I Actually Do increase the compression ratio when I go longer on duration, but to be more precise, I actually increase the CR when going later on the IVC.
I'm sure we've all heard of Dynamic compression ratio versus static or geometric CR terms.
If I have a configuration that I KNOW will NOT knock on a certain octane of fuel, I will generally equalise the dynamic compression ratio (which is a ratio of volumes, but that of the volume in the cylinder at IVC point versus TDC) of this config to the newer configuration. This is only a starting point ,before I go away and do my analysis work and iterative testing.
The later IVC of the new wilder cam will reduce Volumetric efficiency at lower engine speeds. Now Pat primmer is quite correct, but this is the clever bit: When the engine gets into its working range- the volumetric efficiency will be just as high if not higher then the older less "tuned" configuration and hence the cylinder pressure may be high also, however, because it is taking place later in the rev range it is less likely to be as knock limited- due to less real time being availiable. Engines are typically much more knock limited at lower speeds.
The other point of note in favour of the High-CR-late IVC cam is that, much like the Atkinson cycle, because the compression stroke is occuring much later, initial temperatures will be lower- and this will tend to offset the knock condition too.
To illustrate the Knock/high-speed point, I know of a supercharged engine running 95 octane RON fuel, that is running 0.9 bar of boost, that is not knock limited above 4700 rpm upwards and this is an engine that is producing 130-140 % VE!
To illustrate the late IVC point, during some Miller cycle work investigations I found that with IVC occuring at 51 deg ABDC the output torque was limited by knock at a cylinder pressure of 1.35 MPa, when the cam timing was retarded the onset of knock didn't occur until a cylinder pressure of 1.55 MPa!