This is an interesting issue and subject, and I have many years experience with owning a vehicle rated as a "Do not buy" by Consumer Reports. That vehicle is my 1999 Isuzu Trooper. Please excuse me for this long post, but I do have pertinent information to share.
Like the Lexus, Consumer Reports determined that the 1999-2001 Isuzu Trooper were prone to rollover, but not because of oversteering. What the problem supposedly was with the Trooper was that two wheels on one side of the vehicle would leave the ground when performing a certain evasive testing maneuver that CR still performs to this date.
The test involves driving in a straight line at about 40 mph along a lane that has traffic cones placed directly in front of the line of travel of the vehicle. The cones simulated objects that a driver might have to suddenly evade--like a child running into the street, or a driver in front of you suddenly slowing down or stopping his vehicle.
So the tester is driving down this lane at 40 mph and as he approaches this cone, he suddenly has to swerve to the right or left--without reducing speed--as if he might have to do in a real world situation in order to avoid the child, or vehicle or other object in front of him. The maneuver involves swerving around object then immediately swerving back into his lane as if there are oncoming vehicles approaching him.
CR would put outriggers on the tested vehicles to prevent them from rolling over in the event that they pitched too much, and caused the wheels on one side of the vehicle to leave the ground. According to CR the Trooper failed with flying colors. The only other vehicle to ever fail that test prior to the Isuzu Trooper was the little Suzuki Samari which also got a "Do not buy rating."
So, needless to say, CR really hurt (unintentionally) sales of the Isuzu Trooper--Isuzu's flagship model at the time--and undermined consumer confidence not only in their Troopers but in Isuzu products in general. Isuzu sued CR but lost the court battle. Even though a number of other publications like Car & Driver, Motor Trend, and Automobile magazine down-played the issue by pointing out that Troopers weren't going to just pitch over because someone made a hard turn and that the vehicles were in fact safe to drive, everything went downhill for Isuzu. By 2002 stopped making the Isuzu Trooper. A few years after that they stopped producing any passenger vehicles for sales in the U.S.
Isuzu's key argument in court against CR was that the test was not realistic because most drivers were "unable or unwilling to perform that maneuver." If you think about it, that's a very true statement. Next time you're driving your 4R find a long straight two-lane road with little traffic. Accelerate to 40 mph and maintain that speed. Then think about a kid suddenly running out in front of you or a vehicle suddenly stopping. Do you think you have the skills or gumption, and/or sense of mind--without touching the brakes or letting off the accelerator pedal--to make a sudden swerve to the left, clear the object, then suddenly swerve back into your lane? At 40 mph? That maneuver takes some skill and balls and a lot of luck. Even in a car. How about in your 3" lifted 4R with 33" tires?
Well, I actually made a very similar maneuver (no I'm not claiming I had the "skills and balls"), once, in the 14 years that I've owned my Trooper and it occurred about five years ago. A driver in front of me slammed on the brakes in order to avoid running over something in the road ahead of it--turned out to be a large trash can lid. I just reacted spontaneously and without thought and did the CR maneuver--but only at about 35-38 mph (I think. I wasn't exactly staring at my speedometer before, during, or after I made the maneuver). My Trooper heaved over to the left like the Queen Mary in a storm and it scared the sh*t out of me. But my wheels never left the ground and I managed to safely pull back in front of the stopped vehicle without getting slammed by an oncoming car. After I calmed down I thought, "Damn! That was spooky!"
But here's the deal, in all the years since CR reported that issue, to my knowledge there is not one recorded instance--in the real driver world-- in which an Isuzu Trooper ever rolled over as a result of that maneuver being performed. Not one documented incident that can be attributed to a driver making that maneuver. Nor has it been documented, to my knowledge, that Troopers rolled over more often (again, in the real driver world) or were anymore dangerous to drive that any other SUV. My Trooper has never come close to rolling over and believe me, Trooper's are known for body lean, and I do not feel unsafe when I drive it--nor has any other Trooper owner that I ever talked to.
So as others have mentioned in this thread, what it really boils down to is driver sensibility. SUV's are not sports cars. They have high centers of gravity. They are generally big and bulky. They don't handle or ride like other vehicles. Most of them are "tail-heavy," and have a tendency to fish-tail--like pickup trucks--even on hard surfaces. So, they must be driven differently and with more thoughtfulness. But a lot of people fail to grasp this concept.
So now Toyota is suffering from CR's "Do not buy" warning. But they have learned from history that it's best to just stop sales of the "faulty" vehicle until they fix the problem, because most likely it will also turn out to be a no-win situation for Toyota if they ignored CRs testing results.
Now, I haven't read the complete CR report on the Lexus so I don't know all the parameters of the test or the full report. But I just wanted to share this info because I am speaking as a living, uninjured 14-year owner of another vehicle that CR once reported as "unsafe" and "Do not buy." The most dangerous thing about most vehicles is the person sitting in the driver's seat--not the vehicle under their crotch.
BTW, if I have said anything that might be construed as accusing Consumer Reports of any negative and/or causal actions against Isuzu Trooper it was not intended and I apologize for doing so. (Yes. This is a disclaimer).
