If the method you use with your particular camera and lightmeter works then stick to it. I'm just explaining the physics. Super 8 was an amateur format, manufacturers virtually never marked T stops on the lenses or explained about viewfinder light loss probably because it complicated things. Manual mode was most likely intended to be used in conjunction with the built-in meter needle, to overcome certain lighting conditions that auto-exposure doesn't handle very well, like strong back light, snow scenes etc.
With your own findings, there are variables that can affect your results when using an external meter, as mentioned, which is why testing is the only way to really be sure.
Once you move up to 16mm, lenses (particularly zooms) were often marked with T stops and if a reflex system used a prism that ate up some light, it was mentioned in the manual. The Bolex "compensated" shutter speeds confuse the hell out of a lot of people, but it's just trying to include the prism light loss in the exposure table. The Canon Scoopic is probably the closest thing mechanically to a Super 8 camera, its manual mentions that the lens is marked in T stops which accommodate light lost to the CdS cell and the viewfinder, taking it from an f/1.8 lens to T2.5.