Golf Statistics and Shot Value

[Richard Choi]

A fascinating article about how to truly evaluate performance on the golf course.

http://www.slate.com/id/2263079/

I would LOVE to see the shot value diagram of all of the top courses. I think that would be truly fascinating and helpful to understanding golf course architecture.

Also recommend reading the article on ShotLink; http://www.slate.com/id/2263078/

[Tom_Doak]

Richard:

I think you'd get lost in all the diagrams.  Because just as every course has different shot values, so does every individual player.

But, the article was fascinating.  I think it's probably spot on, too, in identifying the real strengths of Tiger's game vs. his peers ... which is funny, because it shows clearly that the more courses try to Tiger-proof their courses and set up longer approach shots, the more they are playing right into his strength.

[George Pazin]

Interesting piece, especially for a math geek like me, thanks for posting the link.

I've been wanting to start a thread on here for quite some time entitled "Developing an analytical framework for golf course architecture", but the biggest problem I see is compiling all the data. I just don't have the time or energy to do the heavy lifting.

For instance, a member of this site once commented that we should be learning something important from the fact that Cabrera hit something like 5 fairways in his final round while winning at Oakmont. The remark was completely ignored, but I've always wanted to sit down and go through his round to see if we could in fact learn anything about 1) Cabrera and (more importantly to us) 2) Oakmont's design.

Maybe Brent Hutto could get a grad student assistant to do the heavy lifting for me...

[Richard Choi]

Tom, wouldn't it be fascinating to see the shot value diagrams of a golf course for a scratch, mid, and high handicappers? It will really tell you whether or not certain risk/reward strategies exist and how valuable they are. It will also be informative on what features can make it easier for the high handicappers while still providing challenges for the scratch player.

George, what do you think about asking Mr. Aggers to join the fray here?

[George Pazin]

Tom, wouldn't it be fascinating to see the shot value diagrams of a golf course for a scratch, mid, and high handicappers? It will really tell you whether or not certain risk/reward strategies exist and how valuable they are. It will also be informative on what features can make it easier for the high handicappers while still providing challenges for the scratch player.

This is almost exactly the crux of my thinking on the subject as well. I actually think in many ways the info would prove more useful in analyzing golf courses than golf in general.

I'll drop Mr. Aggers a note; maybe if a few people do it, he might look in.

[Tom_Doak]

George P:

That is a terrific idea of yours.  When I used to look through those old GOLF ILLUSTRATED magazines from the 1920's and 30's in the basement of the Cornell library, they did exactly what you suggest ... they would have hole-by-hole aerial photos of a course and show every shot Bob Jones hit in the playoff for the 1929 U.S. Open, for example.

To me it was fascinating to see how far he hit the ball and what clubs he used the wide RANGE of distances he hit it from one hole to the next.  But, it would also be fascinating to see how a player like Cabrera approached Oakmont, and whether his misses were always on the correct side (in which case they weren't all misses!).

I don't know why GOLF Magazine doesn't take three pages and do this for every Open.


Richard:

The only problem I have with your suggestion is the implication that "scratch" players or "high handicappers" all hit it the same, when they most certainly don't.

Incidentally, the guys in the Tour design office think they have a great advantage over everybody else because they look at ShotLink all the time; and they are making changes to TPC courses based on what they find.  I think it would be useful to understand the trends, but I do worry that it will lead to paint-by-numbers design that doesn't fit the landscape [if any Tour course actually does].

[George Pazin]

George P:

That is a terrific idea of yours.  When I used to look through those old GOLF ILLUSTRATED magazines from the 1920's and 30's in the basement of the Cornell library, they did exactly what you suggest ... they would have hole-by-hole aerial photos of a course and show every shot Bob Jones hit in the playoff for the 1929 U.S. Open, for example.

To me it was fascinating to see how far he hit the ball and what clubs he used the wide RANGE of distances he hit it from one hole to the next.  But, it would also be fascinating to see how a player like Cabrera approached Oakmont, and whether his misses were always on the correct side (in which case they weren't all misses!).

I don't know why GOLF Magazine doesn't take three pages and do this for every Open.

Thanks, Tom, I'll see if I can find my copy of Cabrera's last round and grind it out.

[Jason Topp]

Thanks for this.  I also highly recommend the book referenced in the article - search for a perfect golf swing - which is old and based on very limited data, but more revealing about analyzing golf shots based on statistics than much else that I have seen. 

[Jason Topp]

One other interesting source of this type of data is from the first chapter or two of Dave Pelz's "Short Game Bible" in which he analyzed the error percentage of every shot for a number of rounds.  He reached the conclusion that on the PGA Tour - a player needed to fall within a certain range of error for average full swing shots in order to succeed (5% - 9% at that time) but that success depended on short game shots - specifically how often a player got the ball within 10 feet of the hole on shots requiring less than a full swing.

Again - I think his data is out of date and it is less sophisticated than this model but is far more useful than all of the ShotLink statistics.

[Richard Choi]

The great series on Golf Statistics continue today with a technical breakdown on putting.

http://www.slate.com/id/2263080/

"The poster boy for the difference between MIT's putts-gained-per-round rankings and the standard stats is Ernie Els. From 2003 to 2008, the South African star ranked 15th in the PGA Tour's putting average. During that same period, he ranked 283rd in putts gained per round—his -0.63 mark means he gave back six-tenths of a stroke to the field each 18 holes with his substandard putting.

In this case, the putts gained per round stat matches what golf fans see with their eyes: Els is a wonderful iron player but a poor putter. The reason Els' putting average is so strong, the MIT team explains, is that—on account of his great iron shots—his "first putt" distance is two feet shorter than the average pro. Putts gained per round strips away those approach shots and reveals the truth: Ernie Els is an elite player despite his putting, not because of it.

The MIT study also confirms a fact that, until this year, was a given in golf statistics: the ridiculous dominance of Tiger Woods. From 2003 to 2008, he ranked first in both putting average and putts gained, bettering the field by seven-tenths of a stroke per round with his putting alone. To put that in perspective, if Tiger was paired with Ernie Els, Ernie would give up almost a stroke and a half to Tiger on the greens."

Good STUFF!

[Tim Pitner]

Yes, we can only hope that golf, like baseball, is taken over by statistical nerds.  (Sorry for the negativity, but I have an aversion to the likes of George Will and Peter Gammons staking the wonks' claim on baseball).  Although I understand the value of statistics, I prefer the art in sport to the science. 

[George Pazin]

Yes, we can only hope that golf, like baseball, is taken over by statistical nerds.  (Sorry for the negativity, but I have an aversion to the likes of George Will and Peter Gammons staking the wonks' claim on baseball).  Although I understand the value of statistics, I prefer the art in sport to the science. 

 

The real art is in interpreting the statistics. At least, when we're talking about statistics and sports. Obviously the real real art is in the sport.

[Jason Topp]

Here is a link to a number of acadamic pieces that appear to be the foundation for the series of articles:

http://www.pgatour.com/stats/academicdata/

I like use of statistics is because so many "truisms" regharding golf stategy are patently false and statistics are a tool for seperating out the wheat from the chaff.  Statistics are only a tool and have flaws (some of which are acknowledged in the papers) but for me they take nothing away from the art of either baseball or golf. 

Examples from golf include:

"Layup to a comfortable yardage" - anyone that can hit a half wedge solid should ignore that advice.  A player is much more accurate from 40 yards than he is at full wedge range.

"Keep it below the hole" - the putting stats appear to debunk that theory.

In baseball - on base percentage is the most obvious stat that used to be overlooked but is no longer.  The value of RBI's has been seriously debunked.  I don't believe the claim that clutch hitting does not exist and probably never will.   

They also provide an excellent benchmark for determining flaws in one's own game.  Based on the putting paper - go to the practice green and see what is the longest distance from which you can average two putts.  If it is longer than 32 feet you are either an excellent putter or your practice green is much easier to putt on than the average tour green.  You can do similar tests for 8 foot putts to see how close you come to the 50/50 standard of a tour player.  The same sort of tests can be developed for any yardage you want.

[George Pazin]

I'm planning on dvr-ing the final round of the PGA and charting the shots for the last twosome, at least as much as possible. Anyone want to volunteer to chart another player's full round?

[Jud_T]

In baseball - on base percentage is the most obvious stat that used to be overlooked but is no longer.  The value of RBI's has been seriously debunked.  I don't believe the claim that clutch hitting does not exist and probably never will.   

It's statistics 101 Jason.  If a guy is a lifetime .300 hitter then regardless of whether he's 0 for his last 25 or it's the bottom of the ninth with 2 outs, the winning run's on second and he's got a 20 game hit streak going,  the odds of him getting a hit are exactly 3 in 10....

[Rich Brittingham]

I just finished reading these articles and thought "wow this nees to be posted on GCA"  sure enough, I sign in and what do I find.
While the 3 articles so far are quite 'Tiger Specific', it was interesting to see unique statistical proof of his dominance.  I was actually surprised by the PGA Tour putting statistics, they only make one 20 footer over the course of a tournament???  That was surprising.

[Richard Choi]

George I will try to chart the round of the next group they feature the most.

[Jason Topp]

In baseball - on base percentage is the most obvious stat that used to be overlooked but is no longer.  The value of RBI's has been seriously debunked.  I don't believe the claim that clutch hitting does not exist and probably never will.   

It's statistics 101 Jason.  If a guy is a lifetime .300 hitter then regardless of whether he's 0 for his last 25 or it's the bottom of the ninth with 2 outs, the winning run's on second and he's got a 20 game hit streak going,  the odds of him getting a hit are exactly 3 in 10....

I've read Bill James' writings on this issue.  I still think there is a flaw somewhere.  

[Jim Nugent]

In baseball - on base percentage is the most obvious stat that used to be overlooked but is no longer.  The value of RBI's has been seriously debunked.  I don't believe the claim that clutch hitting does not exist and probably never will.   

It's statistics 101 Jason.  If a guy is a lifetime .300 hitter then regardless of whether he's 0 for his last 25 or it's the bottom of the ninth with 2 outs, the winning run's on second and he's got a 20 game hit streak going,  the odds of him getting a hit are exactly 3 in 10....

So Tiger's odds of winning the PGA this week are about 1 in 4?  And Jack's, if he teed it up, would be around 1 in 8?  The odds of anyone winning the first time are zero?       

On the article Richard linked, my immediate reaction is to disagree with its major point.  I think current stats probably do give a good picture, of who is the most accurate driver... and who hits the most accurate approach shots.   

BTW, how close can they cut the hole to the fringe on the tour?   

[Phil_the_Author]

Jud,

Growing up my grandmother had a saying, and I honestly don't know why, that there were Liars, Damned Liars and Statisticians.

Not that you're a liar, but I must disagree with your statement, "It's statistics 101 Jason.  If a guy is a lifetime .300 hitter then regardless of whether he's 0 for his last 25 or it's the bottom of the ninth with 2 outs, the winning run's on second and he's got a 20 game hit streak going,  the odds of him getting a hit are exactly 3 in 10...."

Built into that lifetime .300 average is the facts that in 150 plate appearances with two out and two men on the batter got 75 hits while over his career he he batted just .055 while striking out 83% of the time in the first inning. In other words, each at bat has a whole series of individual aspects that can change both the actual average and statistical likelihopod of the batter getting a hit. How many runners are on base? Is one of them a pinch runner? How HEALTHY has he been leading up to the game? How OLD is he? How many years has he played in the majors? Is the pitcher throwing from the Stretch? Is the pitcher the number one reliever or is the starter who has now gone almost 9 complete innings and thrown 134 pitches. Is it a middle reliever brought in because the closer has thrown too often as of late? What is the temperature? What time of year...

There are simply TOO MANY variables that will directly effect the ACTUAL outcome to make a true statistical prediction?

[Jud_T]

Dig Babe Ruth's Grave up, prop him up at the plate and he'd hit a dinger 1 out of 15 times. .

[Sean_A]

Jud,

Growing up my grandmother had a saying, and I honestly don't know why, that there were Liars, Damned Liars and Statisticians.

Not that you're a liar, but I must disagree with your statement, "It's statistics 101 Jason.  If a guy is a lifetime .300 hitter then regardless of whether he's 0 for his last 25 or it's the bottom of the ninth with 2 outs, the winning run's on second and he's got a 20 game hit streak going,  the odds of him getting a hit are exactly 3 in 10...."

Built into that lifetime .300 average is the facts that in 150 plate appearances with two out and two men on the batter got 75 hits while over his career he he batted just .055 while striking out 83% of the time in the first inning. In other words, each at bat has a whole series of individual aspects that can change both the actual average and statistical likelihopod of the batter getting a hit. How many runners are on base? Is one of them a pinch runner? How HEALTHY has he been leading up to the game? How OLD is he? How many years has he played in the majors? Is the pitcher throwing from the Stretch? Is the pitcher the number one reliever or is the starter who has now gone almost 9 complete innings and thrown 134 pitches. Is it a middle reliever brought in because the closer has thrown too often as of late? What is the temperature? What time of year...

There are simply TOO MANY variables that will directly effect the ACTUAL outcome to make a true statistical prediction?

Philip

Of course you are right, but the important thing is the manager's decision at that moment.  Do you think he has all these stats you suggest at hand or do you think he goes with a .300 hitter assuming the guy is in good health?  Much of these sorts of stats are after the fact and mean nothing in the course of the game.  

We could talk about the same thing with Tiger.  I don't think he will make the US team on points, but does a team really not want to be without the best golfer on the planet if he is willing and happy to play?  Recent stats (and Ryder Cup stats) suggest Tiger may be a dubious pic, but...

Ciao

[George Pazin]

George I will try to chart the round of the next group they feature the most.

Thanks, Richard.

Jim N, I can see your point. That's kind of what I thought when I read the article, how much value is there in this, other than perhaps to tell a particular player what he needs to work on? I do think there may be value with regard to architecture - things like showing if there is a preferred side of fairway or green, if a risk/reward is truly a risk/reward. etc.

If one simply wishes to try to determine the best golfer - seemingly the goal of the article and the professor's work - I'm going to suggest score as the metric of choice...

[Phil_the_Author]

Richard,

Actually, yes, the manager DOES have a multitude of facts and tendencies backed up by statistics at his hand in the dugout during the game. Where do they come from? Among the myriad of professional sources producing them for him are the daily CHARTS that the starting pitcher from the day before keeps during that days game. These pitch-by-pitch accounts are compiled and used by managers and players.

Also, just listen to what announcers actually say during telecasts. How often do they describe tendencies like "So-andSo has 17 hits in 23 at bats against the fireball-throwing reliever whereas everyone else on the team has only managed two walks in total in 605 at bats..."

In baseball, situational statistics are far more a predictable tool than lifetime stats and the same is true in golf.

For example, if I told you that one of the players in this years Masters had won several U.S. Opens and a British Open and just last year became the first and only player to win the U.S., British and Canadian Opens in the same year, wouldn't you see him as one of the 2-3 favorites to win?

And then I tell you his name is Lee Trevino and you immediately realize that it would take a parting of the Red Sea type miracle for him to win despite his accomplishsments and talents for there are simply too many right-to-left shots demanded to be played and Lee couldn't hit a snap hook if he played with a set of Medicus clubs... Staistics and past accomplishments need to be applied correctly in order to predict...

[Jud_T]

Phil,

You are right if the sample size is big enough of "clutch" hitting.  My point is simply that a slump, or conversely, a streak can be explained statistically.  I.e. Over the course of a career with 10,000 at bats you expect to find a run of 0-25 or 15-30. 

Jim,

The only thing keeping Tiger from being 3-1 to win this week is the fact that he has not played or practiced as much as his norm this season.  If he shows up at Augusta next year healthy and back to a normal practice/training regimine, I'll take him at 3-1.....