelming evidence against the claim.

To this day, despite all evidence to the contrary, Karen’s online defenders still claim that this search took place at 2:27. In defense of this claim, they generally resort to two (really bad) arguments. First, they claim that the FBI (as part of the federal investigation) validated that the search took place at 2:27. This is not true. Back in 2023, when the bulk of the federal investigation took place, the Cellebrite software update had just been released. Indeed (as I’ll explain below), the existence of the apparent search is likely what kick-started the investigation (initial steps had been taken earlier, but the bulk of the activity happened after). The FBI – like all law enforcement agencies – uses the Cellebrite software, so if they independently analyzed Jen McCabe’s phone data during that time period, they would have gotten the same extraction report as everyone else. Remember, this was before anyone had looked into the issue and really understood what the timestamps in that database meant. But there is no reason whatsoever to believe that anyone at the FBI still believes the search happened at 2:27 or disagrees with the subsequent conclusions reached by Whiffin and Hyde.

The other argument Karen’s defenders make is that other forensic data tools (e.g. Axiom) continue to show that the google search happened at 2:27. This is absolutely not true. It is true that Axiom’s tool (unlike Cellebrite’s) continues to report the timestamps from this particular database. But no one disputes that these timestamps exist. The issue is what the timestamps mean. And Axiom’s user guide now includes a note specifically explaining – in full alignment with Whiffin and Hyde’s conclusions – that the timestamps in that database do not necessarily reflect the time a google search occurred but rather the time the underlying Safari tab was opened.

So long story short, it is beyond reasonable dispute that Jen McCabe did not google “hos long to die in the cold” at 2:27, and the fact that so many people still cling to this belief, despite the overwhelming and conclusive evidence to the contrary, is a testament to how hard it is to dislodge beliefs that have become foundational to people’s belief systems. For whatever reason, many people need to believe that Jen McCabe is the villain of this story, despite how ludicrous that narrative is.

Why didn’t Brian Albert come out of the house in the morning?

I’ve always been a little surprised by the salience of this particular argument to those who believe Karen is innocent. It’s always seemed (to me at least) like a very minor point, but when I ask people why they think Karen is innocent, this is inevitably one of the first two or three points that come up, so I guess it’s worth addressing.

The argument Karen’s defenders make is that the commotion created just after 6:00 a.m., when John’s body was discovered, should have woken up Brian and Nicole Albert. Karen was screaming and there were multiple ambulances, fire trucks, and police cars out front. Brian Albert was also an experienced police detective and first responder. The fact that he didn’t come outside to help or to see what was the matter is, per Karen’s defenders, strong evidence of consciousness of guilt on his part.

Brian and Nicole Albert claim, for their part, that they were sound asleep until around 6:30 when Jen McCabe entered their bedroom and woke them up. By the time they got downstairs a few minutes later, Sergeant Lank was at the base of the stairs waiting to talk to them. By this point, John O’Keefe was already on his way to Good Samaritan, so there was no point going outside to help.

I personally find this version of events to be very plausible. Keep in mind that, by this point in the morning, the blizzard was in full force and the winds were howling, drowning out other noise. All you have to do to understand this is to listen to the audio of the dashcam footage. While you can make out an occasional shriek from Karen over the sound of the wind, you can’t really hear any voices, even when people are very close to the squad car. And while the lights of several of the emergency vehicles were on, there were no sirens. So it’s not clear to me that anything that was happening in the street would have been enough to wake someone up from a deep sleep. And remember that Brian and Nicole had a fair amount to drink the night before and had only been asleep for about four hours by that point. Furthermore, when asked about this at trial, various first responders who were on the scene testified that no one in the entire neighborhood came outside during that time period. So I find it perfectly believable that Brian and Nicole were in fact asleep until Jen woke them up.

Finally, it’s worth pointing out here that this is a perfect example of how conspiracy theories work. Let’s imagine, hypothetically, that Brian Albert had in fact come immediately outside and attempted to help John and/or assist the other first responders on the scene. Can you even imagine what Karen’s defenders would be saying about that? They’d cite his presence at the crime scene as exhibit #1 in their case for conspiracy, alleging that he was tampering with or planting evidence to cover up his crime. So there’s really no way that Brian Albert could have handled this situation that would not have been characterized by Karen’s defenders as evidence of duplicity.

Wasn’t the surveillance footage of the Canton PD sallyport inverted and tampered with?

No. This claim dates back to the first trial, during which the prosecution showed the jury surveillance footage from the Canton PD sallyport without initially explaining that the video was inverted (i.e., it was a mirror-image). The defense claimed this was a deliberate effort by the prosecution to trick the jury into believing that Michael Proctor had been standing next to the left taillight, rather than the right taillight. But that claim doesn’t stand up to scrutiny. The obvious flaw with that logic is that, because the video is inverted, the right taillight appears to be fully intact. Indeed, when the video was first shown in court, Karen herself got super excited and can be seen saying “it’s fully intact!”

Obviously, it’s not in the prosecution’s interest to make the jury believe that the right taillight was still fully intact at that point in time. It would have torpedoed their entire case. So the failure to initially note that the video was inverted was pretty clearly just an oversight.

And as for the video itself, the prosecution made clear that nothing was altered. That particular camera simply records inverted video – which is not uncommon with security cameras. At trial the prosecution showed video taken from the same camera on days prior to January 29, 2022, all of which were also inverted.

It is true that there were delays in retrieving and producing some of the surveillance video from the sallyport and that one of the cameras in the sallyport was not functioning properly (and therefore produced distorted footage), but these issues were heavily litigated by the parties before trial, and there has never been any credible evidence offered by the defense that any of these videos were manipulated or altered.

More importantly, for all of the reasons I explained above, there’s no way that any of these videos are relevant to anything. By the time the SUV was even in the sallyport, the SERT team’s search was already well underway at 34 Fairview, and they found the first pieces of taillight just minutes later. And we have several videos from earlier that day clearly showing that the damage to the taillight had already been sustained. There is simply no reason whatsoever to believe that the taillight was tampered with while the vehicle was in the sallyport. Indeed, even if you’re inclined to believe that Michael Proctor wanted to frame Karen Read, why would he wait until the vehicle was inside Canton PD headquarters, where he had to have known there would be numerous security cameras, to monkey around with the vehicle? Why not do that in Dighton or while the vehicle was being towed back to Canton? None of this makes any sense.

The glass on the bumper doesn’t match the cocktail glass. Doesn’t that mean Proctor planted it?

No. This argument rose to salience in the second trial due to the confusing and technical nature of the testimony from the lab technician who tried to match the various glass fragments found at the scene to the broken cocktail glass found near John’s body. Matching glass fragments in this situation is the equivalent of putting together a jigsaw puzzle when you don’t have all of the pieces. The technician was able, through painstaking work, to match various fragments of glass found at the scene to the broken cocktail glass. She was also able to match glass fragments found on the bumper of Karen’s SUV to a piece of glass found on the street. But she was not able to match those pieces to the cocktail glass.

The defense team argued that this proves that the glass on the bumper did not come from the cocktail glass. And because the matching piece of glass found on the road was found by Michael Proctor, they claimed that the glass on the bumper had to have been planted there by Michael Proctor.

Those inferences are totally unwarranted. The fact that the technician was not able to match the glass found on the bumper to the cocktail glass does not prove that the glass did not originally come from the cocktail glass. It just means there were too many missing intermediary pieces to make the necessary jigsaw-like connection. And it should surprise no one that glass fragments would be missing, given the weather conditions. Pieces that fell onto the roadway that night were at high risk of being further fragmented or carried away by passing plows or other vehicles.

Moreover, the piece of glass recovered from the road that did match the bumper glass was not found by Proctor until well after the pieces from the bumper had been collected in the sallyport. It would make no sense for Proctor to plant glass on the bumper from an unrelated source (and then plant that unrelated glass on the street later) when he had the actual cocktail glass in his possession from day one.

But didn’t the Feds investigate and conclude that Karen was innocent?

No. It is true that a highly unusual federal investigation was opened in late 2022. The investigation seems to have been triggered by a meeting between the leadership of the Boston U.S. Attorney’s Office and Karen’s defense team. To this day, I find it really hard to understand what the USAO could possibly have been told or shown at that meeting, especially back in 2022, that would justify taking the highly unusual step of launching a federal probe into a still pending state case, but for whatever reason, Read’s team found a receptive audience. I suspect that the federal investigation didn’t really begin in earnest until the 2023 Cellebrite software update appeared to show Jen McCabe googling “hos long to die in the cold” at 2:27 a.m.

At that point, the FBI began interviewing witnesses, and federal prosecutors eventually called most of those witnesses before a federal grand jury. By the time of the first Read trial in 2024, it seemed pretty clear that the federal investigation had not uncovered any significant new evidence. In response to a Touhy request, the federal prosecutors dumped over 3000 pages of impounded federal materials on the prosecution and defense just two months before the first trial. These materials, which are still impounded, appear to have largely consisted of grand jury transcripts in which the key witnesses stuck to their original stories. But these materials also included a very limited report from a crash reconstruction firm (ARCCA) and a bunch of personal texts from Michael Proctor, both of which had a major impact on the trial.

Prior to the second trial, in February 2025, the federal prosecutors informed the parties (as well as the Canton PD and MSP) that the investigation had been closed and that no charges would be filed. While Karen’s defense team (as well as Turtleboy) now admit that they knew in February that the investigation was over, they continued to blatantly lie about it. To this day, many of Karen’s online defenders are convinced that the federal investigation is still active.

It’s not, and the fact that the Feds closed it (and announced they were doing so) prior to the second trial strongly suggests that they did not find any credible evidence of a conspiracy or law enforcement cover up in this case.

Wouldn’t Karen have hit Brian Higgins’ jeep if she really reversed at 24 mph?

This is one of those deep in the minutiae arguments where it’s not even clear if Karen’s supporters are claiming that Karen didn’t reverse her vehicle or whether they’re claiming that Higgins’ jeep was never there. I’m not sure even they know.

To back up for second, there was testimony from both Brian Higgins (in the first trial) as well Jen and Matt McCabe (the latter in the first trial as well), that Brian Higgins had parked his white Jeep (equipped with a plow blade) along the street in front of 34 Fairview, by the mailbox. Other witnesses, specifically the three people in Ricky D’Antuono’s truck, did not remember seeing the Jeep. This inconsistency has led Karen’s supporters to claim that the Jeep was never there and that Brian, Jen, and Matt were all lying about this (though to what end has never been clear).

Layered onto this, Karen’s online supporters have misinterpreted the techstream to suggest that, had Karen really reversed her vehicle at 24 mph for 80+ feet, as the data says she did, she would necessarily have crashed into Higgin’s Jeep (again it’s not clear whether they are arguing that the data is wrong or that Higgin’s Jeep wasn’t there). Either way, they are wrong.

The issue here stems from misinterpreting what the techstream data says. Because that data only captures 10 seconds worth of data, it doesn’t tell the full story of the movement of Karen’s SUV at the time. Critically, we don’t know how far the SUV moved forward before it began to reverse. When the data starts, the vehicle is already moving forward at 13 mph and is in fact in the process of decelerating. That means it had likely traveled a ways forward before the recording even started. It then moves 30+ additional feet forward before reversing for over 80+. Given the speed of the vehicle at the onset, it’s very likely it moved forward at least as far as it ultimately moved backward, meaning it very well could have (and probably did) end up very close to the place it started from. That would be consistent with the vehicle colliding with John in the area near the flagpole (likely during the braking process) and would not require the SUV to have ever backed up anywhere near the mailbox.

Moreover, even if the SUV ultimately did back up all the way to the mailbox area, we have no idea where (within the street) it would have been. It’s a wide street, and there’s no reasons whatsoever to assume that the vehicle would have hugged the curb the whole way.

Finally, you can’t have your cake and eat it too. If your position is that the Jeep was never there in the first place, this is an exceptionally dumb argument.

But didn’t Brian Higgins and Brian Albert destroy their phones in violation of court orders?

No, they didn’t. The facts surrounding this aspect of the case are routinely misstated and exaggerated by Karen’s supporters.

It is true that both men upgraded and replaced their phones in the fall of 2022, over eight months after John died. I don’t know about you, but if my phone contained evidence that I had conspired to murder someone, I probably wouldn’t have waited the better part a year to do something about that. Remember, these are both veteran law enforcement officers. So the implication that these phones contained inculpatory evidence related to John’s murder is bizarre.

Now let’s dig a little more into the circumstances, starting with Brian Higgins. In connection with this case, Higgins received notice on September 23, 2022 that the defense was seeking the contents of his phone and that he needed to preserve it (sidenote: Higgins had already voluntarily turned over an extraction of all of his texts with both John and Karen in the early days of the investigation). In full compliance with that notice, Higgins changed his carrier a few days later and got a new phone, while preserving the existing one. This is a perfectly reasonable thing to do when you receive a preservation notice of this kind. It allows you to preserve everything subject to the order while not worrying that your current, actively used phone is going to have to be collected as evidence at some point.

On October 5, 2022, the defense’s motion to gain access to the phones was denied. Shortly thereafter, Higgins specifically asked the DA’s Office if he was allowed to get rid of his old phone. They gave him permission to do so, and only then did he get rid of that phone. So he didn’t violate any orders.

The defense has also tried to make a big deal out of the way Higgins got rid of his phone, stating over and over again that he drove to military base to dispose of it. While it is true that Higgins testified he got rid of his phone at a military base, that’s because he gets rid of all of his garbage there. He has a place on the Cape that doesn’t have garbage service. As a veteran, he has access to the local military base and uses that access to dispose of all of his garbage on site.

As for Brian Albert, he also did not violate any order, though the timing of his upgrade (just a day or two before he received the notice of preservation) does suggest he may have been aware that such an order might be coming. But a couple of things are worth noting here. First, he did not change his number. He simply upgraded his phone on an existing account. So a lot of the most relevant data (texts messages, etc.) simply transferred to the new phone. Second, that is the exact time of year when new iPhones come out and everyone upgrades, so it could genuinely have been a coincidence. Third, there never was any court order requiring Albert to keep that old phone. The motion was denied a few weeks later.

And finally, as noted above, all of this happened the better part of a year after John O’Keefe’s death. Generally speaking, savvy criminals (remember these are both veterans of law enforcement) don’t wait nine months to get rid of incriminating evidence. Moreover, it is perfectly normal and reasonable not to want to turn over your personal phone to strangers, much less a team of lawyers looking for anything they might be able to use to embarrass you. So if you’re not going to get in trouble for doing so (i.e., there’s no order saying otherwise), it’s totally reasonable to get rid of your phone. In other words, there are all kinds of reasons that non-murderers might not want the contents of their phone combed through. It’s silly to assume that the only reason someone might not want to turn over their phone to an aggressive defense team is because they were involved in a murder conspiracy.

Is there anything I forgot?

As I noted above, if there any major points of controversy that I failed to sufficiently address, please let me know in the comments. I’ll update this post as needed.

Finally, although this post is very long, I have in prior posts, expanded on many of these topics in much more depth. If you want to learn more, I’d encourage you to peruse the archives of this blog.