FL Flash, I'm not disagreeing with you, practical advice is very valuable. To say I would take it anytime over theory is a little far fetched. Just remember, I am not downing practical advice, there is plenty of it anywhere you go. I am just trying to offer up something that more or less doesn't already exist.
Larry B, you pointed out one of the ever lasting mysteries, side shock springing. The answer is not a cut and dry one. One thing we do know is that R/C car tires do behave similar to conventional tires; Meaning that the property of load sensitivity does exist. How do we know that our tires behave as conventional tires? Without going into pure tire theory, I will say as practical example, if you put more wedge in our cars (more LR/RF weight), it will tighten a car up.
Load sensitivity in simple terms is that as a tire is more heavily loaded its traction coefficient drops. This is to say, as a generic example, if a given tire with 16 oz. of load makes 16 oz. of lateral grip (=16oz lat/16oz vert=coeff=1). Now if that same tire is loaded with 32 oz., it will now only make 30 oz. of grip. (=30oz lat/32oz vert=coeff=.937). This property is what allows us to tune our cars by changing the dynamic weight transfer.
So, by that simplified theory, more roll resistance across the rear would lead to a higher percentage of the roll stiffness distribution to be carried on the rear. With more roll stiffness distribution carried in the rear a higher percentage of the lateral load transfer, would be carried about the rear tires, the rear tires would be come more unevenly load resulting in the car being looser????? Right? Well not really.
As we already know, this does not work like this in reality. The question is why? Here is an example where a simplified theory yields the wrong answer.
The reason why softening the side springs actually works to free a car up is mostly a transient handling property. As the car initiates turn in, lateral forces are developed before a car takes it's full roll angle set. Therefore, the properties of the roll balance (front/rear) of the car(tight or loose), don't take there effect until car establishes its full roll set, which often never occurs on a short flat track.
Under initial lateral force build up (the front tires forming slip angle as they are steered) the car begins to rotate or yaw, the amount based upon three major variables, front tire grip, CG location and aerodynamic side and downforce centers of pressure.
With softer side springs, these effects are more pronounced for reason that the car takes a longer time to get into it's steady state roll angle and therefore steady state balance set (which is a function of a number more variables but also including those listed before).
Our RC oval cars are heavily rear biased in weight (55-65% rear) so there natural transient handling tendency is to oversteer. The longer the delay for the car to achieve a "tighter" steady state set, the more the car will continue to rotate into a corner a quite possibly continue to rotate past center.
This is a common situation seen in modern Winston Cup racing situations. However the situation is found in reverse form. With Winston Cup cars, the majority of the roll stiffness distribution is handled by a large front anti-roll bar. As the front anti-roll bar stiffness is increased, often one will find that the car actually turns (rotates) better. This is based on the same transient handling concept, but on a different end of the car.
This is not the easiest thing to explain on paper, but one can see especially on a flat track, a car that is softly side sprung, seems to cut more aggressively into the corner and through the center, and if enough yaw momentum is built, the car will actually carry that yaw momentum all the way through the corner.
The best known example of this is a Rally car or dirt track car, the initial steering input creates enough of a yaw moment that the car builds yaw velocity quickly on entry and the rest of the corner is used trying to reduce the yaw velocity to get the car back strait.
Hence the reason why guys such as Ryan Newman and Tony Stewart (guys that are used to transient loose on entry set-ups) have been so successful as of recent in Cup. Getting a car rotated on entry allows one to run a tighter steady-state set-up, which assists in laying down power earlier and more efficiently off the corner.
A lot of these set-up factors are heavily influenced by driver preference. Drivers who prefer a tighter entry set-up give up a lot of speed middle and off of the corner and often complain that they are too loose in, too tight in the center and too loose off.
I can elaborate more on a lot of this stuff but I ought to stop and make sure the explanation up to this point makes any sense to you guys. To summarize in practical form: As a rule of thumb, the less traction that exist in the track the softer you want a car sprung in order to slow down the transients of the car (or to make the car more forgiving to drive). As bite comes up, one wants to stiffen a car as to avoid excessive roll angles and a sloppy handling car.
S & B
Remember, theory isn't wrong, theories are wrong.