Judson Welcher - Direct (Part 1)

MR. WELCHER: Yes, sir.

MR. WELCHER: Yes, sir.

MR. WELCHER: You're welcome. May I refer to my resume?

MR. WELCHER: Morning.

MR. WELCHER: Absolutely. My name is Judson Welcher. My first name is J-U-D-S-O-N. Last name is Welcher, W-E-L-C-H-E-R.

MR. WELCHER: I'm an accident reconstructionist and biomechanical engineer.

MR. WELCHER: Sure. You can think of accident reconstruction and things such as this as the physics and engineering that happens in car crashes. So accident reconstruction tends to deal with everything that happens to the car and outside the car — skid marks, vehicle damage, the electronic recording systems in the vehicle, things that have to do with the motion, forces, and how the physics of everything occurred and how that affected the motion of the vehicles.

MR. WELCHER: So simply, you can think of biomechanics as similar but to everything inside the car. So biomechanics is looking at how the effects of physical forces are applied to the human anatomy. When I'm trying to explain to somebody what I do, I say, "Hey, it's like a civil engineer who's designing a bridge — so many trucks and cars on the bridge, are they going to cause the beams in the bridge to fail? Is that going to cause the cables to fail?" So a human body is governed by the same laws of physics just like the bridge. Obviously it's made of a lot of different materials, but in a human body, my beams are bones. My cables are the muscles and ligaments — different material properties, but everything's governed by the laws of physics and principles of engineering.

MR. WELCHER: So I'm looking at how physical forces — outside of the car — affect people when they are inside the car, seat belts, airbags. The first seat belt was patented by an engineer. It's engineers who quantify the force of the crash and what kind of injury damage there is to the body. So it is force that causes damage like a fracture. And so we're looking at how the physical forces are applied to the body and what that means in terms of what happens to the body.

MR. WELCHER: Sure. So when I was getting ready to go to college, my father, who was a physician, said, "Whatever you do, do not go to medical school," because I was getting ready to go to medical school. So this was — 39 years ago, he said it's just not worth it. So I said okay, so then I entered into mechanical engineering. I grew up riding motorcycles, taking things apart, frequently pissing my parents off when I couldn't put it back together, but I like taking things apart and figuring out how things worked. So I went to university and got my undergraduate degree in mechanical engineering. From there, I happened to be at a Thursday meeting of the American Society of Mechanical Engineers.

MR. WELCHER: A group came in from a company that did this type of work and they showed some examples of how you apply engineering concepts that you learn in mechanical engineering to the human body. And so I was like, man, that's a great job. That's the perfect mix of engineering that I have and my desire to learn the medical side of things as well, without becoming a doctor. So I talked to them. I actually applied to the company and was ultimately hired as a biomechanical engineer.

MR. WELCHER: So I actually very quickly after I started the job, I took a human anatomy class at a community college. I quickly realized that wasn't enough. And so in around 1993, I went to the USC medical school and enrolled through what's called concurrent enrollment, which means you can take the class but you're not actually in medical school. It's a medical school class, but I took gross human anatomy. So it was me and 160 or so other doctors. I took the same human anatomy that first-year medical students at USC took. So it began in 1993.

MR. WELCHER: I ultimately started my own company, continued to work, continued to do research, had the wonderful opportunity to have two beautiful children, and continue to this day.

MR. WELCHER: Sure. So in about October of 1994, I started my own company. Me and the guy who was my immediate supervisor said, "Hey, we can do this better. We can do more research, more testing." And so we started our own company.

MR. WELCHER: It was an accident reconstruction and biomechanical engineering company. We also early on did a bunch of testing and design and work for amusement park rides. The same concepts for car crashes applies to roller coasters, basically anything. So we worked and had some contracts on improving ride safety. And I was also on the ASM committee that looked at amusement park ride designs.

MR. WELCHER: Correct. We took it from two to about 25 people and then we ultimately sold it right around January 1st of 2022.

MR. WELCHER: I did.

MR. WELCHER: So again, I was working full-time. Actually, while I was doing my master's degree, I was also teaching classes at Texas A&M through their law enforcement training division. So it was extension classes, but I ultimately completed my master's thesis in December of 2001. That was from USC, University of Southern California, and I received my master's degree in biomedical engineering. I wrote my thesis — it was basically on automotive seat design, but it was titled "Whiplash Associated Disorders: Related Anatomy, Biomechanics of Injury, and the Relationship Between Injury Potential and Automotive Seat Design."

MR. WELCHER: —it's my resume, not my report.

MR. WELCHER: Sold it in 2022.

MR. WELCHER: I did. So again, ever since 1993, I've been slowly pursuing higher education, both obtaining my master's degree as well as my PhD. During that time frame, I also published 20 peer-reviewed scientific publications. I taught for five years at Texas A&M and served as a peer reviewer for both the Society of Automotive Engineers and for a short period on the Archives of Physical Medicine. So I was working and running a company with my co-creator, as well as teaching, conducting research, publishing also while going to school part-time.

MR. WELCHER: I was.

MR. WELCHER: So in August of 2011, I finally obtained my PhD in biomedical engineering. I wrote my dissertation — reading from my resume — titled "Development, Validation and Testing of a New Sensor Array for Intraarticular Pressure Measurement in Vitro Human Lumbar Spine Intraarticular Facet Testing," which is a lot of big words to say I was testing cadaver spines looking at the loads inside the spines associated with various motions of the spine.

MR. WELCHER: It's actually in biomedical engineering. Biomechanics is within the field of biomedical engineering.

MR. WELCHER: Correct.

MR. WELCHER: I mean, I have a number of peer-reviewed scientific publications. Peer review just means you submit it to a journal and they have other people review it, give you comments — your peers in the field. I do believe there are a number of them that could certainly be relevant to both the physics and, for example, pedestrian injuries.

MR. WELCHER: Sideswipe collision dynamics. So I was actually an invited speaker at the World Reconstruction Exposition — sorry, not the world congress, the World Reconstruction Exposition — they asked me to give a presentation on the theory, science, and mathematics behind sideswipe or side impact collisions. That would include side collisions, glancing collisions, shallow angle impacts such as like a corner of a bumper goes to the side of a car, any kind of glancing or clipping type impact.

MR. WELCHER: It is.

MR. WELCHER: So that was a Society of Automotive Engineers paper. Again, another peer-reviewed paper that we worked on. It was kind of an extension or continuation of all that research where we just looked at the physics and the mathematics — basically the reconstruction of glancing or sideswipe type impacts.

MR. WELCHER: Yeah, I thought that was the same one.

MR. WELCHER: So in the course of my career, I've done probably 1,200 to maybe 1,300 actual crash tests. I've crashed everything from skateboards, mopeds, up to trains, motor homes, and ambulances. I've conducted — relative to this case, probably — or been involved with 50 to maybe 70 pedestrian impact tests. So the low-speed side impact one is we actually put human volunteers in side impact collisions. And what we did is we had accelerometers, which are things that measure acceleration. If you had physics — or remember, Newton told us force equals mass times acceleration. So knowing the acceleration of something, you can get an idea of how much force is applied to it. So we have various sensors on the body, we have various sensors on the car for the reconstruction aspects of it.

MR. WELCHER: So that study was the biomechanics portion, and the first paper you read about the reconstruction was the reconstruction portion. So the first paper you discussed was everything outside the car. Then we looked at how the forces applied to those impacts to people inside the car.

MR. WELCHER: I don't know if I'd say a lot, but I was an invited lecturer to the Accident Reconstruction and Collision Safety Institute conference where I specifically gave a paper and a lecture on what's called pedestrian impact biomechanics, which was exactly as the name implies — what happens to pedestrians when they're impacted, predominantly by cars and vehicles.

MR. WELCHER: Probably maybe 10%, 5 to 10%. Most of it's car to cars.

MR. WELCHER: That is a paper I wrote.

MR. WELCHER: Yes.

MR. WELCHER: So, Toyotas — newer Toyotas particularly — have two types of electronic equipment on board that can record crashes. There is what's referred to as an event data recorder, or EDR. It's somewhat commonly referred to as a black box — it's not exactly right, but it's close. And then there's a secondary system which is really more closely tied to maintenance, that's called Techstream, but it will pick up data that you can frequently relate back to various events. So that particular paper was looking at the event data recorder data in Toyotas.

MR. WELCHER: Yes.

MR. WELCHER: Correct. For example, every year we're part of a conference called SATAI, which is — our company, largely led by an engineer Christopher Ferbish that we work with, has spearheaded us going out and conducting crash tests for that conference for 10 years. We did a lot of the instrumentation and crash testing at the Accident Reconstruction and Collision Safety Institute conference, and then for a number of years the company actually went down and did crash testing for what's referred to as a spine research institute down in San Diego. So we've been hired by various professional organizations. We came out here once — the Massachusetts Association of Accident Reconstructionists — I think I'm messing the name up — had a conference.

MR. WELCHER: We came out here and conducted crash tests for training other experts out here. We did all the instrumentation, the biomechanics of that crash testing.

MR. WELCHER: Yes. So again, frequently at the conference we're doing pedestrian tests. The last actual Accident Reconstruction and Collision Safety Institute conference, there was a series of pedestrian crash tests run there. SATAI, two years ago, maybe three years ago, some pedestrian crash tests that were done there as well. So there are a number of instances on my resume where I have both attended class, and then I mentioned the one time where I gave an hour, hour-and-a-half long lecture on pedestrian injuries.

MR. WELCHER: Correct. I've always continued my education and research, up until — I just got back from the event data — or accident reconstruction and Collision Safety Institute EDR conference, or the event data recorder summit. My company, through my relationship with the founder of that summit, was a sponsor. So Aperture was actually a sponsor of that conference. We actually sponsored it the previous year as well. And also as part of that, I assisted one of the researchers in conducting — it's called front camera module testing — for pedestrian impacts on Chrysler vehicles — excuse me — General Motors vehicles.

MR. WELCHER: I think ACTAR is important. So ACTAR is an acronym — it's A-C-T — and that's short for the Accreditation Commission for Traffic Accident Reconstructionists. And so it's a bunch of professional groups — originally from some concerns, I believe, from the Department of Transportation — got together and set standards for reconstructionists. Again, these are kind of minimum standards you have to show. You send in your resume, you talk to them about your research, and then if you're deemed qualified, you sit for an examination that tests your skills — a day-long test, theory, a practical portion. And then every 5 years you have to have 60 continuing education units to become reaccredited. I've been an accredited accident reconstructionist since February of 1995.

MR. WELCHER: And my accreditation is current and active up through today.

MR. WELCHER: Yeah. Yeah, correct. I don't want to trick you on the math.

MR. WELCHER: No.

MR. WELCHER: I have.

MR. WELCHER: I have.

MR. WELCHER: I have, matter of fact, in Boston.

MR. WELCHER: Yes, I have.

MR. WELCHER: I have.

MR. WELCHER: Yes. You

MR. WELCHER: I have.

MR. WELCHER: Yes.

MR. WELCHER: Correct. Public defender's office.

MR. WELCHER: No. My job is to be a critical analyst of the data. I want to look at the data as objectively as possible and form conclusions based upon the data, trying to remove wherever possible speculation. Try to remove wherever possible reference to testimony. Now sometimes testimony is the only thing we have, and so we need to rely on it for bits and pieces. But generally my job is to try and look at data and see what the data shows. I have no dog in this fight. I don't, quite frankly, care. What I'm doing is trying to do a good engineering job, which is looking at the data and drawing sound principles and sound conclusions from the data.

MR. WELCHER: It's a company called Aperture. I guess we were acquired or merged into it, technically.

MR. WELCHER: Oh god, I should know that. We have them in Massachusetts. We have them in — I think there's three or four in Texas. I think four or five in California. So I want to say there's at least — I don't know — 13 to 15.

MR. WELCHER: I do.

MR. WELCHER: Long Beach, California.

MR. WELCHER: I do.

MR. WELCHER: Five.

MR. WELCHER: One individual is a PhD in engineering, Dr. Umali, and then I have three other engineers and one case reader. I guess it's six people. I have a case manager as well.

MR. WELCHER: 99% of the time it's my own team. Occasionally we'll get a case from another office. I've worked on a couple of cases out of the Texas office. And maybe a couple out of the Nevada office.

MR. WELCHER: Yes.

MR. WELCHER: I definitely cast a very broad range. We try and get as much information as possible, and then weed it down to what is important, weed it down to what is consistent with the data. So start off very big and then get down to sort of the meat and potatoes of it, the important information.

MR. WELCHER: No. I certainly consider everything I look at, but I frequently don't rely on everything, because sometimes it's not reliable or it's irrelevant.

MR. WELCHER: Sure. Your honor, may I read from my report?

MR. WELCHER: Thank you. So, Mr. Porter?, I apologize — it's a lengthy list.

MR. WELCHER: So I have police reports, including the Canton PD incident report, and I have 58 other reports, also the 34 Fairview Road CARS final report. I have the Ring video from Mr. O'Keefe's residence — I believe that's 1 Meadows?. I have the cruiser cameras, which are the police vehicles' onboard cameras, both front and rear cameras — from one, two, three, four, five, six — I believe 10 different videos. I have photos and videos of the scene from Sergeant [unintelligible]. I have photos of the vehicles — 140 photographs of the vehicles lined up together.

MR. WELCHER: I have 25 additional photographs of the scene, 18 separate photographs of the scene, two photographs of the scene, one group of 92 of the scene, one group of 50 of the scene, another video showing the cars lined up bumper to bumper, four videos of the scene. I have a video showing Miss Read's vehicle being pulled into the Sallyport at the Canton Police Department. I have a 360 diagram. Surveillance video files from McCarthy's, surveillance video files from Waterfall Bar, Canton Police Sallyport videos from Dighton. I have the Lexus Connect Verizon records. I have the alarm.com videos. I have a video of the tow truck arriving at the Canton Police Department. One titled native location John O'Keefe's phone. Past witness statements. The report of examination from the coroner.

MR. WELCHER: Jennifer McCabe Cellebrite data. Something called Lexus infotainment. Different past summary testimony of numerous witnesses. Earlier summary testimony of numerous witnesses. The Boston Magazine interview with Miss Read, a [unintelligible] interview with Miss Read, a 2020 interview with Miss Read. Lab reports from Bode, the Crime Scene Response Unit, Crime Scene Response Unit Boston, Crime Scene Service Section Maynard, Crime Scene Service Section Southeast, the Criminalist file folder, trace file folder. UC Davis veterinary report. I have Verizon records, GPS location records, John O'Keefe GPS points, Magnet AXIOM cache location, weather reports, medical records including the Canton Fire Department run report for both Mr. O'Keefe and Miss Read. Medical records for both Mr.

MR. WELCHER: O'Keefe and Miss Read from Good Samaritan. I have — excuse me, one retrograde extrapolation report, and a serum-plasma conversion report relative to Miss Read. I have the neuropathology report. I have the Orange County Medical Examiner's report, the diagrams, the postmortem toxicology report, and a second — same autopsy report, more pages.

MR. WELCHER: I did. So, first what happens is I asked for all the information on the case. I received some information and then what we do is we start doing our homework, the groundwork. Once I received some of the information, what I did was — you know, modern cars have what's called a VIN number, or vehicle identification number. It's a unique 17-digit number. So what we'll do is we'll decode that VIN number, and that gives us information about the vehicles. So I did that for Miss Read's 2021 Lexus as well as Mr. O'Keefe's 2019 Chevy Traverse. Ran — called, ran the VIN numbers, used the VIN numbers. It tells us gear, make, model, safety equipment, things like that. I then got dimensional data from a program called Expert Auto Stats.

MR. WELCHER: And then I have complete part diagrams for the vehicles — literally bumper-to-bumper parts diagrams from something called the Mitchell collision estimating guide. And what that shows is parts involved in the vehicle. So for example, in Miss Read's vehicle, even though I have bumper-to-bumper, I certainly have all the parts around the rear bumper and tail light of her vehicle. And then of course I started additional work on top of that.

MR. WELCHER: Yes, I did.

MR. WELCHER: That included looking at Miss Read's vehicle, looking at Mr. O'Keefe's vehicle, going to the scene of the incident. I ultimately purchased a 2021 LX570 — the exact year, make, and model of Miss Read's vehicle — and conducted testing with that vehicle, as well as doing a whole bunch of research and compiling data for my report and slides.

MR. WELCHER: I did. I provided a PowerPoint presentation on January 30th.

MR. WELCHER: Okay.

MR. WELCHER: It appears — 141 pages. Didn't look at every page, but it looks right.

MR. WELCHER: It is.

MR. WELCHER: Okay. So that's me. That's my first slide.

MR. WELCHER: Please.

MR. WELCHER: Oh yes, sir.

MR. WELCHER: You want to have as much information as possible. I wanted to physically look at all the things so I can actually see with my own hands, my own eyes, lay hands upon it, look at it as best I can. You try and inspect and look at as many things as you possibly can. Sometimes they don't exist anymore. Sometimes they're gone and you have to make a best approximation if you can't do that. In this case, I was able to look at the scene and both vehicles.

MR. WELCHER: I think so. Um, again, it's always good if you're going to talk about what happened to the cars or what happened to the people to look at what happened to the cars — meaning to actually look at the cars, not just look at photographs when possible. So, I have physically looked at the cars. For example, on the Lexus, I climbed in it, over it, underneath it. I downloaded data from it. I did a whole bunch of things looking at Miss Read's Lexus to assist in my analysis to try and add some further objective data to what we're doing here. So, same thing with Mr. O'Keefe's vehicle — looked at the rear bumper, took some pieces off the rear bumper to see what was behind the cover to see what kind of damage we did or didn't have.

MR. WELCHER: So, went to the scene, looked at the measurements of the curb, looked at where some of the lines are on the road because you can see them in some of the photographs. Um, again, it's always good when possible to inspect everything you can inspect.

MR. WELCHER: Sure. So on September 23rd, I had an opportunity to come out. ADA Brennan, that was the first time I ever met you on that day, I believe, and I had received material before that. And what you see here is this is my inspection of Miss Read's vehicle. In the upper left photograph, uh you'll see — oh, this doesn't work very well. Um, upper left photograph, that's Miss Read's vehicle. That's Mr. O'Keefe's vehicle, the Chevy Traverse. And then what you see here is a laser scanner. You can see it on the ground here. The laser scanner is creating a 3D model of the vehicles. So, I use that to create a 3D model of the vehicles. You can see me in the upper right photograph. That watch is this watch right here, my Garmin watch.

MR. WELCHER: And what I'm doing is looking at the construction composition of that right rear corner of the bumper. I'm looking at things such as the dimensions, where various lights and components are, and how does the geometry of that bumper relate to things that would be hit when backed into.

MR. WELCHER: I did. 9/23/24. Correct.

MR. WELCHER: No. It was obviously before this — I had material before I came out to look at the vehicles.

MR. WELCHER: I don't think it was months. No, it was maybe a month if not less.

MR. WELCHER: It was. Yes.

MR. WELCHER: You're welcome. However, some of the evidence stickers — these white stickers here — were still on the vehicle.

MR. WELCHER: I did not.

MR. WELCHER: Correct. They match what I had seen in other photos.

MR. WELCHER: So, just continuing on. Uh, so what you see is again I mentioned looking underneath the bumper. So, we wanted to see what if anything was damaged on the corner. Um, so I'm putting my hand to show where the frame rail ends. This is me looking up underneath to the back side of this tail light. If you can see, this red tail light actually sticks out. And so I'm checking the front here for damage. I'm checking the back for damage.

MR. WELCHER: Okay.

MR. WELCHER: No, there was no damage to the tail light.

MR. WELCHER: There was not.

MR. WELCHER: You're welcome. So, again, looked at the height of it. I mentioned this lower tail light. Notice how the structure of the bumper protrudes out. You can see one of the marks, which is consistent with the contact to Mr. O'Keefe's vehicle here. This scratch mark, scuff mark. Notice how close it is to this light. Again, some of the evidence stickers — this is a better view of those. Then going down the side of the vehicle. So with all that information, I mentioned a laser scan. So you move this laser scanner and each minute-51-second scan is collecting about 50 million data points. And so you scan all the way around and then what you do is you merge all that data together. So even though this looks like a poor photograph, what you're seeing is hundreds of millions of points.

MR. WELCHER: And so you literally can figure out, you know, if you wanted to know the distance from this lug nut to where the lighter is on the ashtray, you could probably do that to an accuracy of about 2 to 3 millimeters. So really, you know, taking tons and tons of data. So it knows the exact position of every one of these millions of data points and knows the color associated with those data points.

MR. WELCHER: Uh, you want to have an accurate model of the car. So, I mean, you probably don't need literally that many points to get an accurate model, but that's just the nature of the way the science is going. Computers are getting better and faster, scanners are getting faster. So you get an enormous amount of data when you use these scanners nowadays. So then as indicated on the underline here, I did a CDR. That's a little different than what you heard me say before. CDR is crash data retrieval. So that's the hardware and software that you use to download — or actually copy or take an image of — what's on the vehicle: the EDR, which is the event data recorder. So you use the crash data retrieval hardware to download the event data recorder, otherwise commonly known as a black box.

MR. WELCHER: So I downloaded the black box from this vehicle, and although you can hardly read it here, I mentioned there are two systems on this vehicle. This is the CDR system or EDR system. There's also the Techstream. The EDR system is primarily designed to manage the safety components, protect the occupants of the Lexus. So things like fire the airbag, fire what's called a pre-tensioner, keep track of the severity of the crash to decide whether we need to fire the airbag. So it's not really designed for pedestrian impacts. So when we downloaded this vehicle, we got what's called a null file. A null file simply means that there's no data there. The system's working properly, we had no fault codes.

MR. WELCHER: But for a car hitting a pedestrian to cause a file here, this is really more designed for car-to-car impacts — something that's going to have a significant impact on the occupant of the Lexus. And again, the Lexus is a 6,000-pound vehicle. Mr. O'Keefe is only 216 pounds, so it's not going to register from hitting a pedestrian.

MR. WELCHER: No, exactly what I expected.

MR. WELCHER: I mentioned the second system. So, as underlined up here, this is the Techstream. So, this is from Toyota Techstream data. You can see the screens here. There are generally two types of files: what's called a vehicle control history and a PCS file. Uh, the PCS file is the — um, God, I forgot the name of it. The protection system. So I also looked at Mr. O'Keefe's vehicle. Um, sorry — PCS is the pre-collision system. Looked at Mr. O'Keefe's vehicle. Same thing. Took a series of photographs, laser scanned it, looked in it, underneath it, behind it. I mentioned taking some of the bumper parts off. So, I took this rear cover off here to look at the damage behind the bumper cover. Um, most cars have what's referred to as a face bar. That's that piece of metal that you see here.

MR. WELCHER: And there's a piece of energy absorber here. That's your face bar. So, I'm looking to see if there's any damage to that. There was no damage to that.

MR. WELCHER: Sure.

MR. WELCHER: Sure. Um, I mean, a lot of data is stored in a similar location. The Techstream system was originally designed according to Toyota for maintenance issues. It seemed to come about around the same time Toyota was having problems with sudden accelerations, or allegations of sudden accelerations. So the Techstream data or vehicle control history will give you information when there's, for example, excessive acceleration or rapid shifting of the vehicles or rapid steering of the steering wheel. So the statements are that that's designed to help the maintenance people if you come in with a problem with your vehicle to diagnose what that problem is.

MR. WELCHER: It has a lot of utility in accident reconstruction because if you jam on the brakes before a crash, frequently the Techstream data will record the crash.

MR. WELCHER: Sorry. In my mind, they're the same thing. Toyota, Lexus — obviously a little different — but Lexus is basically made by Toyota. So Toyota and Lexus — Lexus has the Lexus safety system. Toyota has the Toyota safety systems, which you both read with the Techstream data.

MR. WELCHER: I think so. Yes. The Tech-

MR. WELCHER: So, I independently downloaded it and then I also ran a test with Miss Read's vehicle and I created a new Techstream file so that I had a stamp in the ignition cycles of what the current ignition cycle was. So, I did a sudden acceleration test of the vehicle — on the Lexus — or on the exemplar vehicle — on her Lexus.

MR. WELCHER: Uh, again, you know, you — whenever possible — want to be the one that does it, make sure it's done correctly and be able to say, "Hey, how do I know that's from that vehicle?" Well, I know because I was the one who did it. And so, if I get it from someone else, then they frequently have to bring that person into court to lay — you know — say the other ones that did it. So then I can use it. So, if I do it myself, it eliminates that problem and then I have no concerns over where it's coming from. So, then I know — hey, I did it. This is the right car. This is the right program.

MR. WELCHER: Correct. So again, simply a blue arrow indicating that I looked down behind all this stuff, took measurements of it, and then similarly I have a 3D point cloud of Mr. O'Keefe's vehicle. So, part of the reason we're making these 3D point clouds — and we'll touch on this further, in addition to the scene — you can then basically put all these things back together in the computer. So, instead of having to drive the cars to the scene, for example, or to Mr. O'Keefe's residence, you can take these computer models and additional data and put it back in a computer and reconstruct everything on the computer. So before we had the exact angle of impact. Again, this is for looking at when Miss Read contacted Mr. O'Keefe's vehicle backing out of Mr. O'Keefe's garage at One Meadows.

MR. WELCHER: I lined them up just to generally show the approximate location. A couple things. You notice how again this lower portion of the Lexus bumper sticks out. Notice how the side of the Chevrolet slopes in as you get higher up. So this portion sticks out more than this portion. And this is where you would expect any contact damage to occur, assuming there was an impact. So again, looking at the measurements up the height of the vehicle — of course, as you indicated, Mr. Brennan, the tail light's gone — and then went to the scene. So, same thing, you know, went to the scene, took again 297 photographs of the scene, had our people scan it to create a computer model of the scene.

MR. WELCHER: So, you know, literally, if you wanted to know the distance from this window to, say, some manhole cover on the road, we have the data to figure that out, you know, down to a couple millimeters.

MR. WELCHER: Yes.

MR. WELCHER: My fingers are too fat. Thank you. Okay. Um, so what you're looking at is what I call event 1162. If you see anything in blue on these charts, that's information that I added. I typically put my initials — JBW — Judson Blair Welcher — that's me. And so this is referred to as the key cycle trip. So key cycle is simply you turn it on — not necessarily turn the ignition on — turn it on, turn it off. That's one key cycle. So I labeled this one 1162-2 because there's also one that's on the same key cycle that I'm calling 1162-1. The system also records the mileage. So again, this is a maintenance system generally. So what you see is the T here is the trigger and there's a point. So this time in the Techstream data, you could think of it as like a stopwatch.

MR. WELCHER: It's when it starts — when the car turns on — and stops when the car turns off. And then if it sees a triggering event — a T — it records that information. So this is when the trigger happened. So that's 1,142.2 seconds. And so I've converted that to minutes and seconds because most people don't think in seconds. And so what it also does is once it sees a trigger, it's constantly circulating data through the computer. So when it says, okay, we have a trigger, it saves 5 seconds before the trigger and then 5 seconds after the trigger. So the trigger is in the middle and you have 5 seconds of data before and 5 seconds of data after.

MR. WELCHER: Uh, it's just a continuous clock, meaning it's not a clock like a time clock. It's more like a stopwatch kind of clock.

MR. WELCHER: The same thing. And you start the stopwatch. It's running.

MR. WELCHER: No, not alone. It simply tells you how long it's been on. It doesn't necessarily immediately tie it back to a clock time. You need additional information. And so this, in and of itself, the Techstream data doesn't give you a certain time on a clock like 1:00 or 2:30, anything like that. No, it simply tells you — in this case, when the trigger occurred — it's been on for 19 minutes and 2 seconds.

MR. WELCHER: Correct.

MR. WELCHER: So, it depends on what the something is. So, if there's another trigger, then it will record more data. So, you can have two triggers that are recording things on the same time window as well — or very close to the same time window. So, you need another trigger to keep it going. But in essence, what it does is when it sees a trigger, it saves that 10-second window and then just keeps monitoring things until it sees another trigger.

MR. WELCHER: No. For example, in the trigger event — say 5 seconds later the vehicle is still moving. Well, the data just ends and it says it's going — I don't know — 17 mph, for example. It doesn't continue on. We just know that when the trigger ended, it was going 17 miles an hour.

MR. WELCHER: Okay. Okay. So, we have mileage recorded in this system. So, what we're trying to do is use the available data to see if we can determine what event this is — or is this the event — or is this something that could have occurred outside 34 Fairview on the evening of this event. And so, one of the things we're utilizing is the mileage. Okay. So 12,629 is at the event — end of event number two on the 162 ignition cycle — when the vehicle ultimately comes into the impound yard. And this is actually before that comes into the impound yard. It has 12,665 miles. That's a difference — did the math for you here — of 36 miles. Okay, let's consider where that could have come from.

MR. WELCHER: So, we looked at the various paths and descriptions of where Miss Read's Lexus was driven after leaving 34 Fairview on the night of this event, up until the point and time the vehicle ended up at the Canton Police Department. So, we used Google Maps. We adjusted for the correct date and time of the year. And then what you'll also see is we adjusted for the correct time of day. So, although you can't really see that, let me see if I can blow it up. So you can see that's looking at the mileage and time. Again, really didn't use the time, just use the mileage, on Saturday, January 29th at around 12:00 a.m. That of course does not take into account the weather.

MR. WELCHER: So this is from 34 Fairview to one Meadows Avenue which serves Mr. O'Keefe's residence and there are other mileage that you have there. Correct. So we basically ran through the entire path as we understood it. Again, we didn't have complete information. We didn't know the exact roads. We didn't know if turned around, stuff like that. So you're seeing the suggested available paths and we end up with a range of data. So we go through — oops — all the paths and you can see that this is what we end up at the bottom here, which is very hard to see. So the minimum suggested route is 36 miles, 36.1 technically. You have a range of 37.9 to 39.7, but again that's right in the range of the 36 miles.

MR. WELCHER: So that's a data point that's starting to build our confidence in what is in the Techstream data being related to our event. This is all an effort to match whether or not the 1162 data is relevant to a particular event.

MR. WELCHER: No. So again these things are all tools. You can't take necessarily one data point in a vacuum. You need to consider the totality of the data, which is what I did. So we mentioned the continuous clock that was in there, or the running clock that was in the vehicle.

MR. WELCHER: I am doing that right now. You can see we have the time and again I've converted in blue to minutes and seconds. This is the first trigger on that key cycle. So on key cycle 1162, there were two separate events. The vehicle keeps a continuously running clock. So we're able to see the time each event occurred relative to when the vehicle was started and the difference between the two events. Okay. So at trigger event 1162-1, we know that the ignition has been on for between 10 minutes and 57.4 seconds — or that's when the trigger is — sorry — 10 minutes 57.4 seconds. Again the trigger has plus or minus, so that's from 10 minutes 52.8 plus 10 seconds, which gets you to 11 minutes 2.8 seconds, or as I wrote here, 10 seconds total.

MR. WELCHER: So again the Techstream data indicates at the time this event was written the trigger occurred — again at 10 minutes. You see we have the T, 10 minutes 57.4 seconds.

MR. WELCHER: So you can tell they're separate events because there's a separate amount of time between them. There's about 8 minutes between the events and so they're on the same ignition cycle, meaning the vehicle hasn't been turned off in between. And we have this continuously running clock that records when these things occurred. So it tells us when event one occurred, when event two occurred. We know it's on the same ignition cycle. So if we have some idea of when the car was turned on, we can then tie that back to clock time or real world time.

MR. WELCHER: This is trigger event 1162.1. I was provided with data and a report from Mr. Burgess that indicated the infotainment system was turned on at 12:12:36. Let me zoom in for you folks. That corresponded to a GPS location that looked like it was outside the Waterfall Bar & Grille on Washington Street. So additional cell phone data that we had is that we knew there was a three-point turn made when the Lexus passed Fairview on Cedarcrest, pulled into a driveway, and made a three-point turn. So what you see over here is GPS data from Mr. O'Keefe's phone. And with that information and knowing when the system turned on, we're able to now tie the Techstream data to a real clock, meaning what time of day did that Techstream trigger occur. So we're going from the running clock to a real world clock.

MR. WELCHER: So if you take when it was turned on from Mr. Burgess's report, you can see that the Lexus was turned on at 12:12:36. You add in the Techstream information, you get it at 12:23:38. And now we have a cell phone data point at 154. So at the time my report was written in these slides back in January, that was all the information we had in terms of data point 154. That was the one that was closest in time to it. So we had since received additional information.

MR. WELCHER: If we can start at the top.

MR. WELCHER: Correct.

MR. WELCHER: That was from Mr. Burgess's report.

MR. WELCHER: That came off the SD card, I believe, in the infotainment system.

MR. WELCHER: Uh, no. I'd have to go back to his report, but it was in — this information was in his original report based on the SD card information about when the car was turned on and off.

MR. WELCHER: So when the car is turned on, that's approximately when the running clock starts in it. There may be as much as a 3-second delay in it, but that's approximately when the clock turns on. So this shows us when the stopwatch was started for the Techstream data. So knowing that that infotainment system has date and time stamps on it, that ties it to a real world clock from GPS information. And so now we can — when that infotainment system turned on, it's very close in time to when the Techstream data started recording data. That allows us to tie the two together.

MR. WELCHER: The first event.

MR. WELCHER: Uh, I call it the three-point turn.

MR. WELCHER: Right. We have a whole series of GPS data with heading angles as well as a description that Miss Read gave in an interview.

MR. WELCHER: Uh, it's my understanding that's a clock in the infotainment system and then we are ultimately trying to also tie it back to the clock on his cell phone.

MR. WELCHER: Correct.

MR. WELCHER: So it's my understanding they do not. In my discussions with Mr. Burgess, reading reports, they can have — there can be as much as 30 seconds to a minute of variation.

MR. WELCHER: I have.

MR. WELCHER: That means differences from — I'm sorry. Differences from what? From typically the correct answer. Like meaning that a given data point has variance in terms of how it's measured, when it's measured. So different measurement devices can have a different variance around the actual true number.

MR. WELCHER: Correct.

MR. WELCHER: Correct.

MR. WELCHER: There is.

MR. WELCHER: I did not. I did not receive that until May 8th.

MR. WELCHER: Yes.

MR. WELCHER: Yes, we'll come.

MR. WELCHER: It's the first one.

MR. WELCHER: Correct.

MR. WELCHER: Uh, Cedarcrest.

MR. WELCHER: Correct.

MR. WELCHER: That's coming from the cell phone of Mr. O'Keefe at data point 154.

MR. WELCHER: Uh, I mean, I do the math between those two.

MR. WELCHER: — it better? So there's an infotainment time of 12:12:36. I guess another time — that answer your question. Then we have the 12:12:36.

MR. WELCHER: That's when the Techstream data basically turned on, plus or minus a little bit.

MR. WELCHER: Correct.

MR. WELCHER: Yes.

MR. WELCHER: Sure is. Okay. So, again, we are looking at event 1162-1. The information in blue is speed. It comes out of the Techstream in kilometers an hour. Converted to miles per hour, feet per second. Looked at the total distance traveled. Shift position. So in the Techstream data, a shift position of one indicates you're in drive. A shift position of two indicates you're in reverse. There's no zeros.

MR. WELCHER: Zero is usually neutral and I believe three is parked. So there's also data on the acceleration of the vehicle. Basically, what is the effect of whatever you're doing in terms of stepping on the gas or not stepping on the gas. And then there's also information on the steering of the vehicle. So, again, there's trigger 5 seconds before, 5 seconds after. So over here a positive number equals a leftward steer. So you can see we have in the forward velocity — or forward motion — in drive: goes forward with leftward steer, stops, shifts into reverse, and then continues with a rightward steer back.

MR. WELCHER: Yes, it can have lots of triggering events. In this case, it has two. On key cycle 1164, there's 11. On 1167, there's 10.

MR. WELCHER: Yes. So if you look at the triggering — the indication of what this trigger was — what you see is accelerator opening ratio, which is basically you can think of as how much you're stepping on the gas pedal. And so if it exceeds — in this case 30% — quick, and time after shifting into reverse, that tells the system, "Hey, we might need to start recording this data." So what you can see: went into reverse and then within a very short amount of time the accelerator pedal went above that 30%. That's what caused the trigger in the Techstream data.

MR. WELCHER: Sure. That's how many revolutions per minute the engine's turning.

MR. WELCHER: Correct. There's going to be a little lag. Your car doesn't move instantaneously. You have to step on the gas, then the RPMs go up, and then it starts going faster.

MR. WELCHER: Sure. So we know the speed at various times. We know from this data that there is — you can see half a second between each one of these timestamps. So one, two, three, four, five, six, seven, eight, nine, 10 data points at half a second. That's the five seconds of pre-crash data. So if you know you're going so many miles per hour — again, I converted it to feet per second — feet per second, you're going so many feet per second for half a second, for example. You just multiply those two together and it tells you how far you went. So what I'm doing is looking at speed traveling during the interval — that's during this short time period. And then I add it all together here for total distance traveled.

MR. WELCHER: It is.

MR. WELCHER: We generally cannot. I mean, you might have some — it looks like it's slowing down and then you see right here this brake switch application. So you have two things relative to braking. You have a brake switch, which is simply whether the brake is on or off — just enough pressure to make the brake light come on. And then you have brake oil pressure, which gives you some idea of how much pressure is in the system, which is a rough indication for the braking. But then what you see here is the longitudinal acceleration or deceleration. This gives you some idea of how much the vehicle is speeding up or slowing down. So generally a minus sign is a slowing or accelerating in reverse.

MR. WELCHER: Correct. So, at the end of this window, you can see it's still going about 8.7 miles an hour when the trigger stopped. So, after that trigger plus 5 seconds, the system said we no longer need to monitor it. Things have returned to an okay state. And so then it stopped monitoring the system.

MR. WELCHER: Yes, I did.

MR. WELCHER: You want timewise or location, or both?

MR. WELCHER: So it happened on Cedarcrest. It happened between cell phone data points 154 and 155. And it's a three-point turn on Cedarcrest. And it occurred, according to the Techstream data, about 12 minutes 23 seconds after the vehicle was turned on.

MR. WELCHER: Yes.

MR. WELCHER: 21 to 29 seconds.

MR. WELCHER: So 12:23:38 is syncing the Techstream time to real-world time. So we know how long the Techstream is on. We know when the vehicle was turned on. So we're adding those two together. So if you add the 12:12:36 to — for example — 11 minutes 2.88 seconds for when the end of the trigger is, that gives you 12:23:38. So the end of the trigger in real clock, real-world time — no longer in the running clock — is now 12:23:38, based on the infotainment clock.

MR. WELCHER: From the Lexus.

MR. WELCHER: The Lexus.

MR. WELCHER: So that is Mr. O'Keefe's cell phone data at point 154. Again, from the original data we had just at 154 and 155 and not the in-between points.

MR. WELCHER: Sure. If you look at where the Techstream math works out to be 12:23:38.8 and compare that to 12:23:58, the difference between those two is 19.2 seconds.

MR. WELCHER: Between what the Techstream infotainment system would show compared to what data point 154 on Mr. O'Keefe's cell phone is.

MR. WELCHER: Yes.

MR. WELCHER: You need to advance the Techstream clock 21 to 29 seconds.

MR. WELCHER: After you advance the Techstream ahead 21 to 29 seconds, can you then compare the information, the time that's coming from the infotainment system, Lexus, to Mr. O'Keefe's cell phone, from his iPhone?

MR. WELCHER: Correct. That's what you need to have an apples-to-apples comparison. You need to get them on the same time clock.

MR. WELCHER: So I think we left off here. So that's 1162-1. So what we have is we have the cell phone data, the original cell phone data on the right hand side. And then from that Techstream data, you know, we have whether it's going forward, we have the speeds, and we have the steering angles. So from that information, we ran a simulation to show you what the Techstream data means. This is what the 10 seconds of data is. And again, just because the simulation stopped, that's just where the Techstream data ends. Again, as we showed you earlier, the car is still moving. That's just the last data point we have. So what that Techstream data shows is matching a three-point turn on Cedarcrest Road.

MR. WELCHER: So we have the vehicle motion, the timing data consistent with the interview statements about making a three-point turn.

MR. WELCHER: Sure. So you can see — you'll have to enlarge that. I'm sorry, Doctor. Column N — that's the shift position. So that's looking — we start off going forward right here. And you can see at the start of the simulation we're going about 9.94 miles an hour. And then as you start a three-point turn, you generally start — in this case if you're going to the left, with the left turn. You're welcome. So you can see we have the steering angle here, then the transmission. So again, we have the brake application. So you're coming in, but you're slowing down, and you get down to a speed of approximately zero. Zero, shift into reverse, back up, and then start to turn the other direction as you're backing up. So again, you want the rear to go to the right, you turn the steering wheel to the right.

MR. WELCHER: Right is negative, and then as you start to pull around and pull forward you turn back to correct for the steering, and so that's what this shows. So again, if you put it all into a simulation program — the speeds, the shifting, the distance — keep in mind we only have that 10-second window of data, and the steering angle for that Lexus. That's what it indicates.

MR. WELCHER: Sure. So again, what we're trying to do is we have the infotainment, we have the Techstream. We're now trying to see how that correlates to Mr. O'Keefe's cell phone. So we had originally very coarse data points, 154 and 155. There's additional data points between here. So we're looking at how well does our predicted time compare to those two. So with the cell phone correction, you end up with the Techstream data at 12:23:59 to 12:24:07. So that's actually right in between these two data points. And so we know 154 was closest, but it's on the right hand side of the road. 155 is when it's turned around and heading back in the southerly direction. So the cell phone data corroborates the Techstream timing data in conjunction with the infotainment system data.

MR. WELCHER: Okay. So we talked about 1162-1. That's the first trigger on the Techstream. So we're now going to go to 1162-2. So I've highlighted that in yellow. So we're now talking about the second trigger. So this is on the same ignition cycle. It's on ignition cycle 1162. We have that same continuous running clock, but we've now synced it to real world clocks. So instead of the running clock, we now have some information on the real world clock. So at this time, we know the ignition has been on for 19 minutes and 2 seconds. So at the trigger, 19 minutes and 2 seconds. And again, we have 5 seconds either side of that. So between 18 minutes 57.6 seconds to 19 minutes 7.5 seconds.

MR. WELCHER: Which part?

MR. WELCHER: Sure. So again, this is continuous from when the time comes on. I just converted it to minutes and seconds here so you can see how long it's been on. This is the mileage. These are the speeds over here. And this is some information on how far it traveled. Again, we have the plus or minus on the steering. That gives us some information on the steering. Really, there's very little steering going on at this point. Slight leftward steer.

MR. WELCHER: Yes.

MR. WELCHER: The three-point turn occurred about 8 minutes after —

MR. WELCHER: This event — I misspoke. 1162-2 occurred about 8 minutes after the three-point turn.

MR. WELCHER: Sure. So we have the accelerometer data here. I've highlighted a shift from drive to neutral to reverse.

MR. WELCHER: Absolutely. We have information on the brake switch, brake oil pressure. We have the acceleration data here. And then from that information, using the acceleration data, calculating the speed at which it's backing up. The reason we have slightly different calculations here is we have some evidence that the wheels are slipping on the ground. So we need to use the accelerometer because the wheel speed is simply telling you the speed of the wheel. It's like if you're stationary burning rubber, the speedometer reads a certain speed, but you're not going anywhere. But the accelerometer tells you how much the vehicle is moving. So from all that information, we get information on how far the vehicle's traveled. We've done it in terms of total travel distance.

MR. WELCHER: We've done it relative to the start position.

MR. WELCHER: All the way to the left. Yep. Okay.

MR. WELCHER: So again, that's the continuous running clock. In the second column in blue, that's where I take that time and convert it to what we more commonly use, minutes and seconds.

MR. WELCHER: So this is now 1162-2. So again, it's technically in the system the same key cycle. I've called it 1162-2.

MR. WELCHER: So it's actually at the same mileage as 1162-1. So we know we see they're within the same mile.

MR. WELCHER: Yes.

MR. WELCHER: So we have accelerometer data. We have velocity data, or speed data. Back here — I corrected for what you see down here — and I go through this on a later slide, but right here you can see a significant increase in the RPMs but yet the acceleration is dropping. So what that means is if your RPMs are going up and the speed's going up, that means you should be accelerating more. But because the acceleration is dropping, that means what's actually happened is the wheels are slipping. So what I've done is I've used the accelerometer data to calculate the speed once the vehicle is put into reverse. And once we know the speed — again, feet per second — we know every data point is half a second. So speed times time tells you how far you moved. Feet per second times seconds gives you feet.

MR. WELCHER: That's how far you went. So you get how far you went in each interval. Then you get the total distance traveled. And so this is relative to the start. So zero is the start. The vehicle goes forward about 34 feet. Then you can see it basically stops in this area here. Then begins to back up and ends up backing a total of 53 feet back from where it originally started. So if you add 53 to 34, you get 87. So the total motion: 34 forward, 53 back. The Lexus moved 87 feet total.

MR. WELCHER: No, that's just from when the data starts to when the data ends. So if you look at when the data starts, you can see that the vehicle is actually going 13.67 miles an hour. When the data stops, the vehicle is still going 23 miles an hour. And you can see that it's still at 74% throttle. So this is three-quarters of full throttle in reverse.

MR. WELCHER: That's where the first data point is. Again, the trigger is here. This is 5 seconds before the trigger.

MR. WELCHER: No.

MR. WELCHER: I mean, considering you're at 74% throttle, it likely increased.

MR. WELCHER: I don't know. And you can see that up here she's on the brake. Down here, she's not on the brake. And so, she's not on the brake and at 74% throttle. And what is the percentage of throttle that has to be initiated to cause a triggering event for a shift into reverse and a rapid throttle application? 30%.

MR. WELCHER: Sure. Again, that's the engine RPMs. And so, as I mentioned, you have this jump in engine RPMs and a very dramatic jump in the reported speed of the vehicle. But when you look at the acceleration, the acceleration is actually dropping. So again, that's a telltale sign that the tires spinning because the vehicle is not being accelerated. And so I've used the accelerometer data. Again, if you use a process called integration, you integrate the accelerometer data, it will give you the velocity. So what I've done is it's called piecewise integration to get the speed data that takes into account the tire slippage and from there calculated the speed or the position at each interval going across the top here.

MR. WELCHER: Yep.

MR. WELCHER: Yes.

MR. WELCHER: Well, again, you can see it's increased about 1.9 — excuse me, 1.27 miles per hour. And again, we know from what I already stated that that last time step, you're still at 74% throttle and you have zero brake application.

MR. WELCHER: Not in total. No. We just have this 10-second window. I can tell you what it did during this 10-second window.

MR. WELCHER: No.

MR. WELCHER: Sure. One is drive, zero is neutral, two is reverse, and then brake switch status. Again, a one is you're on the brake pedal — enough to activate the brake lights. Zero is you're off the brake pedal.

MR. WELCHER: 5 seconds.

MR. WELCHER: So, again, it only occurs at one time step and again there's a little timing issue — like we only have data every half second. You don't know exactly where in that time it was.

MR. WELCHER: Next slide or continue on this slide.

MR. WELCHER: No, I'm good. Next. So, similarly, using this information to then try to tie it back to the cell phone data to make sure we have the real-time information. Again, this was written back in January before I got the additional information. So, what you can see is I'll zoom in again. We're now talking about trigger 1162-2. We know that that's 18 minutes that the vehicle has been on. 18 minutes 57 seconds. Excuse me. 19 minutes 2 seconds. This is the range again. The plus or minus 5 seconds. The 19 minutes 2 seconds is when the trigger occurred. Infotainment system. Adding those two together — 12:12:36 plus the 19:07.5 — indicates the end of 1162-2 is at 12:31:43.

MR. WELCHER: It is. It would need to be corrected to the O'Keefe cell phone data.

MR. WELCHER: 12:32:04. So if you simply take this and add 21 seconds to it and add 29 seconds to it. So you get between 12:32:04 and 12:32:12. When you reconcile the two clocks and you have that range of 12:32:04 to 12:32:14 — 12:32:12, I'm sorry. 12:32:12.

MR. WELCHER: Correct. So, you know, again, even within John O'Keefe's cell phone, even within his cell phone, there's somewhere between 21 to 29 seconds difference on his cell phone. So, even that one piece of instrumentation has some variance to it. So, we know it's between 21 to 29 seconds at this approximate time. So, this is taking into account that variance.

MR. WELCHER: Yes.

MR. WELCHER: So I was given information about a final lock event on the phone.

MR. WELCHER: Okay.

MR. WELCHER: No, the Lexus — or excuse me, the Techstream — will only give you a 10-second window.

MR. WELCHER: Understood.

MR. WELCHER: So I have this other event, a lock event, and the information I was given that at 12 — it's 12:32:09. And so we have the Techstream data ending at somewhere between 12:32:04 and 12:32:12, which puts this number right in the middle of that, more or less.

MR. WELCHER: Thank you, Your Honor. Thank you very much. [Lunch recess — garbled remote audio]

MR. WELCHER: Slide 35.

MR. WELCHER: Correct.

MR. WELCHER: So, we already touched on this from left to right. You know, the first column is the continuous running clock. Second column is where I converted it to minutes and seconds. You can see we're on trigger 1162-2. This is the second trigger. The particular boxes that are highlighted are where there's change. For example, in the upper right when you're transitioning from drive to neutral to reverse. And then I already discussed this a little bit, but where the wheels are slipping, you can see you have high acceleration pedal application, increase in the RPMs, but when you look at the effects on the vehicle, you can see that it's accelerating less. So that's indicative of slippage of the wheels. And then I simply show at the bottom the math I use to derive all that.

MR. WELCHER: So again, we touched on most of this. It's the same slide you saw before just to show how corrected for the speed.

MR. WELCHER: RPM column. Go ahead.

MR. WELCHER: Correct.

MR. WELCHER: I mean, it's 3/4 of full throttle. It's 74% of full throttle. If above it, 74.5 — it drops to 73.5 and then back to 74.

MR. WELCHER: So there's — again, this data is referred to as being asynchronous. You're reporting it at particular times, but the computer scanning the network for the data. So it's not all on exactly the same time step. Not only that, but there frequently is a delay from when you step on the accelerator pedal to when that translates to increased RPM at the engine. It takes a little time to build it up or take it down.

MR. WELCHER: You're welcome.

MR. WELCHER: No, it did not come to rest. I misspoke. That's where the Techstream data ends. As you can see here, it's still going. It's still going approximately 23 miles an hour.

MR. WELCHER: So again, as I mentioned, this — the Techstream is taking a window of data. It doesn't take anything outside the window. You can see that when we account for the wheel slippage that the vehicle at the time of the last data point again is at 74% throttle, 3,300 revolutions per minute, and is going backwards approximately 23.9 miles per hour. And you can see it's traveled a total distance backwards of 53 feet back from where it initially started. Or if you consider where it pulled forward 34 feet and then backed up 53 feet, add those two together from the point it pulled all the way forward. So this is 10, 20, 35 forward, and then I'll go 10, 20, 30, 40, 53 back for a total of 87 feet.

MR. WELCHER: So from the maximum point it pulled forward to where it backed up again — that was 34 feet forward, then a total of 53 back, total travel distance of 87 feet. So went all the way forward where it stopped. It then backed up 87 feet. And when the Techstream data stopped, it's still moving. It is still moving about 23 miles an hour, and you're still at 74% throttle and 3,300 revolutions per minute on the tachometer RPM gauge.

MR. WELCHER: I did.

MR. WELCHER: I do.

MR. WELCHER: Correct.

MR. WELCHER: I do.

MR. WELCHER: Sure. It's basically three different routes you can go. Sherman Street, Pleasant Street, Washington, and Sherman Street.

MR. WELCHER: Yes.

MR. WELCHER: So to go 2.3 miles in 4 minutes and 15 seconds, you would need an average speed of 32.47 mph.

MR. WELCHER: I did.

MR. WELCHER: It's 39.5 miles per hour.

MR. WELCHER: 1162.

MR. WELCHER: In 1162, the Techstream recorded two triggering events.

MR. WELCHER: I do.

MR. WELCHER: The actual trigger was excessive rearward acceleration. The event was during the course of a three-point turn.

MR. WELCHER: Yes.

MR. WELCHER: Approximately 8 minutes and 5 seconds.

MR. WELCHER: I do.

MR. WELCHER: It's again— the trigger event itself was a transition into reverse and then high throttle application. It's a forward motion followed by a lengthy high-speed rear impact—

MR. WELCHER: Yep. Yes.

MR. WELCHER: The triggering event was high throttle opening in a transition to reverse. It's a forward motion followed by a higher speed rear motion, and at the end of the trigger event, you're still at 74% throttle.

MR. WELCHER: Do I have an opinion about it?

MR. WELCHER: So, it obviously only captures what's going on during the trigger. It doesn't capture before or after the trigger.

MR. WELCHER: Yes.

MR. WELCHER: That it's 87 feet. So again, the initial forward motion is 34 feet, followed by it backing, crossing the initial point, and then backing up an additional 53 feet.

MR. WELCHER: At the end of that data captured at the triggering event 1162-2, it was accelerating at 74% throttle.

MR. WELCHER: Yes.

MR. WELCHER: Sure. It matches the window of the cell phone data for Mr. O'Keefe that puts it in front of 34 Fairview. So the backing event occurred somewhere between 12:32:04 and 12:32:12.

MR. WELCHER: I did.

MR. WELCHER: Yes, I did.

MR. WELCHER: Sure. So, my approach to this case is three prongs. I'm looking at the downloads, the technical information. I'm looking at this backing incident, and then I'm looking at Mr. O'Keefe's injuries and how they relate to the event potentially. So the second prong, which is looking at this data. So I had a Ring doorbell video showing Miss Read backing the Lexus out of Mr. O'Keefe's residence at approximately 5:00 in the morning. And so we wanted to look at the backing activity and whether that backing, contacting Mr. O'Keefe's vehicle, whether that was sufficient to cause — or what damage it could cause. In particular, was it sufficient to cause damage to the right rear tail light of the Lexus?

MR. WELCHER: Never.

MR. WELCHER: There's never been influence. The influence, again, it's whatever the data shows.

MR. WELCHER: No.

MR. WELCHER: I guess generally yes. I mean, I go through the data, I give you the results, you do with them whatever you see fit.

MR. WELCHER: Yes.

MR. WELCHER: Sure. Probably best on the slides. So, one of the things we first did was to get an exemplar Lexus. What does exemplar mean? It just means I got the same exact year, make, and model vehicle as Miss Read. So, Miss Read's vehicle is on the left-hand side, taken during my inspection. And then I found this exact same year, make, model, manufactured at the same plant, everything — Lexus in Texas. So even though it looks like a European plate, it's a Texas plate. Purchased the vehicle, had it shipped out to California. So I verified that the Lexus is in fact the one I got, the same exact one as Miss Read's. These are all the standard equipment. Miss Read's is on the left. My exemplar is on the right. It says Read Lexus TIS. That's Toyota information services — Toyota information system.

MR. WELCHER: So I ran it through TIS, which is the umbrella program under which Techstream runs. I also ran it through My Lexus, which is the Lexus website, and then started about doing what's referred to as a photogrammetric analysis of the video.

MR. WELCHER: Photogrammetric. We're doing photogrammetry — it's a photogrammetric analysis.

MR. WELCHER: Your normal kind of video camera has about 30 frames per second. Your standard movie is 25 frames a second. A modern IMAX is like 60 frames a second. The frame rate of the camera — how many images it shows per second — because video and films are just single images played really fast, so it looks like motion. And so determining the time between each frame tells you how much time occurred. So if we know the time between each frame, then we can figure out how far the vehicle moved; distance divided by time is speed. So we put it through a couple different programs. This is called Amped FIVE. We got 19 frames a second. This is a specific forensic program called Axon Investigate — 19.8 frames a second. This is a publicly available program called FFMS — same thing, 19.8 frames a second.

MR. WELCHER: Another program, MediaInfo — 19.8 frames a second. Anyways, five different sources we're getting 19.8 frames a second. So then what we've done is add information now. So now that we know how many seconds each frame represents, we now put a clock back in to tie that into time. So it's the same exact video except we have a frame counter now and we have a time. So we have some information now on the time between each frame. So, here's frame zero. This is frame 265 — we're 13 seconds into it. Here's frame 305. Notice you see brake lights on the back wall here. Notice how they're not super red because you still have reverse lights on. Now, they all go off. So, the Lexus brake lights have been turned off at frame 353, 356. I say brake lights begin to come on. 357, they're back on.

MR. WELCHER: Lexus comes to a complete stop at frame 369. 384, the reverse lights turn off. So now look how red this is because we no longer have the white reverse lights in it. And you can see the Lexus starts to drive forward. So frame 825. This is after the Lexus has left the scene. But notice the tracks in the snow. So we have very definitive locations for the wheels of the Lexus. We lay over a grid on the various frames. So we then took the images and did what's called clarification, which is where you overlay images from multiple versions or multiple views of the same vehicle, and it helps sharpen the image, but it slightly dulls the color, so you get a little cleaner, crisper image. So, here's Mr. O'Keefe's vehicle before clarification, after clarification.

MR. WELCHER: Absolutely. Okay. I'll make an effort. So, uh, slide 76. You're welcome.

MR. WELCHER: We're going to focus in on this area here. So what we will actually show is where the snow was displaced in these two spots right across from here and right across from here. If you could focus your attention on those, slide 77. So, slide 78. To bring us all together, we went out to Mr. O'Keefe's residence and took a scan, a laser scan. So you see the laser scanner over here. Laser scanner is right here. Fence hasn't changed position. The camera's changed slight position — we can account for that. So, here's the subject video. We have the correct resolution. There's a new camera. It's going slightly faster frame rate. That just means we have different times between frames on the new camera, but we're simply using it for calibration to the old camera.

MR. WELCHER: It's the same container for the MP4 and it's using the same video codec. A video codec is how the computer encodes the video.

MR. WELCHER: Oh yes, thank you, your honor. Slide 79. Slide 79 part two. So really we care most about is the resolution. So slide 80. You can see these are what are called control points. So from the laser scan we locate these various points. So the 3D scanner knows the exact position of all these points on slide 80 in the image. And so what we can do is use this information in conjunction with the video to create a scaled version of the video. So when we have three-dimensional images or three-dimensional measurements, we can then use that to give us some three-dimensional information from our 2D video. So there's the points on slide 81. Slide 82 — excuse me — slide 81. Second part, we're basically just fading in.

MR. WELCHER: So what you see now, if I go back to the beginning of slide 81, this is the picture from the video camera. And then what you see here is — which is barely any difference — and that's because what we've done is we've now overlaid the points from the laser scan. So now each one of these little speckly points, we know the exact 3D coordinates of those. So we can now know the exact position of all these fence posts, all the positions on the driveway, and we can use that to calibrate. Going to slide 82. So using that information, it's called solving for the camera. We verify that we have appropriate residuals, meaning we don't have too much error. Slide 83. Okay. Subject video. Then this is the — excuse me — the subject video with the control points overlaid on it.

MR. WELCHER: That's a subject video where we're phasing in the laser scans of the scene, slide 84. So these are both on slide 84. I'm just toggling between the point cloud and the video. And what I'm trying to illustrate is how well we were able to overlay the point cloud onto the video data. So, for example, if you watch one of these fence post points up here while I toggle between the video — there's the laser scan, there's the video. So we've done a good job. Again, the residuals show that we've done an adequate job of matching the data. Slide 85. So what I'm doing here is toggling between — this is the video of the car. Remember, I told you I went out and I inspected Mr. O'Keefe's vehicle. I did a laser scan of it.

MR. WELCHER: So we're now taking that laser scan — excuse me — of the vehicle and overlaying it onto the car in the video. So this is the video image. That's the laser scan laid on top of it. So we know all the dimensions in the laser scan. We know all the dimensions in the scene. We have the tire tracks. We can create and take data points along the tire tracks, create a map of those. So green is the right rear. Again, the vehicle's backing out, coming — slide 87 — coming out this direction, backing here, stopping, pulling, pulling, pulling forward towards the bottom of the screen, backing out here, pulling forward like here. Slide 88. So this is the Lexus lined up. This is when the Lexus has come to a complete stop. So this would be maximum engagement.

MR. WELCHER: Again, this is just toggling frame 88 showing we have the time and the frame number. 369 on the same slide. We've now overlaid the Lexus laser scans on top of the image of the Lexus. So there's the video image, a still image that we took from the video. And here's where we overlay the laser scan of the Lexus. So we've overlaid the laser scan of the Lexus onto the Lexus. We've overlaid the laser scan of the Chevrolet onto the Chevrolet. And we've overlaid the laser scan of the scene onto the scene. So now what that allows us to do — we now know the position of everything in there. And we've also tracked the position of the tire tracks. So that's what it looks like if you overlay the computer model of Miss Read's vehicle onto the tire tracks.

MR. WELCHER: Oh, my apologies, your honor. Slide 89. And then I'm just toggling slide 89 where I turn the vehicle off. So you can see how the tire positions match up. Another toggle of slide 89 — how the tire positions line up with where the end of the snow was. From all that information, again, we lined it up in the video, we lined it up in the scene, we have the two vehicles, we can get the exact position of the vehicles when that Lexus stopped. So when that Lexus stop, that's going to be the maximum engagement. So from that information we can get the angle of the vehicle. We can get the degree of overlap, the impact configuration, and it's hard to see because of the angles and because these vehicles are not at right angles to each other, the exact contact.

MR. WELCHER: Now I mentioned — and we're on slide 94 now — that contact on the rear. So we took a whole bunch of data points within those two areas where the snow got dislodged, and we're looking at where that is. So if you look at that on the vehicle — this is where it is on the rear of Mr. O'Keefe's Chevrolet. You see it's about 25 and a half — 25½ inches above ground. 32½ inches above ground. Here's Mr. O'Keefe's vehicle. So these marks are, kind of — this is the Chevy Traverse — the marks are approximately right here. Again, I had the opportunity to inspect Mr. O'Keefe's vehicle. There's no damage where these marks are. So, sure, it knocked the snow off, but it wasn't sufficient to cause damage to his vehicle. We then mapped that data onto Miss Read's vehicle to look at where that height is.

MR. WELCHER: So you can see the only contact marks are down at the bumper level. I mentioned how that bumper sticks out. This is exactly where you'd expect it. So the only evidence of contact is nowhere near the upper tail light. I emphasize the word upper tail light because it is right in line with the lower tail light. But yet the lower tail light also was not broken. So again, 25 and a half — that's 24 right here. 25 and a half — that's right where that scuff mark was. It's also right at again the level at the top of that light. This light is not broken. I've looked at the front of it. I've looked at the back of it. I looked all around it. It's not broken. And it sticks out further than the light above it.

MR. WELCHER: So using those positions, matching the overlay of the vehicles, we can get the speeds from that. We know where the Lexus is. We know how much time it is between each frame. So we know how far it moved. And so what we know is that when it first came to a stop — right before that — it was going about 7 miles an hour. So, realistically somewhere between a half to 1 mile an hour, just to account for some variations, but definitely going 1 mile an hour or less at impact. Slide over, please. Oh, thank you, your honor. Slide 99. That's the dimensions. Slide 102. Similar to what I said, no damage to that light, no damage to Mr. O'Keefe's Traverse in that area. And then if I could, I would like to go back because I realized I skipped over something. So this is slide 85. So this is Mr.

MR. WELCHER: O'Keefe's vehicle before contact. Huh? I don't know what happened to this one. The slide shows that there was snow here and now there's no snow there.

MR. WELCHER: So, referring to this slide, this is after the Lexus has pulled away. Looking in the snow, I see nothing in the snow.

MR. WELCHER: Of course I was. I was looking for tail light fragments.

MR. WELCHER: I do not.

MR. WELCHER: I did.

MR. WELCHER: Consider speed, and then again, knowing we have the positions of the vehicle, we looked at the degree of overlap, and there's only about 2 to 3 mm of overlap between the vehicles, which is insufficient to cause the tail light — that's recessed in from the edge on the right rear of the Lexus — to contact the rear of the Chevrolet.

MR. WELCHER: That impact did not break or crack that tail light.

MR. WELCHER: That's my opinion.

MR. WELCHER: That's to a high degree of engineering certainty.

MR. WELCHER: So the third area of study I'm now focusing in on — the injuries, the at-scene physical evidence, tail light fragments, broken glass, things of that nature.

MR. WELCHER: Yes, sir.

MR. WELCHER: Absolutely. Slide 104, please.

MR. WELCHER: I'm sorry. 105. Yes. Sorry. Slide 105. This is a picture of Mr. O'Keefe's right arm, forearm, hand, and wrist. In anatomy, your arm is defined from your shoulder to your elbow. And then from your elbow to your wrist is your forearm. And then this is your hand. So technically, when you say arm, it really means from your elbow to your shoulder. Most people call this whole thing arm. It's actually the upper extremity. Arm is from shoulder to elbow. Elbow to wrist is forearm and hand. So you can see I wrote arm and forearm. So arm, elbow to shoulder; elbow to wrist is forearm. So looking at the injuries to Mr. O'Keefe's arm, forearm, as well as his hand. So this is me. I'm actually 6 feet and a half inch tall. What you see in the upper right is the coroner's examination report for Mr.

MR. WELCHER: O'Keefe. He's 73 inches tall. That's 6'1" tall. At this time, I was 220 pounds. Mr. O'Keefe at the time of this incident was 216 pounds. So I took measurements of myself. I'm actually wearing the same make and size shoes as Mr. O'Keefe, the same brand of jeans as Mr. O'Keefe, the same brand and size shirt as Mr. O'Keefe. You see the Lexus in the background. We have matched the height to the Lexus. What I'm showing here is both my height and then, keep in mind, where the height of this wing is relative to my right eye. So now I'm wearing, in addition, the same shirt and shoes, same type of sweatshirt, and actually the same kind of baseball hat. We tried to determine the exact glass from the Waterfall Bar & Grille.

MR. WELCHER: Found out they were getting different ones at different times, but I ordered my glass from the same distributor that Waterfall Bar & Grille ordered their glasses from. So you can see in my right hand here I have a glass. Here's a glass. It looks similar. It's from the same location. And then in the lower right photograph, what I've done is I've rolled up my sleeve, taken the sweatshirt off, and I'm going to look at how this contact pattern lines up. So the first set of slides, I'm simply standing next to the Lexus. We know we have a broken tail light. We know we have scratches on Mr. O'Keefe's arm. I'm seeing how those may or may not relate to the height and position of the tail light.

MR. WELCHER: I call it a wind wing. I mean, it could be a wing.

MR. WELCHER: Air dam would be another. Okay, so I then set about to do a little bit of testing with the Lexus. The first test you're going to see is me just walking backwards — or actually sideways — into the Lexus. What I've done is I've painted the Lexus right rear tail light with what's called grease paint — kind of sort of like paint, but it doesn't dry. And then I have backed up sideways, side-shuffled into the Lexus with my arm out to see what the damage pattern looks like, or basically where someone of Mr. O'Keefe's height's arm would contact the rear tail light. So then what you see is I've shown the photographs again. I show the paint transfer onto my arm. The yellow grease paint on my hand was to show where my hand would hit the tail light.

MR. WELCHER: And so just kind of me walking slowly sideways into it. The approximate location of the tail light — and I'll line it up with the broken tail light in a second — lines up with the approximate location of the lacerations on Mr. O'Keefe's arm.