Latest / The Vault: The Epstein Files / Tyler Robinson Preliminary Hearing Transcripts (Day 2 Part 17) (9/30/26)

Transcript

Tyler Robinson Preliminary Hearing Transcripts (Day 2 Part 17) (9/30/26) · The Vault: The Epstein Files

• 0:00
  Speaker 1: What's up, everyone, and welcome back to the program. In
• 0:03
  Speaker 1: this episode, we're going to pick up where we left
• 0:05
  Speaker 1: off with the Tyler Robinson prelim hearing day two. So
• 0:09
  Speaker 1: they get back from the break, and once again, we
• 0:12
  Speaker 1: have the defense questioning Agent Baker. Question, Ms. Baker, would
• 0:17
  Speaker 1: you take a look at your notes? Page 24 of 379,
• 0:20
  Speaker 1: the very bottom of the page. Did you write there
• 0:25
  Speaker 1: that on the 15th of September, 2025, spoke with Special
• 0:29
  Speaker 1: Agent Folger and gave update on report timeline, discussed it
• 0:33
  Speaker 1: upon review. It was decided that additional statistics needed to
• 0:37
  Speaker 1: be run. Report will still be issued with twigs as
• 0:41
  Speaker 1: an assumed contributor per previous discussion with Folger and Carmack.
• 0:44
  Speaker 1: Do you see that? Answer, yes, I do. Question, and
• 0:47
  Speaker 1: what additional statistics needed to be run at that time? Question,
• 0:51
  Speaker 1: so at the time, I believe I ran with just
• 0:54
  Speaker 1: twigs comparing that individual to the sample. If I may refer...
• 0:58
  Speaker 1: I think that was the only additional. Yes, after discussing
• 1:02
  Speaker 1: with my reviewer, it was determined that I was also
• 1:04
  Speaker 1: going to calculate a statistic or compare twigs by himself.
• 1:08
  Speaker 1: Comparing him to the evidence and having that run in
• 1:11
  Speaker 1: my case file. Question. Okay. And could you go to
• 1:15
  Speaker 1: tab 38, page 39? And then they have some issues
• 1:19
  Speaker 1: with this evidence being shown on the screens. And after
• 1:22
  Speaker 1: they get that worked out, the questioning begins again. Question. Okay.
• 1:26
  Speaker 1: And this is, yeah, page 39. Is this the DNA report?
• 1:29
  Speaker 1: I'm sorry, the STR mix report that was produced as
• 1:32
  Speaker 1: a result of your inquiry that involved a comparison to
• 1:35
  Speaker 1: the sample to Mr. Twigs alone? Answer, yes, it is. Question,
• 1:40
  Speaker 1: and is that a six-page report? Answer, yes, it is. Question,
• 1:44
  Speaker 1: and is it true and accurate copy of the STR
• 1:47
  Speaker 1: mix report for that particular sample? Answer, yes, it is. Question,
• 1:52
  Speaker 1: And move that into evidence, Your Honor. This would be
• 1:54
  Speaker 1: Baker 38, page 39 through 44. I'm seeking admission as
• 1:59
  Speaker 1: well as publication to the audience and the camera. Judge Graff,
• 2:03
  Speaker 1: I just want to make sure that I have the
• 2:05
  Speaker 1: label correct. Baker 38, pages 39 through 44. Did I
• 2:09
  Speaker 1: hear that correctly? Answer, that is correct, Your Honor. Judge,
• 2:14
  Speaker 1: all right, to the state. Prosecution. Yes, your honor. I
• 2:18
  Speaker 1: object on grounds of relevance to the probable cause determination.
• 2:23
  Speaker 1: Judge Graff, all right. And defense, would you like to respond? Defense, admitted,
• 2:27
  Speaker 1: your honor. Judge, so I'm not exactly sure what this
• 2:31
  Speaker 1: is in order to determine how it fits into the
• 2:34
  Speaker 1: probable cause analysis. So a little bit more information because
• 2:37
  Speaker 1: as of right now, I don't have enough to admit it. Defense, understood.
• 2:41
  Speaker 1: And I apologize to the court. So the report that
• 2:44
  Speaker 1: was admitted into evidence has a conclusion in it that
• 2:47
  Speaker 1: both samples were contributed, that Mr. Twiggs was a contributor
• 2:51
  Speaker 1: to both samples. And this is the report that the
• 2:54
  Speaker 1: analyst ran in order to make that determination, I believe. Question.
• 2:57
  Speaker 1: All right. And did these reports, were these made prior
• 3:00
  Speaker 1: to State Exhibit 31 being prepared? Answer. Yes, that is correct. Question.
• 3:05
  Speaker 1: All right. I'll admit. So Baker 38 pages 39 through
• 3:07
  Speaker 1: 44 is admitted and may be published. Again, As a
• 3:13
  Speaker 1: reminder for all attorneys, this is a probable cause standard.
• 3:17
  Speaker 1: So if all parties can just keep that in mind
• 3:19
  Speaker 1: as we're moving forward on this, thank you. Now they
• 3:24
  Speaker 1: jump back in. Counsel, do you see a page number
• 3:27
  Speaker 1: on this? Yeah, it's page 39, is the answer. No,
• 3:32
  Speaker 1: the actual dates I'm having to identify in the file.
• 3:37
  Speaker 1: Defense answers, so is this the report that you referenced?
• 3:41
  Speaker 1: In your notes that you needed to run the sample
• 3:43
  Speaker 1: through STR Mix? Answer by Ms. Baker, yes. Defense, now
• 3:49
  Speaker 1: at the top, is this the STR Mix unit dated
• 3:51
  Speaker 1: September 15th and it relates to item 7-1, correct? Answer,
• 3:56
  Speaker 1: that is correct, yes. Question, all right. And then down
• 3:59
  Speaker 1: below that, the kit used, if you could go to
• 4:01
  Speaker 1: the summary of the input, you tell the software program,
• 4:05
  Speaker 1: assume that there are two contributors, right? Answer, yes. Question,
• 4:09
  Speaker 1: you input sample 7A. Answer, correct, yes. Question. And when
• 4:14
  Speaker 1: you input it, it's just the raw data, right? Answer, correct,
• 4:18
  Speaker 1: just the numbers from the graph. Question, the numbers. And then,
• 4:22
  Speaker 1: known contributors under HP, the prosecution hypothesis. Answer, correct, yes. Question.
• 4:28
  Speaker 1: And you put in sample 61, right? Answer, yes. Question.
• 4:33
  Speaker 1: That's Twig's known sample. Answer, correct, yes. Question. And then,
• 4:38
  Speaker 1: the known contributors under the defense hypothesis. You left that blank. Answer, correct. Question,
• 4:45
  Speaker 1: is that because you didn't know what the defense hypothesis was? Answer, no. Again,
• 4:49
  Speaker 1: the defense hypothesis for this situation, if I'm comparing someone,
• 4:54
  Speaker 1: is an unknown unrelated individual that is taken into account
• 4:58
  Speaker 1: during the calculation. Question, okay. And then you input that
• 5:02
  Speaker 1: data and what you get out of that is a
• 5:04
  Speaker 1: likelihood ratio, right? Answer, yes. Question, so let's go to
• 5:09
  Speaker 1: page two of that report. And is this a summary
• 5:11
  Speaker 1: of the results from that test on the 15th in
• 5:14
  Speaker 1: terms of a likelihood ratio? Answer, yes, it is. Question,
• 5:18
  Speaker 1: what is a likelihood ratio? What does that mean? Answer,
• 5:21
  Speaker 1: a likelihood ratio is a statistic that calculates the probability
• 5:25
  Speaker 1: of seeing a DNA profile given one or two different hypotheses.
• 5:32
  Speaker 1: One is that a particular person of interest is a contributor,
• 5:35
  Speaker 1: and the other is that an unknown unrelated individual is
• 5:39
  Speaker 1: a contributor. The probability with the higher number gets the support,
• 5:42
  Speaker 1: and the magnitude of the support is determined by the
• 5:45
  Speaker 1: ratio of those two numbers. Question, that's pretty technical stuff,
• 5:50
  Speaker 1: but the bottom line is that the software computes the numbers,
• 5:53
  Speaker 1: and then you look at those numbers and determine whether
• 5:56
  Speaker 1: it has any significance, right? Answer, I don't know if
• 5:59
  Speaker 1: I would say if it has significance. I report the
• 6:02
  Speaker 1: value that's calculated. Question right, and why is one column
• 6:06
  Speaker 1: these columns relate to different that are called population databases, right? Answer,
• 6:11
  Speaker 1: that is correct, yes. Question, the one in the far left, BAH,
• 6:15
  Speaker 1: is an FBI population database on Bohemians. Answer, an African-American, yes. Question, African-Americans.
• 6:24
  Speaker 1: Another one for Caucasians. Answer, correct, yes. Question, SEH is
• 6:29
  Speaker 1: Southeast Hispanics. Answer, correct. Question, and then Southwest Hispanics. Answer, correct.
• 6:37
  Speaker 1: Objection now. By the prosecution. Defense. That's all I have
• 6:41
  Speaker 1: on that. I'm just trying to explain the chart. Judge.
• 6:44
  Speaker 1: All right. Well, there's an objection. Let's pause and turn
• 6:47
  Speaker 1: to the state. The state, your honor. At this point,
• 6:50
  Speaker 1: I'm going to object on, again, grounds of relevance to
• 6:53
  Speaker 1: probable cause hearing. State versus Drosbeck. State versus Merrill. State
• 6:57
  Speaker 1: versus Ramirez. That the court has cited all hold the
• 7:01
  Speaker 1: court should bind the defendant over for trial unless the
• 7:04
  Speaker 1: evidence is wholly lacking and incapable of reasonable inference. to
• 7:07
  Speaker 1: prove some issues as supports the prosecution's claim. I don't
• 7:12
  Speaker 1: know that any of the analysis goes to whether or
• 7:14
  Speaker 1: not the expert's testimony in her report is wholly lacking
• 7:18
  Speaker 1: and incapable of reasonable inference to prove a result in
• 7:21
  Speaker 1: the case. We're far beyond this. We're hours into the examination.
• 7:25
  Speaker 1: This feels like a 702 hearing to me. I think
• 7:28
  Speaker 1: we're far beyond a probable cause hearing, Your Honor. I
• 7:31
  Speaker 1: would ask the court to sustain the objection and not
• 7:33
  Speaker 1: allow questioning on this or any other matter that does
• 7:36
  Speaker 1: not deal with the issue of probable cause. Thank you,
• 7:40
  Speaker 1: says the judge. Defense? Defense, your honor. The state offered
• 7:45
  Speaker 1: this report on a very technical, complex scientific area of inquiry,
• 7:50
  Speaker 1: and they offered to the court conclusions in that report
• 7:52
  Speaker 1: that Mr. Robinson and Mr. Twiggs were contributors to the
• 7:56
  Speaker 1: two samples that were found at the crime scene. They
• 7:59
  Speaker 1: obviously think that that's relevant to the probable cause inquiry.
• 8:02
  Speaker 1: and they wouldn't have offered the report in the first place.
• 8:05
  Speaker 1: So having done that, I think the defense is entitled
• 8:08
  Speaker 1: to inquire whether those conclusions are reliable, because that, again,
• 8:13
  Speaker 1: is the court's ultimate determination here, reliability of evidence. And
• 8:17
  Speaker 1: the court can decide it's not relevant, I'm sorry, it's
• 8:19
  Speaker 1: not reliable, therefore I'm not going to rely on that
• 8:23
  Speaker 1: in determining probable cause. And that, or the court could
• 8:26
  Speaker 1: say it's reliable and bolsters up other evidence that the
• 8:29
  Speaker 1: court is going to hear from. So the particular report
• 8:32
  Speaker 1: in front of us goes directly to the issue of
• 8:35
  Speaker 1: whether the expert correctly concluded that Twiggs was not a
• 8:38
  Speaker 1: possible contributor, but was a contributor. And the state's theory here, apparently,
• 8:43
  Speaker 1: is that both Mr. Robinson and Mr. Twiggs are somehow
• 8:47
  Speaker 1: connected to this crime scene. So I think when the
• 8:49
  Speaker 1: court sees that, and what the numbers were generated how
• 8:52
  Speaker 1: that compares to her own standards for assessment, the court
• 8:55
  Speaker 1: will see that the conclusion is not justified that she reached.
• 8:59
  Speaker 1: Judge Graff, All right, well, I understand what both parties
• 9:02
  Speaker 1: are saying. And at a probable cause hearing, the issue
• 9:05
  Speaker 1: is whether we walk one mile or 100 miles down
• 9:08
  Speaker 1: the path. And so I want to keep it limited
• 9:11
  Speaker 1: to one mile as opposed to 100 because that would
• 9:14
  Speaker 1: exceed the probable cause standard for a preliminary hearing for
• 9:18
  Speaker 1: what's before me. And so defense, as it relates to
• 9:22
  Speaker 1: the objection, I'm going to sustain it in part. If
• 9:27
  Speaker 1: you want to briefly address this and move on, you
• 9:29
  Speaker 1: certainly may choose to do so, but as it relates
• 9:32
  Speaker 1: to where we're at, my hope is we can keep
• 9:34
  Speaker 1: this very brief. Thank you, Judge, Your Honor, and in
• 9:39
  Speaker 1: interest of doing that, would you bring up the same
• 9:41
  Speaker 1: screen part of her report, People's Exhibit 31? Judge, and
• 9:46
  Speaker 1: let's take that off the screen. Defense, I'm going to
• 9:49
  Speaker 1: need that up because I need to do it. Judge, oh,
• 9:52
  Speaker 1: so that's part of the same because it was on
• 9:54
  Speaker 1: the screen. I want to make sure that what you're
• 9:56
  Speaker 1: calling up on the screen right now is this part
• 9:59
  Speaker 1: of what was already admitted or something else? Answer by
• 10:02
  Speaker 1: the defense, it's part of this, which has already been
• 10:05
  Speaker 1: admitted compared to plaintiff's exhibit 31, which has also been admitted.
• 10:10
  Speaker 1: Judge Graff, I see. Okay, thank you. Question, they both are. Judge,
• 10:14
  Speaker 1: I just wanted to clarify. Defense, sure. Judge, all right.
• 10:18
  Speaker 1: And do you want that published on the screens? Answer, yes. Judge,
• 10:22
  Speaker 1: all right, let's go ahead and do so. Question, and
• 10:25
  Speaker 1: the chart I'd like you to bring up from that
• 10:27
  Speaker 1: report is on page five. Yeah, and just highlight that.
• 10:30
  Speaker 1: Thank you. So what we've blown up, there is from
• 10:34
  Speaker 1: your report, Plaintiff 31, and you provided this chart in
• 10:37
  Speaker 1: your report, correct? Baker, that is correct, yes. Question, explain
• 10:41
  Speaker 1: to the court how this works. What's this chart and
• 10:44
  Speaker 1: what does it represent? Answer, so this is our verbal scale.
• 10:48
  Speaker 1: In addition to the numerical likelihood ratio value that I provide,
• 10:52
  Speaker 1: I also give a qualitative or verbal equivalent to the
• 10:55
  Speaker 1: numerical value that gives some kind of context to what
• 10:58
  Speaker 1: that likelihood ratio means. Question, okay, and so on the left,
• 11:03
  Speaker 1: explain what the greater than 1 million less than 1
• 11:05
  Speaker 1: in 100 means. What does that scale relate to? Answer, sure,
• 11:10
  Speaker 1: so that scale is the numerical value of the likelihood ratio.
• 11:14
  Speaker 1: So if I have a likelihood ratio that's greater than
• 11:16
  Speaker 1: or equal to 1 million, I designate that as very
• 11:19
  Speaker 1: strong support for inclusion. The next example, 10,000 to less
• 11:23
  Speaker 1: than 1 million, I would say that's very strong support
• 11:26
  Speaker 1: for inclusion and so forth. Question, so you need to
• 11:29
  Speaker 1: reach a likelihood ratio of two in order to get
• 11:33
  Speaker 1: even limited support for an inclusion, correct? Answer, correct, yes. Question, okay,
• 11:39
  Speaker 1: if you could not have, reduce that and go to
• 11:41
  Speaker 1: page two of the report on the left. So this
• 11:43
  Speaker 1: is from the chart when you ran Mr. Twiggs against
• 11:46
  Speaker 1: the sample. What was the likelihood ratio that you determined? Answer,
• 11:51
  Speaker 1: I believe this chart is not the comparison to Mr. Twiggs. Question,
• 11:55
  Speaker 1: it's not the report? Answer, the chart at the bottom
• 11:59
  Speaker 1: of the screen is not my comparison of Mr. Twiggs. Question,
• 12:03
  Speaker 1: let's verify that if we could. Take it down, the highlight,
• 12:06
  Speaker 1: and repost 38, page 45, I'm sorry, page 39. That's
• 12:10
  Speaker 1: in evidence. There we go, second page. Is this the
• 12:13
  Speaker 1: likelihood ratio from your tweaks comparison? Answer, yes, it is. Question, okay,
• 12:19
  Speaker 1: now what was the likelihood ratio that you calculated based
• 12:22
  Speaker 1: on this chart? Answer, the likelihood ratio is one. Question one,
• 12:26
  Speaker 1: answer yes. Question, so that doesn't rise to the level
• 12:30
  Speaker 1: of even limited inclusion, correct? Answer, correct. It would fall
• 12:35
  Speaker 1: under uninformative. Question uninformative. And yet you concluded that despite
• 12:42
  Speaker 1: this STR mix reading with a likelihood ratio of 1,
• 12:45
  Speaker 1: that Mr. Twiggs was a contributor to both item 7
• 12:48
  Speaker 1: and item 8, correct? Answer, I did. So actually, I
• 12:52
  Speaker 1: didn't utilize this calculation to determine that Mr. Twiggs was
• 12:55
  Speaker 1: a possible contributor or that I could use him as
• 12:58
  Speaker 1: an assumed contributor. I visually looked at the DNA sample
• 13:02
  Speaker 1: and the DNA. Unlike that major contributor, he also has
• 13:05
  Speaker 1: those peaks. So that was one proposition or one hypothesis
• 13:10
  Speaker 1: that I determined that he was an assumed contributor. There
• 13:14
  Speaker 1: are multiple different ways. We don't have a threshold. I
• 13:17
  Speaker 1: don't have to see two, four, eight peaks to determine
• 13:20
  Speaker 1: I think an individual is present given that the field
• 13:22
  Speaker 1: submitted him as an elimination. I assumed he was a contributor.
• 13:27
  Speaker 1: When I compared him to that data, I did see
• 13:29
  Speaker 1: his DNA present. So when I calculated the statistic for Mr. Robinson,
• 13:33
  Speaker 1: I used twigs as an elimination. Question, let me see
• 13:37
  Speaker 1: if I understand that. The STR mix that you relied
• 13:40
  Speaker 1: upon to include Mr. Rob, you relied on the STR
• 13:43
  Speaker 1: mix to include Mr. Robinson as a possible contributor? Correct. Answer, correct, yes. Question,
• 13:51
  Speaker 1: but when you ran Mr. Twig's sample through STR mix,
• 13:53
  Speaker 1: the result you got was that he was not a contributor, correct?
• 13:56
  Speaker 1: Or that he was not even of limited value including him? Answer,
• 14:00
  Speaker 1: based on STR mix run, it was uninformative. But I
• 14:04
  Speaker 1: did use my own knowledge and training to determine... that
• 14:07
  Speaker 1: I was going to utilize him as an assumed contributor. Question, okay.
• 14:11
  Speaker 1: Your own knowledge and training, is there some standard that
• 14:14
  Speaker 1: you use for deciding when not to rely on STR
• 14:17
  Speaker 1: mix results and when to sort of disregard it and
• 14:20
  Speaker 1: rely on your training and experience? Answer, I can't say
• 14:23
  Speaker 1: that I disregarded it. When I'm comparing someone, that's my
• 14:26
  Speaker 1: first determination. Do I think this person is potentially included
• 14:29
  Speaker 1: or excluded? If I determine that I think they're included,
• 14:33
  Speaker 1: I'll run STR mix to give weight to that inclusion.
• 14:37
  Speaker 1: Elimination of samples are different. If I'm told that an
• 14:40
  Speaker 1: individual is submitted as an elimination sample, I can assume
• 14:44
  Speaker 1: that they're there. Visually, I'm able to look at the
• 14:47
  Speaker 1: DNA evidence and determine, yeah, I think there's enough evidence
• 14:51
  Speaker 1: there to assume that individual, which is what I did here. Question? Okay.
• 14:57
  Speaker 1: All right, folks, we're going to wrap up right here.
• 14:59
  Speaker 1: And in the next episode, dealing with the topic, we're
• 15:01
  Speaker 1: going to pick up where we left off. All the
• 15:03
  Speaker 1: information that goes with this episode can be found in
• 15:06
  Speaker 1: the description box.