PGA's New Strokes Gained stat

[ericpaul2]

www.golfwrx.com/240163/pga-tour-rolls-out-new-strokes-gained-tee-to-green-statistic/

The article above describes the stat implementation and then delves deeper into its importance. Personally, knowing what I do of Broadie's actual concept, I'm entirely underwhelmed and here's why.

Broadie'c concept it to measure strokes gained for each and every shot and compare it to the average Tour performance from the same or similar conditions. For example, if a Tour player hits from 195 y from the hole in the second cut of rough and ends up with a 4 on the hole, while the average score was 4.5, than he is +0.5 strokes gained for that shot. Add up all the strokes gained for each shot for a round and you'll get that metric (strokes gained). You can also categorize by shot type (off the tee, from the fairway, from the rough, from the sand, by distance, etc) in however many permutations you desire, limited only by an understanding that the smaller the sample size of similar shots, the lower the probable accuracy of the strokes gained statistic.

It has always been my understanding that Strokes Gained - Putting was done in exactly that way...measure each putt's length, compare the average shots to hole out from that distance for the tournament to the player, and viola.

My understanding of this stat is that it is completely different. They compare the player's eventual score to the field's average, subtract the Strokes Gained - Putting value, and the remainder is the Strokes Gained - Tee to Green. Absolutely no shot per shot comparison, no differentiation between different aspects of the game, etc. Little more than assigning a number to what we already knew...if a player wins, but their Strokes Gained - Putting stat was average, they must have had good ball striking.

Worse still about that article, the last bit drones on about potential weaknesses of the stat, when in actuality, those are weaknesses of the original concept...those weaknesses are in no way applicable to the current implementation since tracking each shot is not done.

Here's hoping this is just the first step to full implementation.

[Richie3Jack]

I think the ShotLink guys understand the flaws in Broadie's strokes gained -driving and strokes gained - long game and even strokes gained - short game metrics.

I went over this with a couple of friends of mine and a cousin of mine. One friend is an Economics professor with a PhD from the University of Georgia. The other friend has a degree in Biostatics from Brown University and my cousin just earned his PhD in Mathematics from UNC-Charlotte.

We discussed this and found many potential flaws and some glaring. For instance, let's say Bubba Watson bombs a drive that goes 380 yards long. But, he's also in the rough, 50-feet from the edge of the fairway. This leaves him with a 100-yard approach shot (we'll say a 480 yard par-4), but from the deep rough. Also, our research with a few local golf designer groups find that when you miss by a lot off the tee, often times the angle into the green is so odd that it affects the player's approach shot.

Anyway...

Let's say on this particular course the majority of 100-yard approach shots are hit by short hitting golfers on the par-5's. They can't reach them in 2 shots, so instead they lay-up to about 100 yards. And because they are laying up, if they find the rough the ball is likely to go into the first cut of rough where the grass is much shorter than it would be if the ball was 50-yards from the edge of the fairway.

And because they are short hitting Tour players, they usually are the better wedge players on Tour. They almost have to be in order to survive on Tour.

And those great wedge players are getting the ball in the hole from 100-yards away in the rough on average in 2.3 strokes.

However, if they were 50-feet from the edge of the fairway at 100-yards with an odd angle...they would likely never come close to averaging 2.3 strokes. And neither would Bubba.

That's why when you look at Broadie's 'Strokes Gained - Driving' metric, it heavily favors distance over accuracy and consistency. The big one is that it treats all rough the same and we know that the majority of the time, the further you are from the edge of the fairway the longer the rough grass is going to be.



I think the ShotLink people understood some of these glaring flaws and decided to use the 'Tee to Green' metric so they can separate putting from the rest of the game. But, like you mentioned...it really doesn't tell us much of anything.

For instance, it doesn't tell Daniel Summerhays that he's having his best year ever driving the ball, but his iron play has plummeted dramatically from 2012 and 2013.







3JACK