Sub-frame connectors... Please build them correctly!

I have an MS in Mech Engr, though more a thermal/fluids guy. I agree w/ posts 34 & 60. The structural calcs in post 40 is too rough to be useful. It gives max stress (where something will break), whereas we care about max deflection of a point not even on the outlined structure. It also doesn't include the roof and such which adds tremendously to the stiffness. The input loads are also very important. A valid structural analysis might cost $200,000. I am sure manufacturers do that today (routine w/ 3-D CAD models), but unlikely they spent the big bucks to analyze 60's cars.

Simpler would be experimental deflection measurements. To estimate the stiffness improvements, jack the car up in some manner that maximizes "mid sag" deflection, repeating as you weld or bolt in subframe connectors, then weld to the floor. We all know a 2 dt hardtop greatly sags when you jack it certain ways (ex. from F & R bumpers), so much that you can't open the doors (should be measurable). My guess is that welding to the floor would help slightly but maybe just 10%.

If your goal is drag race times, do as post #34 suggests and use that as your metric as you increment the changes, but for good statistics you might require 100 runs in each configuration to resolve a difference, so save your pennies. Might be cheaper to start with a stiffer body, like a 2 door post or a 4 door. Personally, I never expect to lift the front end, so am happy with the light uni-body as Chrysler designed it.