That defies physics.
Disregarding normal force created by aerodynamic devices (since most MR2's don't really have any), the maximum amount of grip a car produces is determined by its mass X coefficient of friction of the tires. That is the total amount of force (grip) allowing a car to corner. You may be right in the fact that a tire has a different coefficient of friction depending on tire pressure, normal force, camber, temperature, slip angle, etc. (Here's a link if you want to get your toes wet with tire data
http://www.calspan.com/transportation/tire-research) So it it absolutely possible that removing a bar can create a scenario where both tires on that axle would then be in a situation where they would be operating at a slightly higher coefficient of friction, and would therefore be creating more total grip for the car. But it doesn't seem logical to me to think that removing the bars would create a significant difference in total grip.
Also, the bar does not change the amount of weight across the car. The bar changes the amount of normal force carried by the inside tires. If you disconnect the front bar, you are taking load away from the inside rear tire and moving it to the inside front tire thus giving the front end more grip compared to the compared to the rear.
Total weight transfer is not changed by spring rates. It is THE RATE OF WEIGHT TRANSFER that is changed by the changed by the spring.
Weight transfer is calculated by, W=((a*h*m)/l)
Where:
W = total weight transfer
a = acceleration
h = the height of the center of mass
m = mass
l = wheelbase for longitudinal acceleration or trackwidth for lateral acceleration.
You have to account for the most basic physics F=m*a (force=mass*acceleration)
As the car is rolling (when the car initiates a turn) or pitching (under braking) the body of the car is accelerating... That means there are forces lost from weight transfer to accelerate the body of the car in pitch or roll. This time duration in which the car pitches or rolls is determined by the spring rates. The higher the spring rate, the more quickly the springs will balance out the acceleration of the body. Imagine driving down the road on super soft springs and you slam the brakes, the nose immediately dives and after some time period stops diving. You do the same thing on a much stiffer spring and the time from when initially hit the brakes to the time when the nose stops diving is significantly less. That means on the stiffer springs, there is a shorter time period when force was being lost accelerating the nose of the car down. In both scenarios, once the nose gets all the way down and is no longer accelerating or diving down, then TOTAL WEIGHT TRANSFER from rear to front axle is absolutely the same (as long as the height of the CG isn't changed by the length of the springs once compressed in front/extended in the rear).
You are absolutely right that the balance of the car can vary from surface to surface. I could go into my own story of car balance changing on different surfaces, but I'll spare you the details. But basically whats happening, is that a different surface will lead the tires to operating under a different set of parameters of friction and slip.Now the balance of the car has to be adjusted to get the tires back to their best operating conditions.
Hope some of this makes sense, it's late and my brain doesn't want to work anymore tonight, so I started short-cutting some explanations. :smile: