Rocker arm recommendations?

B3,
'Ratio is not dependent on the valve side only'. Of course not, never said it was. I said in post #40, point [3] there is the valve side ratio & the p'rod side ratio.

Your drawing above depicts what I was saying [ & Smokey ] about the ratio not being constant on the valve side. You show the figure of 1.477" at max lift as I presume the lever arm length. If you are using that to calculate the ratio [ valve side ] @ max lift, it is incorrect. At max lift, the lever arm, & corresponding ratio, is greater than the nominal 1.500" because the lever arm is measured from the point of contact of the roller on the valve tip to the fulcrum centre. It might be 1.520", as an example.
There is not a valve side ratio and a pushrod side ratio. Maybe there is a problem with verbiage here, since in OZ you say extractors, and we say headers, Or, like we say intake manifold, and you say inlet manifold. Still, you are trying to quote Smokey Yunick, and he was American, so that shouldn't be an issue.

What you have is a valve side length, which is commonly referred to as "fulcrum length". This length is determined by the rocker manufacturer based on the cylinder head it has to fit, and their perception of how they want it to fit. Then there is the pushrod side length, which is the ratio length. That length is determined by what ratio the manufacturer wants the rocker arm to be. Both sides are part of the equation, but they don't both have their own ratio. That's simply not possible.

Maybe you should reread what Smokey actually said. Here are some excerpts from Power Secrets. Page 87, under the heading Geometry:
SY 1.jpg
Specifically, number 2. What exactly is the axis of the roller, if it isn't the center?

Next, the claim that the rocker keeps moving outward and "lengthening" up to full lift. That's not what Smokey had to say. Read the last two sentences. By the way, the last word that was cut off is "inboard".
SY 2.jpg

And here is the little nugget you have been talking about, which is one of the areas I disagree with Smokey. Maybe he just didn't have a decent rocker to work with, but this is a flaw in theory if I've ever seen one.
SY 3  (1).jpg
The lever does not get longer as the point of contact moves outboard, although I know what he is trying to say. It is just a poor choice of words, in my opinion. The rocker is going through an arc motion which pulls the contact point back at any position other than perpendicular to the valve stem centerline. His reasoning is that the ratio increases because the "lever gets longer". Well, so does the pushrod side when the rocker is properly designed, and is on that same perpendicular plane, which is what my previous drawing shows. That "longer lever" offsets the effect of the valve side "longer lever", and the ratio remains the same.

There are always losses in an overhead valve valvetrain, but it isn't because the ratio changes on a roller rocker. Shoe type, yes. Roller, no.