Valve train geometry

I have to disagree. Even with a stock non roller rocker, the pushrod length will not change the geometry at the valve. It will affect how high or low the adjuster is in the rocker, which will have an effect on ratio, therefore increasing or decreasing net lift, which will increase or decrease the width of the pattern. The geometry has not been made better by doing that.

Without getting too technical, suffice to say that what dictates the location of the pattern is the rockers fulcrum length. So, having to move the rocker laterally in relation to the valve stem would mean the rocker is either too long, or too short for the cylinder head design.

Yes, the fulcrum length will affect the width slightly, but the prominent cause is the amount of net valve lift. To just say narrowest sweep possible without knowing what it should be, is flying blind in my opinion. If someone was to play with the pushrod length and reduce the ratio, it would narrow the pattern, but they wouldn't be correcting the geometry. They would just be introducing more inefficiency into the valvetrain.

I approach every engine with pushrod angularity the exact same way. With trigonometry. What about an LA with 59 degree tappet and a cylinder head that has a .700" intake rocker offset. Holy compound angles Batman!

The first thing I compensate when going to a higher ratio rocker on a mechanical tappet cam is the lash. Then, if the lobe is aggressive, I'll be on the phone with the cam grinder for his input. I rely on others for what they are good at, just like they rely on me for what I am good at.

There are an awful lot of rocker arms out there that have the adjuster in line with the pushrod, or reasonably close, at half lift, yet they are still inefficient. My custom rockers take care of this problem, but those design parameters are proprietary. There's a little more to it than just a straight line.