I have no real experience with quarter ellips, so can't offer much there, but I do have a word of caution to add in reference to custom suspension design in general.
There is something to be said for that "just do it" approach, and it generally works fine, until something goes wrong.
Frankly, I think there are primarily 2 categories of "reasons" people feel their "good enough" systems are "good enough" (including my own).
First, the placebo effect. The same effect that makes people think that adding a K&N filter to their mostly stock motor gives more power and or mileage. We want to believe that the results of our efforts (time, money, physical work, etc.) are productive, so we tend to ignore evidence to the contrary and focus on or even (unconsciously?) enhance any supporting evidence. Nah, the rear steer had nothing to do with my problems on that obstacle. No way the excessive anti-squat contributed to my falling off the line and rolling. But look how much flex I have! And so on...
The other category of reason is that what you have now is in reality actually better than what you had, but you don't realize how much better it could have been with a bit more design effort. Of course that is true of so many things that you can drive yourself crazy worrying about it. Built engines, transmissions, suspensions, chassis and so on all fall victim here. At some point you've got to say "enough".
I guess what I'm trying to say is that "just do it" with a few "broad rules of thumb" may well result in a perfectly usable "good enough" system for many people, but I've also seen first hand where it results in an absolutely unusable POS that MUST be completely redone to get a usable rig. This includes rigs that didn't really look "obviously wrong", but will literally flop themselves on level ground; and I'm not talking about doing doughnuts! So don't get too bogged down in details and calculations to the point you don't make progress, but I think you should make sure to understand the risks and trade-offs that may/often/usually result from insufficient "engineering". Like a friend of mine says, "follow the principles of CEE" (Close enough engineering). But only you can define what is "close enough"...