Academics have the h-index, which has been tremendously useful for quantifying impact. Something like the Lars Kestner's k-ratio but targeted at politicians seems like it would be similarly useful
I'm a student in a somewhat "practical" field of academia and my reading of the h-index is a little more cynical. It seems to me that often the people with higher h-indices have more papers (and they're of course often of very high quality) and they get cited by other papers, sometimes in a citation loop. But they aren't necessarily practical! They may not have a deployment, or often even code that isn't "gradware". I may be way wrong and I look forward to a counter view. :)
I actually agree with your points! The h-index also misses that often many people don't even read the papers they cite! Though I think the h-index should be replaced, it strikes me that it would be an interesting project to do an h-index for politicians, e.g. how many pieces of legislation have they contributed to which we're then subsequently used. It has all the same problems as a metric, but would be interesting nonetheless.