I haven't seen the Cal Poly study but to consider Molinari's test to use fringe conditions is a little weird.
There's a narrow window where if both the ball speed and the entry point are within a small range that it's detrimental in vs. out. In all of the testing I've seen, this window is very small, and overall, the flagstick is better off in than out.
Remember, too, we're only talking about the middle 2.18" of the hole. Anything outside that and the flagstick in or out is irrelevant.
You earlier suggested his high speed was no more than three feet by because of some other unrelated commented he made.
His comments were not at all "unrelated."
His conditions are sound, if not scientific...
You don't know that. You seemingly
want to believe it, but you don't know that. There were a number of issues with his testing, including how far away from the hole he was when he was rolling the ball.
The tests I conducted were more sound than his, and I've said here I wouldn't raise them to the level of scientific standards. I didn't, for example, test or measure or ensure consistent soil density and/or firmness and/or moisture levels at the cup over my two days of testing. They undoubtedly changed quite a bit, and since I tested starting in the middle of the hole and working my way outward (alternating quite often in vs. out), and because I changed holes and angles frequently, those weren't consistent.
EM probably wasn't even hitting the same 1/8" section of the hole with 90% of his rolled balls.
FWIW, because you're sold on the data you've seen so far doesn't mean it's fact...
I have a background in the sciences. These kinds of things are just second nature. I'm not "sold" on the data, but it's pushing in one direction.
Look at the video in the Cal Poly study. The ball enters the hole in quite different locations even on subsequent rolls. Even from 32" away the ball's path can be altered by a hundred different little bumps and things. That's one of the reasons I tested from 18" away.
-----
Edit:
My advice will remain the same until the data prompts me to say otherwise:- If the flagstick is leaning a lot or it's not blowing back and forth, have it tended or take it out. These situations are rare.
- On putts where you're pretty darn sure you'll control your distance well (~3' by the hole or less), take it out if you want*.
- On putts where you're pretty darn sure, leave it in. These will tend to be longer putts, and the flagstick, overall, will provide assistance.
* Or if you're Adam Scott and you feel it helps you aim from three feet, leave it in. But these types will probably be in a small minority.
Will this save the average golfer 20 shots a year? No. Will it save you a few? Sure, it can. There will also be golfers, partly because they don't play as much, they miss the hole a lot, they take it out more than they could and maybe should, their speed is good, or some combination who aren't helped at all. Just like there will be the incredibly rare situations where a golfer is helped four or five shots in a round.