My setup mantra is to start with what is fastest for each parameter. So don't compromise anything for anything else. No changes at the front of the car to fix oversteer. No changing camber or toe to tune handling bias. Definitely no tuning bias with tire pressures. Max them all out (so that means figure out if -3.0, -3.5, or -4.0 is the best camber setting, not just "max it out"). Then make smart adjustments that make the driver/car combo more consistent and faster. Fight the urge to do things to the car that make the driver/car combo "less consistent but faster" as long as possible.
1.) As to the question of why I personally so much preferred to eliminate anti-roll bars from my setup, which I was asked here directly, I've talked about this freely on here. My number one reason is consistency. Before I dropped the front bar, the car handled differently at different venues, with changes in the weather, etc. Without the bars on the car, the handling only changes if the tires start going off, or I make a setup change. Secondarily, I believe there is more ultimate grip in cornering to be had without the bars. It is not a free lunch. The car does not launch quite as well, and braking requires slightly more skill for the same level of deceleration.
2.) Discount setup advice that comes from a different operating environment, unless you can understand the differences. In road racing, having more front stiffness doesn't help put the power down coming off a corner in the same way as it does at autocross corner exit speeds (note that I'm saying same way not meaning specifically the magnitude of the effect (which is also true) but meaning the mechanism for the effect. Toe differences overwhelmingly have a greater effect on transients in autocross than at road course speeds. And so on. A good autocross setup for a given driver can be, but isn't always a good road course setup, and vice versa!
3.) The bias of camber front to rear seems a little more than I'd expect is necessary with that much stiffness. The stiffer the car is in roll, the smaller the required disparity in camber front to rear. Remember, the principle reason for the disparity in static camber is that the rear gains more camber than the front for the same roll angle. Since the car ALWAYS rolls the same amount front as rear, and the tires are generally of the same camber preference front and rear..... the outside front tire often also has more load on it at the critical moment than the outside rear does, so that can mean needing a little more static camber too (even if the camber curves are similar, which is not the case for our cars). If you really need the -4.0 in the front, you might take advantage of running less than ideal for cornering in the rear for purposes of being able to get the throttle down sooner on corner exit, so I'm not saying it is wrong, but pointing to it as a deviation from my setup mantra.
4.) I can not explain why with any certainty, but my experience has been that rear lower than front works better than vice versa. I think many of us here have independently come to that same conclusion, some of us without knowing that others had come to the same conclusion when we did our testing (a blind test is always more convincing).
5.) Toe. Baseline should be zero. Zero is usually best for tire wear. In a steady state corner, zero is damned close to the fastest setting. Overall, zero is plenty fast, and has no evil habits. No evil habits is what you need for all your baseline settings. People use toe in at the rear of the car to compensate for other problems; it should not be presupposed that you need toe in at the rear. We use toe to tune transient behavior, not limit behavior. You have a limit behavior problem that is overshadowing any transient behavior issues. Set the toe to zero, fix the big problem, then go back to messing with toe later.
6.) If your goal is to learn to tune the car, change only one thing at a time. This can be hard. To test changing just the front bar, you need to increase spring rates at the same time as removing the bar, or you are testing a bias change, roll stiffness change, % contribution in roll from bars, maybe bump stop contribution, etc, etc, etc.
7.) It sounds like you have your rake, camber settings, toe settings, damper choice, roll stiffness bias, and driving technique all in the mix here. If you care less about understanding how it works and learning to tune, and more about driving the damned car, I'd suggest you build a plan here to copy a good setup as your next step, and tune from there. I say build it here because when you eventually deviate from some part of a good setup, people can help you know in advance what to do to adjust for it. Taking a good setup that uses super stiff front springs and running it with softer front springs, you either need to compensate by running a front bar, or by softening the rear springs and dealing with the extra body roll, for example.
I think dropping the front bar off the car, the toe to zero front and rear, and the rake to zero or slightly negative (raise the front as much as you lower the rear, then change ride height later), probably without a spring change, and before you get better dampers will make the car into something you rave about. From there it will just be fine tuning.
