Demonstrative Video
Evidence That Survives
Daubert

Chapter 01

What Demonstrative Evidence Is, and Why It Wins Before Trial Starts

A demonstrative is evidence you make to prove a fact, not evidence you found. A photograph pulled from the scene is real evidence. A video built afterward to show how the incident happened is a demonstrative. The distinction sounds academic until a judge excludes the second one and the case goes to trial with only the first.

Evidence versus proof of evidence

Real evidence is the thing itself: the firearm, the rope, the security footage that happened to be running. A demonstrative is something you build to explain or prove what the real evidence means. It didn't exist until your team made it. That single fact is why courts scrutinize a demonstrative that they would never scrutinize a photograph someone happened to take at the scene.

Where the video actually does its work

Most demonstratives never reach a jury. They reach a claims adjuster reviewing exposure before a mediation. They reach opposing counsel deciding whether to recommend a settlement number to a client. They reach a judge deciding a summary judgment motion. The video that changes a case usually does its work in a conference room, months before trial, when someone with authority to settle watches it and reaches a different number than the one they walked in with.

The cost of getting it wrong

A demonstrative that gets excluded doesn't just disappear. It tells the other side your case had a hole they can now point to. A demonstrative that survives to trial but reads as staged or misleading can cost you credibility on everything else you say from the witness stand. The video is never a free swing.

What this book assumes you already know

This book starts where Photography and Videography for Expert Witnesses ends. That book teaches you to capture evidence: camera settings, lighting, scene documentation, chain of custody at the point of capture. This book assumes you can already do that, and teaches you to build the finished exhibit that carries a case: the legal standard it has to clear, the foundation it needs, and the argument it has to survive.

Chapter 02

Illustrative Aid or Substantive Evidence: The Line That Sets Your Standard

Two exhibits can look identical on screen and answer to completely different rules. One is a diagram a witness points to while explaining an opinion already in evidence. The other asserts, on its own, that this is what happened. Courts call the first illustrative and the second substantive, and the label decides how hard you've to work before anyone sees it.

What makes something illustrative

An illustrative aid supports testimony that's already admissible on its own. The witness could describe the same thing in words. The exhibit just makes the words easier to follow. Because the aid carries no independent weight, most courts don't require it to pass Daubert on its own, though the underlying testimony still has to.

What makes something substantive

Substantive, or independently probative, demonstrative evidence asserts a fact by itself. A simulation that reconstructs a fall, a recreation that shows a firearm discharging without a trigger pull: these don't just help the jury follow testimony, they tell the jury what happened. Because the exhibit itself is doing the proving, it has to clear the same admissibility bar as the opinion it represents.

Why the same video can be argued two different ways

Opposing counsel will often argue your illustrative aid is actually substantive, because the substantive standard is harder to meet. Expect this. The way you introduce the exhibit, and the limiting instruction you request, decide which label sticks more than the content of the video does.

How the label changes your foundation testimony

For an illustrative aid, your foundation testimony connects the exhibit to testimony already in evidence. For substantive evidence, your foundation testimony has to independently establish reliability, meaning methodology, data, and error rate, before the judge lets the jury see it. Know which foundation you're laying before you take the stand, not while you're on it.

Chapter 03

Daubert, Frye, and Federal Rule of Evidence 702: The Gatekeeping Rules You're Building For

Every demonstrative you build answers to a judge before it answers to a jury. Federal Rule of Evidence 702 is the rule that judge is reading. Daubert is the case that explains how to read it. Knowing both cold, before you shoot a single frame, separates evidence that survives a challenge from evidence that becomes one.

Federal Rule of Evidence 702, the rule you're actually building to

Federal Rule of Evidence 702 lets a qualified expert testify in the form of an opinion if the testimony will help the jury, rests on sufficient facts or data, is the product of reliable principles and methods, and reflects a reliable application of those principles and methods to the facts of the case. The amendment effective December 2023 made explicit what many courts already required: the proponent, meaning you and the retaining attorney, has to show admissibility by a preponderance of the evidence. The rule no longer gives a demonstrative the benefit of the doubt.

The five Daubert factors

Daubert v. Merrell Dow Pharmaceuticals gave judges five questions to test reliability: can the method be tested, has it been subjected to peer review and publication, does it have a known or potential error rate, do standards exist that control how the technique is applied, and is the method generally accepted in the relevant field. No demonstrative clears all five with equal strength. Know your weakest factor before opposing counsel finds it.

Why Kumho Tire matters for your kind of testimony

Kumho Tire Co. v. Carmichael extended Daubert's gatekeeping role beyond laboratory science to technical and experience-based expertise, exactly the kind of knowledge a stunt coordinator, an armorer, or a wildfire behavior analyst brings. You don't get to skip Daubert because your expertise comes from decades on set rather than a research lab. You do get to argue that the five factors apply differently to skill-based methodology than to a controlled experiment, and that argument has to be made explicitly, not assumed.

The states that still use Frye

A minority of states, including California and Illinois, still apply Frye v. United States: general acceptance in the relevant scientific community, without the broader reliability inquiry Daubert requires. A Frye jurisdiction asks a narrower question, but general acceptance can be harder to prove for a novel reconstruction method than reliability is under Daubert. Confirm your jurisdiction's standard before you design the methodology, not after.

General Electric Co. v. Joiner, and why losing at the trial court is hard to undo

General Electric Co. v. Joiner set the standard of review for a Daubert ruling at abuse of discretion, the most deferential standard an appellate court applies. If a trial judge excludes your demonstrative, the odds of reversing that ruling on appeal are low. The work in this book happens before the motion, because there's little room to fix it after.

Chapter 04

How Demonstratives Get Excluded, and How to Build Around Every Reason

Judges exclude demonstratives for a small number of recurring reasons. Learn the list before you build, and you spend your budget avoiding them instead of discovering them in a motion.

Condition mismatch

A recreation shot under different lighting, distance, materials, or mechanical condition than the real event invites exclusion on similarity grounds alone, no matter how sound the rest of the methodology is. This is the single most common reason a recreation gets thrown out. Chapter six covers how to prevent it before you film.

Undisclosed methodology

A demonstrative built on a method the report never explained is a demonstrative the other side never had a fair chance to test. Courts treat that as a discovery problem as much as a reliability problem, and discovery problems get sanctioned harder than reliability disputes do.

Missing or overstated error rate

Every method has an error rate, even if the honest answer is that it has never been formally measured. Claiming certainty a method can't support is worse than admitting a wide margin. Judges have seen both, and they trust the expert who names the limitation.

Prejudice that outweighs probative value

Federal Rule of Evidence 403 lets a judge exclude even reliable evidence if its power to inflame outweighs its power to inform. A recreation shot for maximum visual impact, rather than accuracy, is the exhibit most likely to lose this balancing test.

Late disclosure

A demonstrative disclosed after the expert report deadline gets excluded on timing alone, regardless of how sound the science is. Chapter twelve covers the calendar. Miss it, and nothing else in this book matters.

Editorializing that reads as argument, not evidence

Music, dramatic pacing, and loaded captions turn a demonstrative into a closing argument wearing an evidence exhibit's clothes. Chapter ten draws the line. Cross it, and a judge will draw it for you, in front of the jury.

Chapter 05

Recreation, Simulation, Animation, Illustration: Choosing the Right Category

Four words get used interchangeably by people who haven't had to defend one in front of a judge. Each names a different category of evidence, and each carries a different burden of proof. Pick the wrong word in your report, and you may have picked the wrong burden.

Recreation: staging the real event again

A recreation stages the actual event again, using real people, objects, and conditions. It claims to show what happened by physically doing it again. Because it claims the most, it has to prove the most: substantially similar circumstances, documented and defensible, covered in the next chapter.

Simulation: modeling the event from data

A simulation builds a computer model from data rather than staging the event physically. Its persuasive power comes from the data and the validated software behind it, not from resembling the real scene visually. A simulation lives or dies on whether the underlying model has been tested and whether its error rate is known.

Animation: depicting an opinion that already exists

An animation depicts an expert's opinion about what happened, without independently asserting that the depiction is itself proof. Properly limited, with the right jury instruction, an animation can clear a lower bar than a recreation or simulation, because the jury is told to weigh the underlying opinion, not the video, as the evidence.

Illustration: the lowest bar, and the least persuasive

An illustration is a static or simple diagram that aids testimony without depicting the event happening. A trajectory line drawn over a scene photograph is an illustration. It carries the least persuasive weight and the lowest admissibility burden, and it's often the right tool when the underlying opinion can't support anything stronger.

Chapter 06

Substantially Similar Circumstances: Building the Foundation Before You Film

Courts don't require a recreation to be identical to the real event. They require it to be substantially similar in every respect that could change the outcome. The word doing the work in that sentence is material, not identical, and knowing the difference is what makes a recreation admissible instead of theater.

What substantially similar actually means

The standard doesn't ask whether every detail matches. It asks whether the details that could plausibly change the physical outcome match. A recreation shot in different weather than the incident isn't automatically inadmissible. Weather has to matter to the physics in question before it matters to the court.

The variables that matter most

Distance, lighting, and material condition usually matter. The exact make, model, and wear state of a rope, firearm, or rigging component usually matters. The color of the shirt a stand-in wears usually doesn't. Build a list specific to the physics of this case, not a generic list borrowed from a different discipline.

When you can't match a variable

Sometimes the original component is destroyed, degraded, or unavailable. Three options exist: substitute a documented equivalent and disclose the substitution, control for the variable mathematically and disclose the adjustment, or don't attempt the recreation and move to simulation or animation instead. Silence about the mismatch is the only option that gets you excluded.

Documenting the match before you roll camera

Build a side-by-side comparison sheet, incident conditions against recreation conditions, before the shoot, not after. It becomes the foundation testimony that gets the recreation into evidence, and it's far more persuasive built in advance than reconstructed after opposing counsel asks why it doesn't exist.

Chapter 07

The Underlying Data: Ballistics, Photogrammetry, Motion Capture, and Sensor Logs

A demonstrative is only as strong as the data feeding it. Beautiful video built on thin data is still thin data. This chapter inventories the data sources that hold up, and what each one needs to survive cross-examination.

Ballistics data and drop-test protocols

A firearm drop-fire recreation stands or falls on the drop-test protocol behind it: drop height, orientation, surface, number of trials, and the failure rate across those trials. One successful drop proves nothing. A documented protocol run enough times to state a rate proves something a jury can weigh.

Photogrammetry and scene measurement

Photogrammetry converts photographs into measured three-dimensional data, and it's only as accurate as its reference points. Log the calibration objects, the camera positions, and the software's stated margin of error. A measurement without a stated margin of error isn't a measurement, it's a guess with a decimal point.

Motion capture and biomechanical data

Motion capture data reconstructing a fall, a stunt, or a struggle needs a stand-in whose relevant measurements, height, weight, reach, are matched and documented to the person involved. Unmatched biomechanics produce a technically smooth animation that an opposing expert will dismantle in one sentence.

Sensor logs, event data recorders, and timestamps

Vehicle event data recorders, rigging load cells, and body-worn sensors generate their own chain of custody problem: the data has to be extracted, preserved, and time-synchronized to the footage without alteration. Document the extraction method with the same rigor as the footage itself.

The chain from raw data to rendered frame

Every step between raw data and finished video is a step opposing counsel can ask you to explain. Keep a version history: what software, what settings, what manual adjustments, and who made them. An unexplainable gap in that chain is worth more to the other side than a weak data point.

Chapter 08

Methodology and Error Rate: Writing the Record Daubert Actually Reads

Judges rule on Daubert motions from the paper record most of the time, long before they watch the video. The methodology section of your report is doing more work than the video is. Write it first.

Write the method before you shoot, not after

A methodology written after the fact, to justify footage that already exists, reads like exactly what it's. Write the protocol first, get it reviewed, then shoot to the protocol. The report should describe what you planned to do and then confirm you did it, not reverse-engineer a plan from what happened to come out of the camera.

Stating a known or potential error rate honestly

If a formal error rate has been established for your method in the literature, cite it. If it hasn't, say so plainly, and explain the steps you took to control for error anyway: repeated trials, independent review, matched conditions. Judges don't expect certainty. They expect honesty about the absence of it.

Peer review and publication, and what to do when your method has none

Peer review doesn't require a journal article. A methodology reviewed by another qualified expert before trial, documented in writing, counts. If no publication exists for a novel method, build the closest available substitute: an independent review, cited industry protocols, or testimony from others who use the same method in their own practice.

Standards controlling the technique's operation

Where an industry standard exists, an ASTM International standard (materials and testing), a Society of Automotive Engineers standard, a National Fire Protection Association code, or a manufacturer protocol, cite the specific standard and confirm you followed it. Where none exists, say that plainly rather than implying one does. Verify every citation against the current version before it goes in the report. Standards get revised, and citing a superseded version undermines the entire section.

Chapter 09

Chain of Custody, Metadata, and On-Set Production Discipline

A demonstrative can be scientifically sound and still get kept out because nobody can say, under oath, exactly where the footage has been since the camera stopped rolling. Authenticity is a paperwork problem before it's a legal one.

What metadata a court will ask for

Timestamp, camera and lens identification, location data where available, and file hash values at the point of capture. Strip none of it during editing. If your workflow requires converting formats, preserve the original file untouched alongside the working copy, and be ready to produce both.

Chain of custody from card to courtroom

Log every hand the footage passes through: who pulled the card, who made the first copy, where the original is stored, who had editing access. A gap of even a few hours with no name attached is a gap opposing counsel will fill with a theory of their own.

Equipment logs and calibration records

Cameras drift, sensors drift, load cells drift. Log the calibration date and method for any measuring equipment used in the shoot, not just the cameras. A calibration record six months stale is an easy target, and an easy target is exactly what a motion to exclude is looking for.

The production log that saves your foundation testimony

A same-day production log, who was present, what equipment was used, what conditions were recorded, what takes were kept and why, turns a foundation question on the stand from a memory test into a document review. Build the habit on every shoot, not only the ones you expect to be challenged.

Chapter 10

Editing Without Editorializing: Where Persuasion Ends and Prejudice Begins

Federal Rule of Evidence 403 lets a judge keep out evidence whose power to inflame outweighs its power to inform, even when the evidence is entirely reliable. Every editing choice adds to your side of that scale or the other side's.

Cuts that inform versus cuts that argue

A cut that removes dead time between relevant moments informs. A cut that removes the one frame contradicting your theory argues. The test isn't whether you cut, everyone cuts, it's whether a viewer shown the uncut footage would say your version is fair.

Music, slow motion, and color grading: the line

Music has no evidentiary function. It exists to produce an emotional response, and courts know it. Slow motion can be legitimate when the real event happened too fast to observe at normal speed, and illegitimate when its only function is to linger on the worst moment. Color grading that shifts a scene toward menace or sympathy does the same work as a tone of voice, and a video doesn't get to have a tone of voice.

Captions and callouts: labeling fact versus asserting conclusion

A caption reading "fourteen feet" labels a measurement. A caption reading "the dangerous condition" asserts your conclusion inside the exhibit itself, where the jury can't weigh it as argument because it looks like evidence. Label what's measured. Argue in your brief and your closing, not inside the exhibit.

The version you show the judge before the version you show the jury

Build a clean version, free of any element opposing counsel could plausibly call prejudicial, and be ready to explain every element you kept as necessary to the demonstration itself. The version that survives a judge's private viewing in chambers is the version built to inform, not to persuade by itself.

Chapter 11

The Expert Declaration and Report: Writing the Words That Defend the Video

The video doesn't defend itself. The words around it do: the report, the declaration, the disclosure. Written well, they make the Daubert motion harder to write than it's to lose.

Describing methodology so it defends itself

Write the methodology section as if the judge will read it with no other context: what you did, in what order, measured against what standard, with what result. A methodology section that requires the video to make sense has already failed. The words have to stand on their own.

Materials considered, and why the list matters more than it seems to

List every document, photograph, dataset, and prior statement you reviewed before forming the opinion the video depicts. An incomplete list is an invitation for opposing counsel to ask what you left out, and why, in front of the jury.

Anticipating the Daubert challenge before it's filed

Read your own report as opposing counsel will: looking for the weakest of the five factors, the thinnest citation, the assumption stated without support. Fix what you find. A report written to survive a challenge that never comes is still a stronger report. A report that assumes no challenge is coming rarely survives the one that does.

What to attach, and what to reference

Attach the underlying data, the protocol, and the calibration records. Reference, rather than attach, material already in the record elsewhere, and cite it precisely enough that anyone could find it in thirty seconds. A report cluttered with everything is as hard to use as a report missing something.

Chapter 12

Disclosure Timing and Federal Rule of Civil Procedure 26: Don't Let Procedure Kill Good Evidence

A brilliant demonstrative disclosed one day late can be excluded with a one-line order. Federal Rule of Civil Procedure 26 sets the calendar this entire book depends on, and no amount of methodology fixes a missed deadline.

The disclosure deadline, and why demonstratives get missed

Federal Rule of Civil Procedure 26(a)(2) requires expert reports, and the exhibits that support them, by the court's scheduling order deadline. Demonstratives get missed because they're often finished last, after the report, as a kind of victory lap. Build the video on the report's timeline, not after it.

Supplementation under Federal Rule of Civil Procedure 26(e)

A demonstrative refined after initial disclosure is a supplementation, not a new disclosure, only if the underlying opinion hasn't changed. Refining the edit is safe. Changing the underlying methodology after the deadline isn't, and courts can tell the difference by asking one question: does the new version prove something the report didn't already say.

Rebuttal disclosure timing

A demonstrative built specifically to answer the other side's expert follows the rebuttal deadline, which is typically shorter than the initial disclosure window. Confirm the rebuttal deadline the day the opposing report arrives, not the week the trial date approaches.

State court variations

State court disclosure rules can differ meaningfully from the federal rule, in both deadline length and required content. Confirm the applicable rule for each jurisdiction at the start of the engagement, and don't assume a federal timeline transfers to a state court matter on the same docket.

Chapter 13

Answering the Daubert Motion: Structuring the Rebuttal Affidavit

A Daubert motion isn't a request for you to explain everything you did. It's an argument that one or two specific things you did are unreliable. Answer the argument that was actually made, not the one you were afraid of.

Reading the motion for its real target

Strip the motion's rhetoric and find the specific factual claim underneath: wrong drop height, no error rate, dissimilar conditions. Most motions attack one or two of the five Daubert factors, not all five. Answer precisely what was attacked, at length, and touch the rest briefly. A scattered response to a focused attack reads as evasion.

Structuring the affidavit factor by factor

Organize the rebuttal affidavit around the Daubert factors the motion actually raised, in the order the motion raised them, so the judge can check each answer against each accusation without hunting for it. A rebuttal that reorganizes around your own preferred narrative makes the judge do work they shouldn't have to do, and judges remember which side made their job harder.

When to request a hearing

Request a live Daubert hearing when the dispute turns on something better shown than described: a demonstration of the method itself, live testimony the judge can question directly, or a side-by-side comparison a written brief can't convey. Skip the hearing when the dispute is purely legal, such as which jurisdiction's standard applies, since live testimony adds nothing to a question of law.

What the case law asks the judge to weigh

Federal Rule of Evidence 702 asks the judge to weigh reliability, not correctness. You don't have to prove your opinion is right at this stage, only that it was reached through a reliable method, reliably applied. Keep the rebuttal focused on process, not on relitigating the underlying facts of the case.

Chapter 14

Direct Examination: Laying Foundation for the Video, Live

Everything in the report becomes real to the jury only when you say it out loud, under oath, in an order the judge can follow. Foundation testimony is a script. Write it before you need it.

The foundation questions, in order

Qualify the witness, establish the methodology, connect the methodology to the specific exhibit, and only then move to admit it. Skipping ahead to the exhibit before the methodology is established invites an objection that stops your momentum at the worst possible moment.

Authenticating the recreation live

Walk the jury through the comparison sheet from chapter six before you play the video, not after: here's what was matched, here's what couldn't be matched and why, here's how it was controlled for. An audience that has already seen the homework trusts the video more than an audience seeing the video cold.

Introducing the exhibit without over-arguing it

State what the exhibit shows in one plain sentence, offer it into evidence, and stop. The video does the persuading. Testimony that argues on top of the video reads as a lack of confidence in the video itself.

Handling the judge's questions from the bench

A judge who interrupts with a question is giving you information: this is the part of the foundation they aren't yet satisfied with. Answer that specific question directly, then return to the script. Don't treat a bench question as an attack to survive. Treat it as the one thing standing between the exhibit and admission.

Chapter 15

Cross-Examination: Defending the Video When Opposing Counsel Attacks the Method

Cross-examination on a demonstrative follows a predictable order: attack the conditions, attack the data, attack the edit, attack the expert. Knowing the order in advance is most of the preparation.

The standard attacks, and the order they come in

Expect, in sequence: a challenge to whether conditions were truly similar, a challenge to the data or its error rate, a challenge to editing choices, and finally a challenge to your qualifications or objectivity. Prepare each answer separately. An answer prepared for the wrong attack sounds rehearsed and wrong at the same time.

Conceding the small point to keep the big one

Opposing counsel will find a genuine limitation. Concede it plainly and immediately: yes, that variable wasn't identical, here's why it doesn't change the result. Fighting an obviously true small point costs more credibility than the point itself is worth, and it signals to the jury that you'll fight anything, true or not.

When to say I don't know

A limitation you didn't anticipate deserves an honest answer, not an improvised one. Improvised answers under cross are where careers end. Say what you know, say plainly what you don't, and let redirect and your report carry the rest.

Redirect: repairing what cross actually damaged, not everything

Redirect exists to repair specific damage, not to relitigate the whole exhibit. Identify the one or two points cross actually landed, address those, and stop. A redirect that reopens everything reminds the jury of every weak point cross found, including the ones that landed softly.

Chapter 16

The Settlement Lever: Showing Opposing Counsel the Exposure Before Trial

Trial is the most expensive and least likely outcome for almost every case this book applies to. A demonstrative built to survive Daubert is also, without any extra work, the exhibit most likely to end the case before trial starts.

Why a finished demonstrative changes a claims adjuster's math

An adjuster values a case on the range of likely trial outcomes, discounted by the odds of losing key evidence along the way. A demonstrative that has already survived, or is clearly built to survive, a Daubert challenge removes one of the biggest discounts from that math. The number goes up because the risk went down, for the other side.

When to show it, and when to hold it

Showing the finished exhibit early, before a mediation or a settlement conference, can move the number without ever going to trial. Holding it can preserve surprise value for trial or for a later, more critical motion. The choice belongs to retaining counsel, but the choice is only real if the exhibit is finished and defensible well before the moment arrives to make it.

Mediation and the demonstrative that ends the day

A mediator's private caucus with the other side is where a strong demonstrative often does its most concentrated work. The mediator carries your exhibit's implications into a room you aren't in, translated into the language of exposure and risk that moves an insurance carrier.

What happens when the other side has already seen it and settles anyway

The exhibits that end cases before trial rarely get a formal ruling on the record. They get a phone call, a revised number, and a case that closes without ever testing the video in front of a jury. That outcome is the actual goal of everything in this book. A demonstrative that never gets watched by twelve strangers, because it already did its job in a conference room, isn't a wasted effort. It's the whole point.

Chapter 17

The Ethical Standard That Keeps the Video, and the Witness, Credible

Every technique in this book can build a video that persuades without being honest. The difference between advocacy and misrepresentation isn't a legal line, it's a personal one, and it's the only line that protects every other case you'll ever testify on.

The test before you build anything

Before building any demonstrative, ask whether it would survive being built by the other side's expert from the same underlying data. If the honest answer is no, the problem isn't the editing, it's the underlying opinion, and no amount of production value fixes that.

Where advocacy ends and misrepresentation begins

Advocacy chooses which true thing to emphasize. Misrepresentation asserts something the data doesn't support. A demonstrative that omits a fact you weren't asked about is advocacy. A demonstrative that omits a fact that changes the outcome is misrepresentation, whether or not anyone asked.

What happens to your other cases when one video crosses the line

A demonstrative found to be misleading in one case follows an expert into every case after it. Opposing counsel in an unrelated matter, years later, will find the ruling and read it to a new jury. A trial record built over decades doesn't reset because of one shortcut. It only takes one.

Disclosing limitations instead of hiding them

The instinct under pressure is to hide the weak point and hope no one finds it. The stronger move, and the one that holds up over a career instead of one case, is to disclose the limitation before anyone asks and explain why it doesn't change the conclusion. A limitation you disclosed voluntarily helps you. A limitation opposing counsel discovered on cross helps them.