Hungry yet?

It is always good to figure out how you die....



Crash Course

The science of reconstructing car accidents.
By Jascha Hoffman
ANDREW RICH IS NOT YOUR TYPICAL POLICE OFFICER, and not because his thick, Jersey baritone and slick double-breasted suits give him the appearance of a capo in a crime family. What really makes Rich stand out from his colleagues in law enforcement is his erudition.

"Good morning, my name is Andy, and I'm a geek," he announced recently at a lecture at the IBM Watson Research Center in Yorktown Heights, N.Y. As if to prove the point, he opened his suit jacket to reveal an extra-large pocket protector.

After 11 years as a highway patrolman, Rich became a detective with the Fatal Accident Investigation Unit of the Bergen County Prosecutor's Office in 1999. As an accident reconstructionist, Rich spends his days determining how car accidents happened, gathering all the hard evidence available, and using the laws of physics to run the crash in reverse. On this morning, however, Rich was giving a talk at a seminar organized by the New York Statewide Traffic Accident Reconstruction Society, or NYSTARS to its members.

Like the recon profession itself, those who attend the quarterly NYSTARS seminars comprise two types of professionals: police detectives like Rich, who do reconstruction for their departments, and mechanical engineers, who typically work for private reconstruction firms that contract their services out to insurance companies and lawyers.

The detectives, most of them men and many of them mustachioed, tend to have developed something like a sixth sense for how an accident occurred, having seen hundreds of them while on duty. The engineers prefer calculus. The two groups tend not to mix much outside events like this one. For cops, the work is a distinction, albeit a nerdy one. Most engineering majors, on the other hand, don't dream of spending their careers reconstructing gory car wrecks. Rich is one of the few in the business who belongs to both camps. Back when he was on the highway patrol, he worked nights and took engineering courses at Bergen Community College during the day. Now he works days, and is in the process of earning a degree in mechanical engineering from the New Jersey Institute of Technology.

Rich wears his erudition lightly—he refers to himself as Professor Protractor. He treats his work with a clinical detachment, referring to pedestrians as "it." Rich's first topic of the day was three-point airborne trajectory analysis, which is used when an accident causes a car to make a brief (and invariably unsuccessful) attempt at flight. Rich flashed a slide of a severely wrecked sedan spattered with blood and battery acid. According to the police who were chasing the driver of this car, Rich explained, "our young hero" jumped a guardrail and hurtled off an overpass, taking down two trees in his path. Miraculously, he survived. His girlfriend in the passenger seat did not.

In a case like this, the recon expert's task is to confirm that the driver was speeding. It's possible to estimate how fast the car was going when it hit the ground by looking at the damage to the car. But to accurately estimate the driver's speed when he left the ground—a crucial question in court—you have to know the takeoff angle.

Where a car has taken off and landed isn't hard to determine—the former is where the skid marks end; the latter is where the crumpled wreck was found—but a third point, necessary for three-point analysis, can be elusive. If, however, a reconstructionist is lucky enough to find a snapped tree branch or, as in this case, some scrapes on the side of an adjacent overpass, he can apply three-point analysis. From there, it's just elementary calculus, Rich explained, but decided not to test the audience's patience with a complete derivation of the formulas involved. "Not if I don't want to get shot," he reasoned. The cops in the room did not object.

Reconstructionists usually rely for their data on police investigators who take photos, measurements, and samples of all the evidence at a crash scene: spattered blood, automotive debris, clothing fibers. With these raw materials, reconstructionists get to work, running lab tests, calculating approach angles, and—a somewhat recent development—running computer simulations to check their work.

With the right physical evidence, reconstructionists can determine how fast a car was going and whether the driver was braking or swerving in the moments before impact. Sometimes, they are called on to make grimmer determinations—in a truly bad wreck, it can be difficult to tell who was driving the car. In some crashes, the force of impact does all the work: The person with the brake pedal's pattern etched into the sole of his shoe was probably the driver.

Other incidents are trickier to decipher. In a case this spring, a reconstructionist testified that a woman was driving when she and a Hartford businessman careered out of control on Connecticut's Route 9. The attorney for the woman, on trial for manslaughter, claimed that it was the man who was driving. The lawyer suggested that the woman's injuries, including a ruptured left breast implant, were consistent not with driving but with performing oral sex on the late businessman when he lost control of the car.

AFTER A SHORT BREAK, Rich picked up a new topic, explaining why hit-and-run cases involving pedestrians are so terrible: They take forever to reconstruct. Arriving on the scene, Rich said he can often identify where the accident happened, how far the pedestrian was thrown, and how far he slid. He can then use formulas honed from thousands of crash tests to estimate the driver's speed within a close range. To use these formulas, however, he must first determine what kind of accident occurred, and that means looking at the injuries to the victim.

Typically there are two kinds of injuries, those from the initial impact, and the ones from hitting and sliding on the asphalt, known as "road rash." To illustrate the different types of impact a pedestrian can suffer, Rich cued up a series of video clips on his laptop. The first one showed a well-dressed man with a briefcase in each hand caught crossing a busy Manhattan street. Suddenly, a white minivan blindsided him, causing a "fender vault" that tossed the man three feet into the air, still holding one briefcase. A taxi approaching from the opposite direction then launched him into a textbook "roof vault," sending his remaining briefcase flying and hurling him headfirst onto the pavement. This was not a walk-away accident.

The lecture hall stirred with uneasiness. Rich ignored his audience's squeamishness and moved on to a clip of a man carrying a pizza box who, propelled into a full forward flip by a speeding coupe, landed twitching on the pavement 20 feet ahead. At this point, the crowd could not control its discomfort. "Is this real?" someone asked anxiously from the back. "It's an Australian public safety video," Rich barked in response. A cop asked whether the first one was a simulation as well. Rich ignored the question.

Having forced his audience to witness these gruesome happenings, Rich continued by explaining that it is a central tenet of the accident reconstruction business that eyewitnesses are more or less useless. Drivers lie to protect themselves, and given the speed of most accidents, even the memories of honest souls can't be trusted. Witnesses will swear that a pedestrian was launched 200 yards by a collision—a feat that could only be accomplished if the offending vehicle were doing something like Mach 3. They will claim a black car was white, or a white car was red. Once, three cops commenting on the same accident told Rich respectively that the conditions were wet, dry, and icy.

In the absence of reliable witnesses or conclusive injuries, Rich explained that it can take a trained eye to identify what's not an accident. He recalled arriving at the scene of what looked like a typical hit and run. Since the force of an accident can scatter personal effects, they are not always a good way to determine the point of impact, but some rules tend to hold. While eyeglasses drop straight down, shoes and hats "go into orbit like the space shuttle." In this case, Rich found a body in a pool of blood with a yarmulke suspiciously close by. ("Maybe he used a lot of bobby pins?" volunteered one officer. "Just one," Rich shot back.) Further examination showed no denim fibers spread on the asphalt from the inevitable slide, which led Rich to conclude that this was no car accident. When fresh tar from the deceased's boots matched tar from the roof of the building across the street, Rich's suspicions were confirmed. Quoting an old axiom of police work, he said, "Things are not always as they appear."

Around 1 p.m., Rich wrapped up the morning's events by discussing how to decipher various medical clues, which he described with the help of a horrific series of slides depicting protruding femurs and sheared skin. "Everyone hungry now?" he asked when he had finished. Everyone was. The crowd filed out impatiently for lunch, refueling for the afternoon session, a three-hour lecture on collisions with poles.