I thought about this while watching the 3rd round of the Scottish Open. I don't think I've seen it discussed yet on the board.
Just how much does having spectators help with scoring? Specifically, lowering the incidences of lost balls.
I remember 2 notable lost balls in recent weeks. The one from (I forget the contender) one of the last groups to tee off in the US Open, and Phil Mickelson on the 10th hole during the second round at the Ozarks.
I've also seen (at least anecdotally to me) a higher number of searches for balls. I just don't think those happen at nearly the same volume with spectators present.
I also think that even with the smaller search groups, there are still more folks participating in a search than would be involved with a routine foursome. There are caddies, usually a marshal or two, a rules official, and perhaps even on-course reporters and broadcast personnel that have the benefit of replay to limit the size of the area to be searched, which decidedly improves the odds of finding the ball.
I would think that all of these factors lend themselves to encouraging swinging for the fences, which will increase the size of the dispersion pattern, which all things otherwise being equal, will increase the number of lost balls and raise scores.
That's a lot of words to say that the implied strokes lost under penalty of loosing a ball seems to be less than the implied strokes gained by shortening the distance required by the shorter shot.
I'm very interested to hear other opinions on my theory, and what the potential implications are for course architecture.