1999 Michigan State Police Patrol Vehicle Test

The Volvo sedan is the interesting shape here. It is slower off the line than the Police Interceptor to 60 mph — 8.86 against 8.55 — and then beats it to 100 by four full seconds. A turbocharged 2.3-liter Volvo with 236 horsepower against a 4.6-liter V8 with 215.

Four drivers — Sergeants Stevens, Clark, Rothermel and Pokora — ran each car over the 1.635-mile road course at Michigan International Speedway, sixteen timed laps per vehicle.

A note this data settles. The report’s methodology section says the score is the average of the fastest of at least nine timed laps, while its tables have long carried a footnote saying the fastest twelve — a contradiction visible in the bulletins going back years. With the individual laps now in hand, the arithmetic answers it: every published average in this table is the mean of that vehicle’s fastest twelve laps, exact to the hundredth in all ten cases. The footnote is right; the methodology paragraph is loose.

The human variable is also visible. Across every single vehicle, one driver is consistently the slowest and another consistently among the quickest — the spread between them running from about a second and a half in the Volvos to nearly three seconds in the Tahoe. Which is exactly why the program averages twelve laps from at least three drivers rather than printing a best lap.

The 1999 book is unusually candid about its own conduct, and every item below is disclosed in the report’s preface rather than dug out of the data.

The Volvo sedan was retested, and scored on the worse run. The manufacturer felt the S–70 was not running correctly; the test driver and observer agreed. Volvo asked for a retest, it was granted, and the second set of results came out worse than the first. Michigan used the second set for the score anyway.

The Camaro ate its brakes. The pads and rotors on the automatic Camaro required replacement partway through the dynamics session, after two of the four drivers had completed their practice and timed runs.

The natural-gas car ran out of fuel. The CNG Police Interceptor needed refueling after three drivers had finished at the speedway — unsurprising for a car carrying four tanks holding the equivalent of ten gallons against the gasoline car’s nineteen.

Two Fords arrived with green brakes. The Explorer’s and Expedition’s brakes had not been burnished by the manufacturer before testing, so the report advises that their Phase II figures — taken after the first six stops had bedded things in — are the more representative numbers.

Only one Camaro was braked. Because the two cars have identical braking systems and nearly identical test weight, only the automatic was submitted for brake testing. This also explains the missing Camaro brake row in the 1997 and 1998 bulletins, which never gave a reason.

Other conditions worth recording: every vehicle ran with a clean roof, no lightbar and no A-pillar spotlights, and the report says plainly that once the emergency equipment is installed, performance will be somewhat lower than reported. The Jeeps were driven in “3” rather than overdrive; the 6-speed Camaro made its top-speed run in fifth. The Volvos’ traction control was switched off for dynamics and their transmissions were set to “S” for Sport.

The full report prints the scoring method in seven steps, and it is worth understanding because it explains every “winner” on every page of this archive.

Raw scores are recorded in each of the six categories in their own units — seconds, ft/s², miles per hour, points, miles per gallon. In each category the best-scoring vehicle becomes the benchmark and receives a deviation factor of zero. Every other vehicle’s deviation factor is the absolute difference between its score and the best score, divided by the best score. That factor is multiplied by the category weight — dynamics 30, acceleration 20, braking 20, top speed 15, ergonomics and communications 10, fuel economy 5 — to give a weighted category score, and the six are added together into a total weighted score. Lower is better: a perfect vehicle, best in all six categories, would score zero.

Then comes the part that makes these books a purchasing instrument rather than a magazine test. Michigan takes one percent of the lowest bid price received as its bid adjustment figure, multiplies each vehicle’s total weighted score by that figure, and adds the result to that vehicle’s actual bid. The vehicle with the lowest adjusted bid price wins — and the state then pays the real bid price, not the adjusted one. Performance doesn’t win the contract on its own; it buys a discount against the sticker.

The report illustrates this with worked examples using hypothetical figures rather than the year’s real bids, so no bid prices are quoted here. The 1999 winner and the actual bid table are not in the published report.