I think that putts with subtle breaks are the hardest to read. I also think that matching the read, line and speed is very difficult for 18 holes. Some days I just "see" lines better. I do know I always hit a better putt if I make up my mind fairly quickly and don't change my read when I get up over a putt.
Here's how I do it. I look at the putt from the side first to determine how much uphill or downhill. So I set the speed first in my mind. Then I look at the putt from whatever is lower, behind the ball or behind the hole. If I'm still confused, I'll look at from the other direction.
I "see" a line that I think the putt should travel on just like you would if you were putting on dew covered greens. When I get over my ball I trace that "dew" line from my ball to the hole, then from the hole to my ball, then fire away.
Sometimes when I can't make up my mind if a putt breaks 12" or say 16", I visualize where would my ball end up if I putted my ball straight at the hole. Ok, it would fall 12" below the hole, that's how much loop I would play.
See, the only thing I might disagree with here is that it was even a true test of the 10,000 hour theory since a majority of people with knowledge of golf could have told Dan he wasn't taking a very good path. It was an underfunded, unscientific study that appeared to also be testing the advising professionals pet theory about the best way to learn the golf game at the same time as the 10,000 hour thing, which then introduces even more variables into an already variable-laden type of test.
The 10,000 hour theory I know is hotly contested by another study that actually claims to have found the percentage difference in performance that deliberate practice makes, but Dan doesn't seem like he actually did much testing on the 10,000 hour theory. He had pretty clear goals from the onset, but I don't believe he ever actually defined "expert" as it would relate to golfing. If it meant getting to the top 5% of all golfers in the world then he would have achieved the goal by getting down to about a 2.7. If it meant that he had to be in the top 1% it would mean he'd have to be better than a +1. It's all very dependent on how you define "expert", which is what makes it so difficult.