The counterpoint here is that a guy like Rasmus is considered an everyday player because he is just good enough to play against same handed pitching. He might not mash against left handed pitching, but he's a good enough hitter that even when you apply his notable platoon split him vs. LHP is better than what you'd have on the bench.
And a guy who is protected against tough same handed pitchers, well, it's for a reason. If he could hit the mediocre same handed pitching with some level of confidence then he wouldn't end up getting protected against the nasty guys.
So yes, what you're saying might be a consideration, but at the same time, some of what you're saying is kind of already "taken care of" in the data or in the regression.
If a lefty gets guarded against strong same handed pitching, then he'll have less PA against them, so even if his split vs. LHP is artificially high, we'd regress it further to the mean anyway.