For the rest of his data, he does not have a validly scientifically derived sample from the sample space. So statements about all golfers are not scientifically valid. He even goes to absurd lengths by even calculating strokes gained for a round Bobby Jones wrote about, and stating that Bobby lost 0.7 strokes to the average PGA tour putter. He made no mention whatsoever about why that is a total nonsense stat!
100% agree with this. Also, something that's being missed in all of this is that no golfer is "average," and many golfers have wildly out of the norm traits in their games. Try to tell my buddy with the pitching yips to "get as close to every green as possible," as a strategy, and you just gave the guy the worst possible way of playing golf ... for him ... on certain holes.
I've seen the guy make double, triple, and worse many, many times from 20 to 50 yards in front of a green where he has to cover a bunker with a pitch shot. But with a full wedge (80 to 120?) he makes par, or birdie (and some bogies, of course) virtually every time. Guy was a 1 to 2 who had the full game of a +3/+4, the putting of an 8, and the pitching game of a bad beginner.
Strokes gained or Fawcett can't tell him what to do because they don't have HIS data. If they did, they would tell him "Hey, you probably don't want to ever leave yourself a "tweener" shot over a bunker if you can help it. You tend to make huge numbers from there and very reasonable numbers from 80 to 120."
And in reality, the system would never pick up how bad his game truly is with those shots, because he's usually picking up at bogey or double depending on who he is playing in match play (where the overwhelming majority of golf is played for amateurs)!
Golf statistics are in their infancy for the average golfer. I'm excited to see what new we can learn in the future, but right now there just is not enough accurate data on average golfers, and the stats guys don't know what to do with outliers. They don't want to have to bother with them, as they are just "outliers" if you are looking at the data statistically only. However, each outlier is a living, breathing golfer whose strategy needs to be adjusted based on who they are as a unique golfer, not adjusted based on the "average."
There is much to be learned from averages for sure, but I'd just like the language that the stats guys use to be tempered more frequently: "For most golfers ..." "The overwhelming majority of golfers..." etc.
I think I'm making sense here. But who knows anymore. The older I get the more I doubt myself.
FWIW, Broadie would never tell your buddy with the chipping yips NOT to play holes accordingly. Here's a quote: "While the importance-of-the-long-game principle applies for all groups of golfers, from the best pros to the worst amateurs, individual golfers have unique areas of strength and weakness. To get better, golfers need to understand where they stand." And a bit earlier, Broadie even has a section where he talks about "awful" shots, and urges amateurs to track these in order to better address their scoring issues AND identify their best strategies for playing.
I love Broadie's work; no secret there. But I've never made the mistake (or misrepresentation; take your pick) of saying that Broadie's book is an instruction manual of any sort; it isn't, and it wouldn't be useful that way. If a golfer reads the book, and isn't led to try to better analyze THEIR OWN GAME, but instead thinks that Broadie is telling him/her or ANY other golfer how to proceed, then he or she has completely missed the point.
The most interesting thing in all these pages of discussion, and not only on this thread, are the criticisms of Broadie's data, with ZERO data offered in it's place. Broadie's data, regardless of perceived imperfections, is the best in the game, and there is no second place. To say that he needs more data from high handicappers, while ignoring the fact that he has the best data set on high handicappers in the world, is sort of silly. As are anecdotal criticisms of the data set based on "this guy that I once played with...".
The book was published in 2014; in the years since, if anybody in the world has complied data which proves a single thing in Broadie's work invalid, I am unaware of it.
Thanks for that. FWIW, I was not really talking about Broadie, specifically. More his and Fawcett's sycophants who can't have a discussion without telling you why you "don't understand" the data.