I think it is important to be aware of that probability drop because you quite often have dogmas on the board that are competing with Physics. 67%*three cards is expected return of 2 cards, to grossly oversimplify the benefit of the first physics. 57%*3 is about 1.77 cards. 1/3rd of a card is a pretty big hit that makes it a lot easier for other available dogmas to overtake it as the best choice. For instance, in a situation where age matters little, Calendar and Wheel with 5's and below missing can overtake the dogma now, and you should consider switching. Usually it's more apples and oranges than that though.
I also think you should think about it in terms of geometric loss rather than a flat 10% loss. The second play -accomplishes- 15% less than the first play.
But more importantly, it's not the second play of physics that really concerns me it's the third one. The third physics after 2 fails has pretty rough odds of success. But I see people playing physics three times in a row even when they have other powerful dogmas available