This will be a little controversial, but elements of it have been presented to several "experts" who were unable to deny the factual part. Here goes:
1. Contrary to most commonly accepted wisdom, IMHO, the air hitting the prop (front engine) is not actually slowing down or at least very little. Rather, it is hitting the prop at an angle to the free stream as it diverts to go around the cowl. Prop designers deal with this as if it were slowdown, using the leg of the triangle as the slower speed. The bug streaks on my prop are sufficient proof that the angled hit takes place. I'm seeing between 30 and 45 degrees.
2. In my view, the problem with this is simply that the air near the base is (thus) seeing an airfoil with a longer chord than if it were measured parallel to the stream. Different airfoils have different behaviors. Until very recently, prop designers did not handle this or at least not in they way that I think they should.
3. If the spinner could be made larger AND the prop designed to correctly handle the airflow, then I think some gain might be possible because the farther out on the prop you go, the higher the Q and the higher the Reynolds number. The mass of air hitting the prop disk is the same but how it is further accelerated is where there is room to improve.
4. Craig Catto is experimenting with fences on props in the middle area to alter the spanwise flow. He is seeing some improvement. He also pays a lot of attention to the inner or root area. You can speculate as I do that a larger spinner fits this view of potential improvement. Craig's gain may just be reduced loss to induced drag at the tips, though. In his recent work his prop was used to set another record. That airplane has a very long, pointy "nose". I really don't think that's an accident.
https://www.facebook.com/cattoprops/photos/a.303998146424926.1073741832.303902023101205/530160723808666/?type=3&theater
OK, flame away.