10 Contenders, 1 Winner: America's Best-Handling Car Elimination Rounds

[mikeyb]

The Contest: Crown the best-handling production car in America

The Contestants: Cars built and certified to Yankeeland standards, and available for sale to regular and/or working-rich folks at dealerships in 2007.

The Criteria: The supreme grand champion handler may be a car that generates big hairy test numbers on the skidpad-or not. It may snake through a slalom course in record time-or not. It may go around a racetrack faster than any other-or not. It will do all of the above with exceptional finesse while eliciting broad smiles on its driver's face. And our selection will likely provoke feverish haranguing in the blogosphere.

Round One Eliminations

Our first cut at the 250 or so available nameplates involved a morning of caffeinated debate over donut holes around the conference table at One Motor Trend Tower. Then, taking a cue from the sports world, which makes a multi-billion dollar business of picking winners, we divided the good cars into four divisions:



<LIST>Front-engine/Front-drive
Front-engine/rear-drive
Front-engine/all-wheel-drive
Mid- or rear-engine/rear- or all-wheel drive


When the advocating, name-calling, and carbo-loading was over, we had eight finalists in each "division" ready to advance.

For Round 2, we departed from the sports-world's one-on-one tournament style and staged four eight-way virtual comparison tests. By combing past road tests and drive reports to jog our collective driving experience memory banks for clues to each car's handling vices and virtues, we advanced only the very best-balanced, most sublime twisty-road dance-partners to the finals. Here's how they shook out.

Front-engine/Front-drive:

FINALIST: Honda Civic Si-This 2006 Car of the Year winner charmed our judges with its slick, smooth power delivery, feather-light yet communicative steering, and seemingly custom-tailored cockpit. It manages to generate high grip numbers without ever feeling muscle-bound. Corner photographers caught every driver grinning.

Acura TSX Type S-The TSX is an emotional favorite based on its supreme balance and light-footedness. It is a joy to drive hard, scoring exceptionally well on the smile meter, less so in all objective measures. While power isn't essential to great handling, most felt this zippy sedan could use a few more horses to run with the best of the front-drive coupes and sedans.


Chevy Cobalt SS Supercharged-Chevy's sport-compact spoiler is a serious numbers-generator, scoring big on the figure-8, the skidpad, and the drag-strip too. But a tendency toward big understeer and a lack of meaningful communication through the steering wheel kept this rice-rocket wannabe from advancing to the finals.

FINALIST: MazdaSpeed3-Remember the Focus SVT we all loved? Well, this might have been it, had Ford's product plan not run off the rails. Splendid power delivery, delightful steering, and a superb ride/handling compromise combine with best-in-class braking and high-ranking numbers on the other tests to make this an easy choice to advance to the final round of competition.


Dodge Caliber SRT4-If the Chevy is all about making numbers, this Dodge is about beating those numbers, and we expect it will (test cars are not yet available). But we have grave concerns about how SRT could possibly preserve the steering feel and finesse we so treasure while the front wheels grapple with 300 horses and 260 pound-feet of torque. We hope to be proved wrong.

FINALIST: Mini Cooper S JCW*-Take the original grown-up's go-kart and make it faster and better handling, and you have yourself a shoe-in for any handling contest. Everyone quotes power-to-weight, this one also trumps most on tire contact-patch per pound. This is surely the quickest-reacting, most visceral front-drive runabout on planet Earth.


Nissan Sentra SE-R Spec V-It looks like a true re-creation of the original BMW-2002-like SE-R. The spec sheet calls for modest 200-hp thrust through a limited-slip differential to big tires, wheels, brakes, and a heavily starched suspension. But can it really be expected to out-nimble a Mini or out-run a Mazdaspeed3 with Civic Si levels of finesse?


Volkswagen GTI-There's a lot to love here-retro style cues without the retro three-legged understeer. Modern brakes, electric power steering, and a proper independent rear suspension deliver strong numbers, but in a package that somehow feels more mature, aloof and detached than the Honda, Mazda and Mini S.


*Okay this one required a teensy bending of our 2007 rule, as Works versions of the new Mini coupe are not yet available. Cooper S JCW convertibles are available in 2007, and the new Works GP coupe surely won't handle any worse than this one...

Front-Engine/All-wheel-drive:

Audi A3 3.2T Quattro-Audi's super-fun TT is sitting 2007 out, but we're on record calling the A3 a TT with a back seat and luggage room. Little of the Bauhaus bahnstormer's grippy, hard-charging charm is lost in the translation to sensible bodywork. But deep within each beats the heart of a GTI, and there's a nose-heaviness that is hard to overlook in a pure handling contest.

FINALIST: Mitsubishi Lancer Evolution IX MR-Staff opinions are split like Mac/PC or Coke/Pepsi when it comes to the Evo and the WRX STI. The current camp holds that the pricier Evo is the more sophisticated implement; more neutral in its responses with better body motion control. These attributes clearly add up to better track times and test numbers. Perhaps out of fear of the entire competition devolving into an Evo/STI shouting match, the decision was made to bring just one, and the STI got voted off the island.


Audi RS 4-Fill the nose of an Audi with a 420-horse V-8 parceling out 60-percent of its torque to the rear axle and perch it on a sophisticated pitch- and roll-controlling Dynamic Ride suspension with big brakes and 19-inch rolling stock, and you're definitely on to something. But electronics and big rubber never fully compensate for a 59-percent front weight bias, which is always noticeable.


BMW 328xi-Renamed to announce its enhanced 230-hp direct-injected 3.0-liter engine, the 328 still enjoys BMW's rear-biased (40/60 torque distribution) all-wheel-drive system, which delivers better handling than most 50/50 setups. Spec the sport-package, and it's a definite contender. But in a handling contest, a four-wheel-drive 3-series suffers by comparison with its rear-drive sibling.


Infiniti G35x-Adding all-wheel-drive to a great-handling car doesn't have to diminish its handling capabilities, but both BMW and Infiniti hold back some of the sporting goods on their "x" models. The biggest horsepower, manual transmission, best tires and wheels are reserved for the rear-drive G35 Sport, forcing the G35x to fight with one hand tied behind its back.


Mazdaspeed6-While a Subaru Legacy beat our Mazdaspeed6 in a comparison, the Mazda was heralded as a clear winner in terms of handling prowess, its programmable AWD system helping generate big numbers on the track as the chassis drew praise for its crisp turn-in and eagerness to change direction. But in the big picture, the lighter, smaller, nimbler Evo is just way better to drive.


Subaru Impreza WRX STI-This lame-duck Scoobie remains a top-pick among the autocross set, largely due to its broad range of readily available modifications. It offers a smoother ride and a bit more roll than its archrival Evo, with oversteer in reserve for experienced drivers. Also auguring against the STI is its recent comparo loss to the Mazdaspeed3. The '08 WRX is sure to up the ante substantially.


Volvo S40 T5 AWD-This Volvo shares its Ford C1 platform underpinnings with the much-loved Mazdaspeed3, but its 2.5-liter five-banger is a bit less enthusiastic, and the rest of the chassis is calmed down by a similar amount. It's a deft handler, with supple ride motions and reasonable chassis response, but by comparison with the zootier Mits'ubarus, it's bound to suffer.





</LIST>

 

[mikeyb]

Front-engine/Rear-drive:

FINALIST: BMW 335i Sport-The magic of BMW's benchmark 3-series starts with a nearly ideal 51/49 percent weight distribution. The chassis sorcerers then add a supernatural strut front suspension and one of the most communicative and intuitive steering systems on earth. Throw in a stability control system that can distinguish "Oh boy!" from "Oh s-t!" and twin-turbo I-6 power, and you have a shoe-in finalist.


BMW M5-This is one extreme sedan, from its 500-hp V-10 to its SMG tranny and aggressively balanced Bavarian chassis, "it all adds up to a machine that screams 'track car.'" That comment came with the caveat that "you'll have to work a lot harder to see it," noting that deft hand-foot coordination is required to keep the M5 from backing off the road. Then there's the frustrating iDrive fiddling required to access full power...

FINALIST: Chevy Corvette Z06-Who'd have thunk just a few years back that Chevy's blunt-instrument 'Vette could matriculate into a thoroughbred track star? But sure enough, replacing a steel chassis with a stiffer aluminum one, swapping carbon-fiber for fiberglass panels, and re-tuning the suspension to suit a 7.0-liter 505-horse engine, has made the Corvette Z06 a potential world-beater.


Dodge Viper SRT10-Dodge has wrought a near miracle with the Viper. A comparatively light roadster with a gigantic 510-hp V-10 should be borderline undrivable. But by fitting similarly ginormous tires and tuning the suspension astutely, the Viper generates big grip numbers and reasonable track performance. But its overall demeanor remains a bit brutish and blunt by comparison with the best handlers.

FINALIST: Honda S2000-Every Honda, from the lowly Fit to the forthcoming HondaJet aims for top-of-class dynamics, and most achieve it. Folks therefore expect great things from a Honda sports car, and the S2000 doesn't disappoint. It's not a bar-room numbers generator, scoring about average in most tests. But there's a magnificent precision in the steering, shifter, brakes, and throttle that foster a unique oneness of man and machine.


Mazda MX-5-The original "British roadster done right" has been a smile-generator since day one. It has never scorched skidpads or slalom courses and yet it is always nominated to handling contests. That's probably because the car truly qualifies as wearable transportation; an anatomical prosthesis for rapid, enjoyable transit. Trouble is, most everything the MX-5 can do, the S2000 can do slightly better, and with no less finesse.


Mazda RX-8-Stick a high-revving pony-keg-sized rotary engine behind the front axle of an already light car, then apply zoom-zoom chassis tuning, and you're onto the formula for a sweet handling 2+2. Like the S2000 and MX-5, it's a finesse player, not a performance-spec-generating braggart, and in the final analysis, our jury chose the S2000 to wave the finesse flag.


Nissan 350Z-This squat, short-wheelbase wide-track sportster is a screamer in the slalom and a high performer in most handling tests-especially for the price. This year horsepower and price are up. But relative to the other bargain handler in this division, the S2000, the 350Z's primary advantage is in straight-line power, and the car feels like a heavier, blunter instrument by comparison.

Mid- or Rear-engine/Rear- or all-wheel drive:

Lotus Elise-Few brands have adhered as faithfully to the aircraft-design maxim "simplicate and add more lightness" as Lotus, and the elemental aluminum Elise is the ultimate expression of this idea. Weighing less than a ton with 1.8-liter Toyota power and a track-tuned suspension the Elise is low on creature comforts, high on handling brilliance. About the only car more tightly focused on handling is its stablemate, the Exige S.

FINALIST: Lotus Exige S-To the Elise's brilliant sports-car formula, add a roof for increased structural integrity and a Lotus-developed supercharger good for 220 horsepower (which is evidently tunable to 240 or 255 with ease). The result is the ultimate Lotus (for now), a 2050-pound purpose-built track-star with a 38/62 front/rear weight distribution. Motorized transport doesn't get much more fun than this.


Bugatti Veyron 16.4-Newsflash: a $1.44 million sticker price affords the development budget for a fab chassis! This latter-day Bug boasts all the mod-cons: hydraulically adjustable suspension, manhole-cover brakes, active aero, etc. It's all needed to cope with the brutal 1001 advertised horsepower. Our thin rationale for its exclusion: at 4530 pounds, it feels kinda heavy. Real reason: it's danged hard to borrow one.


Porsche 911 Turbo-We've dubbed it the 24/7 Supercar. A 480-horsepower missile that can be driven by anyone, any time, thanks to incredibly astute chassis tuning and stability-enhancing electronically controlled all-wheel-drive. But in a purely handling-oriented contest, the Turbo trailed its more track-tuned 911 stablemate, the rear-drive GT3 in jury voting.

FINALIST: Porsche Cayman S-This is the Goldilocksian "Just right" Porsche. Its blend of just enough horsepower (295), slight rear weight balance, and sane levels of grip allow mere mortal drivers to probe the limits of adhesion at quasi-sane speeds and cause journalists to wax poetic when emerging from a stirring drive on a twisting road or a closed handling circuit. It's simply our favorite Porsche.


Ferrari F430-Ferrari got serious about delivering on the marque's performance promise with the F355. Two generations later, aluminum construction and continued refinement of the suspension geometry have further honed the performance of the V-8 Ferrari. It's a stunning piece of work for the road and-in Ferrari Challenge dress-for the track too. But a Porsche 911 GT3 outperforms the Ferrari in all our standard handling tests, and does it for $60-70,000 less, so we invited it instead.


Lamborghini Gallardo-It's as bodacious as any previous Lamborghini, but adoptive parent Audi has worked some professional engineering in under that wild skin making this mid-engine masterpiece "easily the best-handling Lambo in history." But by comparison with its archrival from Ferrari, the heavier, all-wheel-drive Lambo feels isolated and requires more work to extract similar performance.

FINALIST: Porsche GT3-Its ride-height, camber, and roll-stiffness are mechanically adjustable; its shock, engine, and traction-control tuning are electronically switchable; and there's no slop in any of the controls. It comes by its 415 horsepower the natural way, with big lungs and a lofty redline, no peaky turbos. It is the nearest thing to a Factory racer for the road. We couldn't resist taking it to the track.

- Motor Trend

<!--endclickprintinclude-->

 

[genericuser]

articles like these are why I'd love to be an automotive journalist.. I can always dream

 

[mikeyb]

And the winner is....

 

[nazgul350r]

Duh!

 

[jred321]

is that seriously the winner? it seems like they forgot the second half of the contest where the finalists competed with eachother

 

[dmitrik4]

maybe i'm misreading, but it sounds like there was no actual "competition," only a bunch of editors arguing over the "best-handling" car.

For Round 2, we departed from the sports-world's one-on-one tournament style and staged four eight-way virtual comparison tests. By combing past road tests and drive reports to jog our collective driving experience memory banks for clues to each car's handling vices and virtues, we advanced only the very best-balanced, most sublime twisty-road dance-partners to the finals.

i'd expect no less from Motor Trend.

 

[Jeph]

No there were a whole bunch of tests and stuff, i don't think the whole article got posted.

 

[ZoomZoomH]

ironic that they tested on MAZDA RACEWAY at Laguna Seca, and the RX-8, arguably one of the best handling car in the world, is absent from the contest

EDIT: oh wait it is.... PWND (flame2)

 

[mikeyb]

Here are the tests for this article.

Which is the best-handling car in the land? Car magazines have been answering that question for a half-century. Early on, tire and suspension technology was so rudimentary that auto scribes like Uncle Tom McCahill could easily sort out a field of contenders by taking each for a brisk drive around a racetrack or some nice twisty roads with no more instrumentation than a stopwatch and the seat of his pants. Today's worst-handling car is probably as agile on its well-shod feet as some of the most exotic mid-century sports cars were, and sorting out the crme-de-l'handling-crme now takes more than the simple biological assometer. As technology has advanced, vehicle-dynamics engineers have dreamed up better and better -ometers with which to quantify the myriad constituent elements of handling, and our crack technical team-especially Gizmo-Wizard Reynolds-has been itching to outfit a squadron of spry sportsters with the latest crop of g-meters, height-sensors, potentiometers, and, yes, even a real-live electronic assometer in order to once again answer The Question.


Step one in any such venture is to select the contestants. To shed some meaningful light on the topic of dynamic handling, we decided to invite top-handling cars representing the four major driveline layouts-front engine, front drive; front engine, rear drive; front engine, all-wheel drive; and mid- or rear engine. We brainstormed a list of 32 great-driving cars, our eight favorites in each category, then combed through past road-test data, reread our subjective reviews, and winnowed that list down to 10 finalists. (You'll find the full list and our reasoning behind the eliminations at motortrend.com.) We also endeavored to keep the finalists relevant to our readers by eliminating ultra-rare, unattainable exotica. Let's meet the finalists.

Our lowest-priced entrants are the remarkably agile and neutral Honda Civic Si ($21,885) and the aggressive factory-tuner hot-hatch Mazdaspeed3 GT ($24,550). The bulldog Mini Cooper S John Cooper Works GP ($31,150) and the Mitsubishi Evolution IX MR ($37,424) test cars are both 2006 models because their successors weren't yet available, and we couldn't bear to let either of these two staff favorites sit out the competition. Honda's sleek, sure-footed, and high-strung S2000 served as our entry-level rear-driver at $34,845, followed by the sublimely intuitive new BMW 335i coupe (Sport package, hold the Active Steering, for $42,995). Porsche's polished and impeccably balanced Cayman S and Lotus's uncivilized, supercharged Exige S served as our mid-engine entrants, priced at $63,300 and $65,100, respectively. At the top of the class are Chevy's lightweight fire-breather, the Corvette Z06 ($70K) and Porsche's race car for the street, the purist's naturally aspirated rear-drive 911 GT3 ($115,700). For the record, the almost-affordable Ferrari F430 was eliminated because the GT3 outperforms it in all our standard handling tests-for $60,000 less.

All cars were tested in factory-spec condition on their original-equipment tires. Some will suggest we should've shod them all in identical rubber to eliminate tires as a variable, but a manufacturer's choice of tire is such an integral element of the chassis design, that swapping tires would muddy the results as much as setting all suspensions to the same alignment specifications. Tire pressures were adjusted to the factory settings for all evaluations except racetrack hot-lapping, for which all were increased by 5 psi.

So how does one define exemplary handling? Lay persons in the auto-scribbling business often dodge the question, essentially paraphrasing Justice Potter Stewart's Ohio obscenity-case commentary stating that they "know it when they feel it," and proffering a few test-track numbers as backup. But this exercise demands more rigor. The numbers that magazines typically generate on skidpads, slalom courses, lane changes, and even our figure eight quantify only a few constituent elements of vehicle dynamics like cornering, directional response, and transitional behavior. The broader term, "handling," encompasses myriad other vehicle qualities, many of which provide the driver feedback that helps maintain control and enhances the enjoyment of the experience.

These other elements are trickier to quantify, but, with some electronic doodads and a whole lot of ingenuity, many can be measured. Others can only be articulated qualitatively. To cover the spectrum of analysis as thoroughly as possible, we rented the vast airstrips at the retired El Toro National Guard base in Orange County, California, and put all 10 cars through an intense regimen of special test procedures conceived to tease out as much objective handling data as possible. We then drove hours north to Mazda Raceway at Laguna Seca, where we again instrumented the cars and sent them out with professional race driver Max Angelelli, who knows the track well enough to bring each car up to its full potential within a handful of laps and whose experience testing and tuning cars enables him to articulate a car's dynamic strengths and weaknesses. Finally, we put our own experienced editorial staff in the drivers' seats for hours of comparative driving on hilly, twisty, Central California wine-country roads and solicited their feedback.

In all three venues, the qualities we're seeking to evaluate include:


<LIST>Ultimate grip and cornering power: The familiar skidpad g-force performance.
Transitional behavior: How the car behaves when turning the steering wheel, especially while accelerating or braking.
Path accuracy and directional stability: How well the car goes where it's pointed and how faithfully it tracks straight ahead.
Steering and brake feel: How well the controls convey what's happening where the rubber meets road.
Chassis composure and control: How confidently the suspension absorbs and reacts to bumps, dips, and changes in the road surface.
Cockpit ergonomics: Control positioning and function, and the seat comfort and support contribute to good car control.</LIST>

So with that preamble, let's hit the tracks and the road in search of America's best-handling car.

The Basics
A primer on how to get those feel-good chemicals pumping

Whip a car into a sweetly banked turn, zigzag through a necklace of S-bends, clip an apex, and cock the tail out a few degrees with the throttle. It's a good day, you muse, grinning. Feel-good endorphins fire-hosing your brain will tend to do that. There are few better sources of mood-enhancing chemicals than a great chassis performing the elaborate dance of vehicle dynamics.

What differentiates the automotive equivalent of Balanchine's choreography from Billy Ray Cyrus doing the cha-cha on "Dancing with the Stars" are subtle behaviors that aren't easy to discern subjectively. They need to be considered by objective instruments able to detect events scant hundredths of a second apart and sense minute accelerations and movements. And it takes time and patience, which is why we decamped for two days of elaborate vehicle testing that would spotlight each of our contestants' basest handling ingredients.

In the following pages we'll take a close, instrumented look at each of our best-handling gems, inspecting them facet by facet. On-center sensitivity? Step-steer time? Ride-quality analysis? Never heard of such things? Settle back and read on for a vehicle-dynamics meet-and-greet session that'll leave you on a first-name basis with the nitty-gritty of how cars handle. It's unlikely any automotive publication has ever covered a car with as many sensors and wires (often, we were recording 11 data channels at once) or has conducted as methodical a test regimen. When we left (in the dark) 25 hours later, we'd recorded 420 megabytes of data to sort through and tease-out meaning from. Consider the material on the next few pages in sequence, as we're approaching our understanding of each car in a building-block way. The goal? Move on to Laguna Seca with a firm grip on each car's basics.

1. On-Center Feel

In our REM-sleep fantasies, "handling" means one thing: flat-out, neck-bending, tire-marbling cornering. Ah, the smell of sizzling rubber and carbon black in the morning. But after we wake up, shower, dress, and get into our car, all this fantasy becomes just a teensy (and, let's be honest, maybe even a nonexistent) sliver of our real-world life behind the wheel. Most of our time is spent noodling along with the wheel at close to dead-center-a hand's width of steering-wheel-rim deflection either way being about all we need to traverse tens of miles of highway. On-center steering is the humble meat and potatoes of our relationship with handling. And that's exactly why the likes of BMW, Porsche, and Ferrari sweat over it.

While the controls of just about everything we operate require motion (wheel turning) and resistance (most authorities opine that resistance counts more), here we've measured the off-center sensitivity of each car's (easier to quantify) steering motion-each of our entries was steadied at 50 mph and then languidly steered left, then steered back to center, then steered right, back and forth, again and again. This isn't a violent reaction test. In fact, it's just the opposite-typical in the real world.

The results here are the degrees of steering angle needed to muster a subtle 0.1 g of lateral cornering, averaged from several of these slow off-center weaves. The "Starbucks Triple-Latte Sensitivity Award" in this test goes to the 911 GT3, but what's more notable is that this isn't the simple demonstration of steering ratio you might suppose it to be.Move the rim with your hands and compliances have to be taken up, stictions overcome, distortions developed in tires to produce the necessary slip angles. If steering ratio were all that mattered, how to explain then the GT3's sensitivity in view of its slow steering ratio-a lazy 17.1:1 just off center?

2. Skidpad

Cadillac's vehicle-dynamics icon Maurice Olley once remarked, "Don't let anyone persuade you that the skidpad tests don't mean anything. They mean pretty much everything if we just take the trouble to interpret them." Some 70 years after Cadillac began routine skidpadding, we still hold true to this fact.

Testing technology is permitting us to learn even more when orbiting those tiny circles. The popular automotive press-us-generally reports from skidpadding the peak lateral g, averaged from right and left cornering. It's a single number, but a popular one almost every car guy wants to know. However, for this extravaganza, we slapped on every gizmo in sight to aim for even greater insight.

Either the left or right direction's sequence of lapping started with a slow, walking-pace lap around the circle's circumference. Why? To record each car's "zero-speed" steering angle. Which is just what it sounds like: the amount of steer angle required for a particular turn radius when no tire slip whatsoever is occurring. That done, the speed is slowly built up, and, as it is, increasing amounts of steering usually have to be added to that original zero-speed angle to hold the path. We're measuring two things here: the relationship between steering angle and the car's rising lateral g along with its growing, steady-state understeer.

The concept of understeer is that straightforward-it's simply how much steering needs to be added (as speed rises) to the minimum required to creep along a corner's path. As an aside, neutral steer is when the chassis's own slip angle (really, that of its rear tires) inadvertently adds exactly enough yaw rotation slip that no additional steering is required. And steady-state oversteer occurs when the contribution of chassis slip angle is so large you have to reduce the zero-speed angle to hold course. (In reality, steady-state neutral and oversteer are virtually nonexistent; you need to toy with the throttle to provoke them.)

In the nearby plots, you'll notice characteristics of the steering angle versus lateral g curves. For instance, the slope of their initial rise defines a car's cornering stiffness (the steering's responsiveness during casual cornering once beyond its narrow on-center window). The straight portion is called-guess what?-the steering's linear range. How gracefully it then tapers to its lateral grip peak suggests how articulately it can telegraph the front tire's ultimate grip. Before the peak, however, we've additionally sampled each car's steering effort at 0.5 g (and, gads, does the unassisted Lotus need muscle to turn!).Done? Not on your life. On each car's flanks we mounted ride-height sensors in such a way that, as cornering rose, body roll angle could be documented as well. (None embarrassingly turtled-over, of course, but it's fun to see how roll angle correlates with ride quality, which we'll examine in more detail later on). A final dollop of insight came from a 0.75-inch-tall plastic cable strip fastened to the road transversely to the car's path (think of one of those industrial-grade cable enclosures you see laid across concrete floors). Each lap it sharply jolted the suspensions; a car might grip like a gecko on typical tarmac, but bumps like this are often tossed at us by the real world-and usually at the worst moments.

However, the joker in our skidpadding deck turned out not to be a bump, but a subtle layer of grime lurking on the surface (a common feature of sometimes breezy airstrips). While just about all the cars blithely traversed it, the 911 regarded it as an invisible banana peal, making its grip just erratic enough to let its brother, the Cayman S, modestly best it. How do you explain that at the Porsche dinner table? (Its only saving grace was barely nipping the Corvette Z06's solid 0.980 g.) When it did hook up, though, the 911 gripped big time: A dissection of the data demonstrates it periodically punished the pavement at nearly 1.2 g where the asphalt was unblemished. Wow.

 

[mikeyb]

3. Step-Steer

Think our highways are vehicular pinball games already? Try driving them without direction-ensuring stability. The trick here is in picking the right low dose of straight-line-stability to give a performance- car pumpkinseed-between-the-fingers maneuverability without pumpkinseed-between-the-fingers instability. How do we gauge such cornering reactions? Let's introduce the step-steer test.

While cruising along at a predetermined speed (here, 50 mph) you suddenly snap the wheel to a precise angle and time how long it takes for the car to assume some characteristic of steady-state cornering. The mechanics of doing this are straightforward: A couple of trial runs establish the steering angle needed to nail a desired cornering rate (in our case, 0.5 g), and a fabric strap pinched in the side-window jam provides a reliable steering-input limiter.

For this exercise, we picked a percentage of a car's final yaw (rotation) rate as our cue that each cornering mission was accomplished. It happens that other cues are sometimes chosen, too-a particular lateral-g level or even an angle of body roll, for instance. All of them are sequentially involved in the fast unfolding events of a car initiating a turn.When a steering wheel is twisted off center, the front wheel's angling against the road perturbs a toroidal twist in the tire that quickly evolves into lateral-force-producing slip angles at the front contact patches. That force directs the car's nose to head off laterally, creating the yaw (rotation) we're measuring here. Realize, however, that only after the car's chassis starts yawing (due to the front tires' lateral forces) do the rear tires begin contributing lateral forces of their own-the real mark of full-blown cornering. If yaw and lateral-g response times are measured, the former always occurs first; and every time you turn your car, these physics play out all over again.

4. Lane-Change/Swerve Stability

So far, we've been examining targeted aspects of handling in a sequence of progressively rising vehicle-dynamics complexity: moving the wheel just a bit to capture on-center sensitivity; cornering gradually faster around a circle; seeing what happens when you introduce a known, destabilizing steering jolt. All of these are "open-loop" tests: You do something to the car and let the data-logger watch what happens. Driver involvement is nil or minimal at best.

This next one, the lane change, is a clear "closed-loop test" and a different animal. Here we're introducing human judgment into the middle of things.


Regular readers might be scanning this page with raised eyebrows. Lane change? What, no slalom? Nope. For this outing we're taking a different route to explore the more real-world emergency-avoidance lane change (when a ball rolls into the street, for example). It's a test employed quite often in the auto industry.

There are a number of ways to bake this particular handling potato; for instance, you might enter the cones at an agreed upon speed. But that's just not us; we blistered through as fast as possible, partly because it better challenges each car's grip limit, and, well, it's a whole lot more fun that way.

Although we're showing in our results each car's average speed, in truth the numbers are secondary here (I can't believe I'm saying this). The international standard for this test (ISO 3888-1) actually admits as much, noting that goosing the throttle at the maneuver's end can give a powerful but middling-handling car a result that's highly scattered from the driver's savvy opinion. Even they recommend downplaying the numbers. Hence, so will we; confidence in a car matters a lot more than a fraction of a second in elapsed time. Test equipment doesn't sweat and palpitate. People do. So read the comments.

5. Ride Quality


Ride quality, we hear you thinking? What's that doing in the middle of a giant handling test? Telling "the rest of the story," as Paul Harvey famously quips. Ride and handling are among vehicle dynamics' central tug o' wars-move the rope closer to razor-edged handling, and the other end gets yanked that much away from ride-quality bliss. Or at least that used to be the case. Adjustable suspensions (at least, effective ones) have put some potential stretch in that old tug-o'-war rope, but until every ride-influencing component becomes equivalently adjustable (bushings and tires, for instance) consider the old tug-o'-war rope still unsevered.

Here, we offer you a spot check of electronically documented ride-quality measurement. Quantifying ride is a fun, repeatable exercise, rich in insight, and decidedly imperfect in its resemblance to human experience.

Nevertheless, we can learn a lot from it. The setup we used here was to fit accelerometers beneath the driver's bum (er, nestled in a rubber pad) sensing vertical "bumps," with two more attached to the seat quantifying "pitch" and lateral "roll" vibrations. All three are sampled 100 times per second (100 Hz) and later projected into a frequency spectrum (think of your sound-system's jiggling bar graphs for a reference). And after being subjected to a weighting filter that approximates human perception of vibration, all three accelerometer results are ultimately summed into a final ride score according to a method prescribed by ISO 2631. With me? Alrighty then.

Below you'll find a few of the vertical g (rump-buster) frequency results: At the "bass end" of the spectrum is the springs' natural frequency (usually between 1.5 or 2.5 Hz). Higher up is the frequency at which the driver oscillates on the seat (the driver and seat amounting to a mini suspension system all its own), and next are the jitters of the wheel and tire packages, agitated by the tire's undamped sidewalls. Next time you're scorching down the road, see if you recognize these vibrations. Often you can.


Conclusion

With phase one of our analysis three-rounder complete, do we see a handling champ in the making? Not yet. But the 911 GT3 has bubbled to the top in nearly every test scenario, only faltering in our expectations around the skidpad (if you call second place a slip) and numerically in ride quality, where its last place showing wasn't subjectively bad enough to hobble it. The Porsche Maximus might the early leader by a nose, but the Cayman S, Lotus Exige, and Z06 will be nipping at its heels all the way to Laguna Seca.

 

[mikeyb]

Track Attack! Laguna Seca
Go, stop, toss, turn - to the max!

An empty road-racing course designed with a good smattering of corners-fast and slow, increasing and decreasing radius, uphill and down, on-camber and off-is an ideal place to assess a car's dynamic handling behavior, especially for a driver who's logged countless hours lapping the track. Max Angelelli knows every braking point, FIA curbed apex, and blind corner of Mazda Raceway at Laguna Seca, which frees up a lot more of his mental capacity to concentrate on assessing the car's behavior than might be available to mere auto-scribes with vastly less experience on this track. To facilitate meaningful comparisons, we ran Angelelli through the cars in ascending order of anticipated performance, starting with the front-drivers and ending with the fire-breathers.

Mazdaspeed3 GT
Best lap (min:sec): 1:50.375
73.0 mph avg
1.31g peak lateral acceleration

This hot hatch delivered better-than-expected traction that didn't degrade a bit over five hot laps. Angelelli reported being able to carry surprisingly high (for a front-driver) mid-corner speeds through the fastest turns, though he cautioned, "If you have to brake going into the corner late, you can go into an oversteer situation that's very hard to control." Thanks to the Mazda's neutral handling, its tires showed no shoulder or sidewall wear, and its seventh-place performance overall topped that of the S2000. Angelelli praised the light, direct steering, but felt more anti-lock activity than expected from the brakes.

Honda Civic Si
Best lap (min:sec): 1:54.990
70.1 mph avg
1.55g peak lateral acceleration

Relative to the Mazda, the Civic's suspension struck Max as allowing more body roll while feeling stiffer on bumps and permitting more understeer. Its brakes also were noticeably less powerful, evincing fade by the end of the second lap. With less peak power and a somewhat "lazy" steering response, "everything is in slow motion, so you have a lot of time to react." Angelelli's recommended changes: "First, I would improve the traction by changing shocks, springs, or the diff, and then increase power for enjoyment." The right-front tire sidewall took a beating en route to a last-place finish here.

Mini Cooper S JC Works GP
Best lap (MIN:SEC): 1:51.733
72.1 mph avg
1.28g peak lateral acceleration

ANGELELLI emerged from the Mini smiling. "This car makes me laugh, it's such a little sports car; a little racing car." Of the three front-drivers, it corners the flattest, exhibiting excellent traction and steering effort that builds naturally with cornering loads. Power is modest, but the gearing is perfect. Mini recommends identical tire pressures front and rear, but Angelelli would run a bit less in the rears to add bite and reduce oversteer in the fastest corners. The Mazdaspeed3 entered and left seven of the turns faster than the Mini, giving it a 1.4-second advantage, though the Mini sustained higher average gs in six turns.

Mitsubishi Evolution IX MR
Best lap (MIN:SEC): 1:47.926
74.7 mph avg
1.36g peak lateral acceleration

ANGELELLI's hot-laps in the Evo were among the noisiest, and its Yokohama Advans came in showing shoulder wear front and rear. That's because it drifts like a rally car at the limit, even in "tarmac" mode. A quick steering ratio keeps countersteer flailing to a minimum, but Angelelli found himself correcting a lot in most corners. He offered high praise for the Evo's powerful brakes and stellar traction. "With four-wheel drive, going out of the last corner of the track you can really smash the throttle and take off like a champ." Indeed our MR entered the front straight fastest of all at 58.7 mph.

Honda S2000
Best lap (MIN:SEC) : 1:50.738
72.8 mph avg
1.40g peak lateral acceleration

Honda's F1-inspired engine and close-ratio gearbox obliged Angelelli's inclination toward gentle, controllable power-on oversteer exiting corners of varying speeds, and he appreciated the S2000's light, direct, intuitive steering and neutral demeanor. He found fault with the brakes, which faded badly enough to cut his run short by a lap, and with the chassis's overall rigidity, which trailed those of the closed cars. "Going into the corners you can feel a twist in the beginning, and then it settles to the end of the corner." He recommends softening the rear springs for track duty.

BMW 335i
Best lap (min:sec): 1:49.038
73.9 mph avg
1.32g peak lateral acceleration

The 335i has a smoking habit on the racetrack. "This car needs a tighter rear diff. The inside wheel is just spinning." There's no mechanical limited-slip device offered on the 335i, as it's intended for road use, where electronic aids snuff out the smoke. Long-legged gearing is also autobahn-optimized, and Angelelli found this big coupe's general chassis feel a bit heavy and "lazy," with lots of body roll. Natural understeer transitions easily and safely to oversteer on demand, and its twin-turbo engine and powerful brakes helped Angelelli hustle it around to a sixth-place finish here.

Lotus Exige S
Best lap (min:sec): 1:45.818
76.1 mph avg
1.54g peak lateral acceleration

Angelelli felt so at home in this one that he stayed out for an extra lap. "This is a proper sports car. It has the same handling, the same characteristics as the [Pontiac-Riley LMP] race car I'm usually driving." It is demanding, however. The steering is heavy, and the rear end breaks loose with little warning. "You need to use the steering wheel a lot-to turn in and then correct." The brakes are direct, powerful, and linear, though Angelelli prescribes a bit more rear bias. He'd also like to lower the rear spring rates for more grip and to add 100 horsepower.

Porsche Cayman S
Best lap (min:sec): 1:47.577
74.9 mph avg
1.43g peak lateral acceleration

Compared with the Lotus, the Cayman struck Angelelli as soft and refined, but he praised its balance and poise. "This one at the moment is the easiest [to drive], not the fastest-because of the weight, because of the power-but definitely the easiest and most enjoyable." Neutral in flat corners, it verged on oversteer in downhill turns and understeer in the uphill ones. He found the brakes indefatigable and the powerband appealingly broad, all of which contributes to smooth track work. As for improvements? "Maybe give it more power at the top end." Amen.

Porsche 911 GT3
Best lap (MIN:SEC): 1:39.517
81.0 mph avg
1.69g peak lateral acceleration

"The Porsche 911 GT3, to me, is perfect. I wouldn't change anything. I like the engine, the handling, the [ceramic] brakes. This is the car every race driver would like to have on the rack. It's neutral and has a wide powerband, the brakes are stable, it's well balanced, it has good traction. It's perfect." Angelelli also admitted that even with all his experience, he probably only probed 85 to 90 percent of this car's astounding limits, and he opined that its behavior probably becomes unpredictable at the limits-especially under power. When asked whether a Cayman with the same engine, brakes, and tires wouldn't be better still, he fretted about its less rigid chassis. We'd sure love the chance to try that option.

ANGELELLI saved the Z06 for last because prior experience suggested it would be a handful. "This one is definitely the most difficult to drive, because of the low rear grip and high power." He doubts tires are the only culprit. "The front seems soft and heavy, and the rear seems too stiff and very light. So when you go into all the corners, you have enormous oversteer. Off power, on power, on brake-everywhere." Angelelli also felt the chassis could use more torsional rigidity. He praised the steering and the limited-slip differential, however, and reckoned he'd love to take a Z06 out hunting Ferraris on the Italian autostrada, but for a twisty road-he'll choose the 911 or Exige.

Our fancy electronic equipment corroborated most of Angelelli's seat-of-the-pants impressions, adding some additional color commentary, shown in our multihued graphs. For example, the lowly Civic, which logged the slowest lap times proved to be remarkably well behaved. Angelelli called it neutral, tending to understeer at higher speeds, and indeed below 0.80 g, the chassis slip-angle chart shows it behaves as well as or better than the 911 GT3 in terms of going where it's pointed-albeit in slower motion. At the other end of the civility spectrum, Angelelli deemed the Corvette most difficult to drive, followed closely by the Exige S. Our gear suggests maybe his familiarity with mid-engine race cars might have colored his opinion here, as our gear identified the Lotus as twitchiest, wagging its tail with an average chassis-slip-angle change rate of 1.96 degrees per second, to the Z06's 1.76. He pronounced the Cayman the easiest of the faster cars to drive at Laguna, and indeed, though its tires generate less grip than the 911's, the chassis slip angle rises smoothly and predictably as cornering loads build, and the average rate of slip-angle change is just 1.45 degrees per second. The mighty-mite Mini is more adept at generating smiles than big track numbers, but its lateral-g and slip-angle traces are smooth and well behaved, suggesting there's no dark side to its cheerful tossability. The same can be said for the big, well-mannered BMW. Angelelli's ambivalent impressions of the Honda S2000 were backed up by disturbing electronic results. Not only did it rank eighth in lap time and seventh in average grip, it also displayed squirrelly behavior while doing so, as evidenced by its squiggly lateral-g traces and its third-highest slip-angle-rate figure (1.63 degrees/second).

Our pleasant surprise at Laguna was that the biggest number generator also managed to circle the track leaving no electronic trace of misbehavior. The 911 GT3 logged the quickest lap time, the highest-peak lateral grip, the least amount of chassis slip at nearly all levels of grip, the least degradation in grip at increasing speeds, and the lowest "scare-factor" in the group (rate of chassis-slip-angle change). Hence, we have little choice but to concur with Max Angelelli and crown the Porsche 911 GT3 as winner of the racetrack segment of our contest, followed closely by the Porsche Cayman S, Lotus Exige S, and Mitsubishi Evolution IX MR.

 

[mikeyb]

Road Loop
How was it out in the real world?

So enough of the "professional driver, closed course, kids: don't try this at home" stuff. What are these 10 MVPs like to drive on public roads paved with imperfect asphalt and populated by unpredictable cars, children, and animals lurking around unfamiliar corners? Clearly the parameters are way different when we're forced to share the road. The intensity of the driving experience is necessarily dialed back from what it is on a racetrack, but the sounds, smells, tingles, and twitches generated by a great car can bring abundant joy on a Sunday drive. It was our editors' task to quantify that joy.

When we reached our top-secret 15-mile driving loop, we distributed voting ballots that focused each driver's attention on seven areas of specific interest, with each rated on a scale of 1 to 5, 5 being best. The first three categories rated the steering systems on feel (1=uncommunicative, numb; 5=tells all about road's texture and grip), accuracy (1=imprecise, nonlinear, requires midcourse correction; 5=crisp, responsive, precise), and weighting (1=artificially light or heavy; 5=effort builds naturally with cornering force). Next we examined the overall chassis's composure (1=excessive body motions; 5=minimal pitch, roll, and float), response (1=delayed or nonlinear response to control inputs; 5=immediate, predictable response), and control (1=limits and limit-behavior unpredictable; 5=limits easily probed with confidence). We scrutinized cockpit ergonomics, awarding low scores to unsupportive seats or poorly placed controls and top marks to cars that fit us like a driving shoe. Finally, and perhaps most important, we ranked how much fun each car was to drive on a hilly, twisty, wine-country lane. It's interesting that the fun-to-drive rankings mirror the total finishing order almost perfectly. The spider-chart presents our averaged rankings at a glance, and our discussion here will start from the center and work outward along this web of handling excellence.

Perhaps the best thing about the Honda Civic Si's handling is that its limits of adhesion are low enough to be explored safely with ease, and indeed the chassis-control rating was its highest. The old saying about having more fun driving a slow car fast than a fast car slow might have elevated the Civic, but in this company, its slightly numb steering, nonlinear brake response, and underdamped suspension earned last-place marks in every category except cockpit ergonomics.


Our Corvette Z06's surprising ninth-place ranking owes much to its last-place chassis-control ranking. Quite simply, this car's surplus power and surfeit of rear grip overwork the stability control, the driver's nerves, or both, on public roads. Combine that with imprecise steering and a chassis that's easily upset in bumpy curves and pretty soon in a handling test, you're looking at a low ranking. Considering the rest of its impressive performance envelope, it's certainly a bargain supercar, but it turns out that handling isn't its longest suit.Compared with the Civic, the Mazdaspeed3 cranks everything up one noticeable notch. There's added grip for better turn-in, tighter chassis control, lots more power-and, yes, unwelcome torque steer. Efforts to mask the latter stifle steering-feel, which robs some of the joy of handling, though the helm remains nicely weighted and linear in its response. Drivers felt well bolted into the seat, too, awarding the Mazda fourth place in ergonomics. All reported being pleasantly surprised and entertained on these roads, but seven cars delivered more joy.Our Honda S2000 divided voters. Some praised its steering and chassis responses as accurate and precise, with little lost motion and superb overall control at the limits. Another camp-in which the campers all stand well over six feet tall-felt a lack of crispness and an artificial heft in the controls that robbed them of confidence in the car's limited handling capabilities. Clearly, fitting comfortably into its tidy cockpit is essential for deriving full pleasure from an S2000.

BMW's 3 Series has been a staff favorite for brisk country drives since its origin as the 1600/2002. Evolutionary gentrification of the breed improves compliance and comfort and adds a lag-free twin-turbo engine this year, but every logbook commentator lamented a loss of road feel from the base steering system. Our sport-package tester ranked second only to the Cayman in chassis control, however, and editor MacKenzie pronounced it "the best everyday real-world ride/handling compromise here," sealing a solid midpack finish, with only overt sports cars ranked ahead of it.

As a smile generator, the Mini Cooper S JCW GP is tough to beat. After all, it's a cartoon of a sports car. Everything is exaggerated for effect-acceleration, braking, cornering, and bumps in the road. There's almost no pitch or roll, ever, but a tall seating position and limited side bolstering had all voters hanging onto the wheel for dear life. Astute stability-control programming keeps the car on track with minimal intrusion, contributing to a third-place ranking in chassis control, fifth overall.

car can truly be deemed idiot-proof (there'll always be a better idiot), but the Mitsubishi Evo comes mighty close. This nearly neutral, vice-free, high-limit chassis breaks away gently, sliding all four wheels on asphalt just as it does on gravel in rally duty. Add strong power and stronger brakes and a snug-fitting seat and cockpit, and the driver's confidence level increases accordingly. Even the steering manages to transmit more road-surface info than those in the front-drivers. It's a strong fourth-place finisher.
The Lotus Exige S was the only car to receive a unanimous 5.0 score-three, actually, for steering feel, accuracy, and chassis response. It goes to show what's possible when no concessions are made to comfort or civility. Every voter felt soldered directly into the control system of this car, praising every aspect but ride quality and the cockpit (ninth place). Many also admitted fear of misbehavior at the limits, which were too high to probe in public. This one's for purists only.

That leaves the two Porsches duking it out for top honors. At the outset of this exercise, the smart money was on the Cayman S. The two cars share an almost ideal cockpit layout (the 911 GT3's offers a bit more legroom), and both boast sublime steering and brake systems that communicate nearly as well as the Lotus's without beating the driver up as much. The Cayman's ideal mid-engine weight balance and more modest grip levels seemed sure to result in a more fun drive, enabling surefooted sliding that would be impossible with the GT3's Herculean grip. And indeed the Cayman's approachable, benign limits earned it first place in chassis control (the 911 ranked fifth). But it turns out that enormous rear tires with a tread compound that could wear out before the first oil change can indeed rewrite the laws of physics, erasing all predicted ill effects of the 911's inherent rear-weight imbalance. The GT3's 1.9-point margin of victory speaks volumes as to how civilized and livable the new adjustable suspension makes this car. It also hints at the degree to which we were impressed with its 415 linear and smoothly delivered horses and its ceramic brakes, which seem powerful enough to activate the airbags and yet are easily modulated, never grabby-even when stone cold. Confessed Caymanophile St. Antoine put it so: "Every virtue the Cayman demonstrates, the GT3 does just that much better." Editor MacKenzie concurs, adding, "This thing makes you feel like God and Schumacher, rolled into one." Amen, and joy to the world!

The Winner: Porsche 911 GT3
Put this critical number on your speed dial

Which is the best-handling car in the land? We've instrumented and tested 10 great cars, microanalyzing the vehicle dynamics of each. We've spent hours at the wheel assessing the nuanced feedback each car transmits to its driver. The logbook notes, interview tapes, and 420 megabytes' worth of objective test data agree: The clear winner is the Michelin-Porsche 911 GT3. We're augmenting the nomenclature here to acknowledge the immense contribution the Michelin Pilot Sport Cup tires make to the tail-heavy Porsche icon's handling. They're the closest thing to a full-on racing tire that is DOT-legal for street use. They're fair game because GT3s come from the factory with these tires, and we doubt that fitting conventional summer rubber would lower the GT3's performance to a second-place finish here. Conversely, a set of Cup tires probably wouldn't have enabled any of the competitors to outperform the GT3.

But we're nagged by the fact that the astonishing fair-weather grip generated by these new Michelins will degrade markedly when the short-lived gooey tread compound wears down, when temperatures fall, or when anyone so much as wet-sneezes on the pavement. If actual rain is falling, park this car or limp it home as though negotiating a blizzard in Buffalo.

Need a car that's not afraid of the rain? The other nine contenders are less weather sensitive, but the second-place finisher is less obvious. The Evo's excellent steering, approachable limits, and eagerness to slide earned it high praise as one of the easiest cars (and certainly the best sedan) in which to hustle, but it scored too few podium finishes. The visceral Lotus Exige entertained all drivers, finishing tops in step-steer response, second in the lane-change, and third at Laguna, but lost big points for its tricky limit behavior and dodgy ergonomics. And so the deft, sophisticated Cayman S earns our Miss Congeniality prize for nipping at the wild Lotus's heels at Laguna and in other transient maneuvers like the lane-change while out-scoring it handily in ride quality, slip-angle shenanigans, and on the public-road drive.

Of course, all these contenders are terrific handling cars, and any might respond well to some savvy chassis tinkering-just don't expect springs, shocks, or Cup tires to turn any of them into a GT3.

 

[jred321]

wow the gt3 blew everyone away on the track (no times posted for the z06 though). some day...

 

[ZoomZoomH]

that angeleli feedback post is great mike, awesome read

 

[Jeph]

Nice, the whole story I am amazed how fast the MS3 was through the track.

 

[Antoine]

Props to the new kid on the block...The MS3 is a mere "sport compact" but even in stock form it can really run with the "BIG DOGS"...You're my hero MS3 (first)(drive2)

 

[SilverBulletES]

Cool article but nothing in the results was a surprise.

Still.... I would rather have a Boxster S over any of those. 98% Cayman handling with top down happy ending. When I win the lottery.

And for everyday practicality, Mazdaspeed3 is really the only choice... duh.

The GT3 is basically a race car you can buy in a dealership, seems a bit unfair. And you can own both the above cars for what? 30 grand less? Or you could have the Boxster S and the 335i for about the same price as the GT3. Plus that ridiculous wing on anywhere but the racetrack screams.... I'm a ****!

 

[Kosh]

The MS3 beat out the s2000...which beat out the rx-8...

Never would have guessed the ms3 was "better-handeling" than the rx-8. According to this test it is.

 

[Kooldino]

Great write up. The Exige S is amazing, but that GT3...good GOD.