Skip to main content
CrossCoach™
Sign inRequest access
Guided readings

The reports they’ll quote at you, read before they do.

12 short, plain-English walk-throughs of the landmark reports and studies behind forensic testimony, each grounded in the source document itself — with the questions you’ll face about it built in.

A single small figure standing alone at the foot of a vast marble courthouse rotunda, dwarfed by towering grey stone columns rising into a coffered dome under flat overcast daylight.
Reading · Foundations14 min

What the NAS Report Actually Says About You

The report that rewired how courts scrutinise expert evidence, in seven short readings: whether your method holds up, the bias that can move it, and the words that get you impeached on the stand.

Begin reading
An empty wooden witness box under a single hard shaft of light: a vacant chair, the polished timber rail, and a microphone on a gooseneck waiting, with the darkened public gallery behind.
Reading · The cross-examination15 min

The Cross-Examiner's Playbook

The lawyers’ field guide to attacking forensic evidence, read as your preparation. It covers validation, bias, error rates, and the words that get you impeached.

Begin reading
A bright clinical test bench where a long rack of identical labelled test samples is being run through a stainless steel measuring instrument, a controlled validation experiment in progress.
Reading · Foundations15 min

The Two Questions PCAST Will Ask Your Method

The 2016 White House science report that gave courts a two-part test for forensic validity, set out across seven short sections: foundational validity, black-box studies, and the words that get you impeached.

Begin reading
A vast grid of hundreds of near-identical grey paper comparison slips, every one matching, with a single slip glowing stark red, the lone error exposed in a sea of correct decisions.
Reading · Fingerprints15 min

What's Your Error Rate?

The first large-scale black-box study of latent print examiners. The real false-positive and false-negative numbers, and how to state your error rate on the stand.

Begin reading
A precision analogue gauge whose slender needle rests dead centre at the exact tipping point between a teal-lit half of the dial and an amber-lit half.
Reading · Decisions13 min

“Inconclusive” Is Not the Safe Answer

The trickiest word in your testimony, in everyday terms. Why two equally skilled examiners disagree, how “inconclusive” hides errors, and how it sets your discipline’s error rate.

Begin reading
A close-up in cool near-darkness: one anonymous person leans in to whisper, a hand cupped to their mouth, directly into the ear of a second person who stays calm and intent. A single warm beam catches the cupped hand and the listener's ear while both faces stay in shadow.
Reading · Cognitive bias16 min

The Same Print, A Different Answer

How context can flip an expert’s own call on the same print: the study that showed it, the wider evidence that it isn’t a fluke or confined to one discipline, the six fallacies that keep examiners exposed, and how to answer “could you have been influenced?” on the stand.

Begin reading
An extreme macro view inside a hard drive in the dark: the mirror-black platter fills the frame, the read-write head poised a hair above it, the whole surface lit cold electric-blue with a single thread of green light tracing across it.
Reading · Digital forensics16 min

Digital Forensics: The Evidence That Looks Like Maths

Digital evidence reads like arithmetic. Six cross-examination moves, from the Horizon presumption to hashes, bias and timestamps, that show it is not.

Begin reading
A dense overhead surveillance grid of many near-identical anonymous human faces in desaturated grey-green, each face caught from above in low-resolution CCTV, all of them slightly soft and indistinct, the whole wall reading as one watched, interchangeable crowd.
Reading · Facial comparison16 min

Faces from Images: How Good Is the Match, Really?

Matching two images of a stranger is the hard direction, and the method has no measured error rate. What the science shows, and how to testify to it without overstating the match.

Begin reading
A lone studio microphone standing in near-total darkness, the air around it alive with a glowing violet-to-magenta waveform that radiates outward like a voice made visible, the only light in the frame.
Reading · Speaker comparison15 min

Voice on Trial: Forensic Speaker Comparison

The voiceprint myth, the likelihood-ratio paradigm, the Cllr you must measure on case-realistic recordings, and how to testify to a voice match without overstating it.

Begin reading
A lone anonymous figure walks across a vast empty car park at midnight, small in deep darkness, crossing through the single pale pool of light thrown by one overhead lamp, the rest of the lot lost in cold blue-black.
Reading · Gait & body15 min

Gait, Body & Clothing Comparison: Identifying a Person from an Image

Walk, build, clothing and height, all sold to a jury as a "match." Four feature comparisons, no measured error rate, and the six moves a cross-examiner uses to pull them apart.

Begin reading
A close human portrait filling a near-black frame, split down the middle by a hard seam: the left half a crisp, cleanly captured face, the right half tearing apart into corrupted pixel blocks and a magenta-and-cyan RGB-split fringe, as if the same face cannot decide whether it is a real photograph or a synthetic one.
Reading · Image authentication16 min

Is the Image Even Real?

Container metadata, PRNU camera fingerprints, deepfake detectors and C2PA credentials, and the one honest answer to the new question every exhibit now faces: is the image even real.

Begin reading
A single sworn written statement on heavy cream paper lying on the worn honey-toned oak ledge of an old witness box, a brass-buttoned panel rising behind it, lit by a soft warm pool of low lamplight that falls off into the shadowed courtroom beyond.
Reading · Foundations16 min

How Conclusions Go Wrong: The Shared Backbone

The spine of CrossCoach: the analysis can be sound and the sentence to the jury still false. The logic of a conclusion, the three ways it breaks, and the five questions every standard generates.

Begin reading