The Sad State of Golf Stats

« **Reply #25 on:** September 30, 2004, 08:11:36 PM »

Quote from: redanman on September 30, 2004, 07:38:14 PM

Pelz, now there's a guy to listen to, "Leave the flagstick in place."

Depends on you and your game, Dave, not some robot.

So what is magical about your game that taking the flagstick out makes the ball go in the hole?

So the ball is rolling toward the hole and it notices the flagstick is still in. Does it think to itself "Well, if a robot had hit me I'd go in but since it was redanman I think I'll stay out" or is the other way around "Flagstick out, I think I'll go in since it was redanman that hit me instead of the robot" or does it work some other way?

There is no circumstance where a normal flagstick sitting stationary in the middle of the hole will keep a rolling ball from dropping in. It just doesn't happen, not because Dave Pelz says so but because that's how balls and holes and flagsticks work. There is occasionally the situation where a ball rolling too fast hits the flagstick and stays in the hole whereas it would have miss if the flagstick were out.

The vast majority of the time it doesn't matter whether it's in or out. On the rare instances where it does matter, it matters because being in will help. Being in never hurt unless the flagstick is sitting leaning toward you. That's not statistics, it's not robots, it's not science, it's just common sense.

« **Reply #26 on:** October 01, 2004, 08:57:50 AM »

redanman,

Now I understand where you're coming from.

Thanks.

« **Reply #27 on:** October 01, 2004, 10:54:44 AM »

Redanman:

I didn't read your assessment, but I've never understood how Pelz can make a blanket statement to leave the pin in when it makes no assessment of speed control.

A good chipper with a good lie will either make or leave the 30' chip stoney. Jimmy Sclaffenhack will miss long and short and left and right. "Leave the pin in..." For everybody??

Anyway, back to Pelz. He does have an understanding of scoring. He measured variance of ballstriking in the 70s and 80s and realized that short game is so important because most players at Tour level are proficient ball strikers. His information is misapplied when he says an overwhelming percentage of the game is played within 60 yards. NOT IF YOU CAN'T HIT IT!!

My point is just to point out that stats are kept by Sal Johnson, the Tour, Dave Pelz, and many others. Like all statistics, they mean nothing on the surface. Someone needs to interpret them for them to be meaningful.

There is a high correlation between GIR and money won. Very little correlation between # of putts per round and $$. Yet putting is important. So BCrosby probably just needs to dig deeper to get the "good" putting stats. Hit rate from 10-20 feet? % of greens 3-putted? etc...

« **Reply #28 on:** October 01, 2004, 11:05:37 AM »

Quote from: John_Conley on October 01, 2004, 10:54:44 AM

I didn't read your assessment, but I've never understood how Pelz can make a blanket statement to leave the pin in when it makes no assessment of speed control.

I'm not sure what "assessment of speed control" means. Here's his method:

1) He rolled balls at a hole with a flagstick in it at every speed from just barely gets there to way, way too hard. He varied the aim from dead center to all the way out on the edge.

2) He rolled balls at a hole with the flagstick out at every speed from just barely gets there to way, way too hard. He varied the aim from dead center to all the way out on the edge.

What he did not find was any possible speed at any possible angle for which the flagstick kept the ball out of the hole. It just didn't happen which make perfect sense if you just look at golf balls rolling toward a hole with a flagstick in. The other thing he found was that occasionally (not often at all) the flagstick did keep the ball in the hole. The only time that happens in when the ball hits the flagstick dead center and it was going so fast that it would not have dropped if the flagstick weren't there.

His conclusion was that it doesn't matter much whether you leave it in or take it out. However, since the only time it does matter is when the flagstick helps you ought to leave it in.

People seem to dream up elaborate scenarios where the flagstick somehow magically knocks the ball out of the hole when it would have otherwise gone in. The only way that can happen is if the flagstick is moving or if it's leaning toward you rather than begin centered in the cup. And BTW, Pelz' suggestion in those situations is to take it out if there's any chance of it moving or leaning.

« **Reply #29 on:** October 01, 2004, 11:16:14 AM »

Quote from: John_Conley on October 01, 2004, 10:54:44 AM

My point is just to point out that stats are kept by Sal Johnson, the Tour, Dave Pelz, and many others. Like all statistics, they mean nothing on the surface. Someone needs to interpret them for them to be meaningful.

There is a high correlation between GIR and money won. Very little correlation between # of putts per round and $$. Yet putting is important. So BCrosby probably just needs to dig deeper to get the "good" putting stats. Hit rate from 10-20 feet? % of greens 3-putted? etc...

What people really want to know is the extent to which putting and ballstriking determine success on Tour. As a stand-in for "ballstriking" they use "GIR" and as a stand-in for "putting" they use "putts per round" or maybe "putts per GIR".

It's my hunch that "GIR" is actually a halfway decent measure of "ballstriking". The two main problems are that "GIR" doesn't include any information about how far the miss occurred or, more importantly, how far away were they attempting the approch shot from. But it must be at least an OK measure because it consistently correlates with success both on the Tour and for amateur golfers.

It's more than a hunch that "putts per round" or "putts per GIR" are nigh-useless as a measure of "putting". That's for one simple reason, those stats don't reflect how long the first putt was. The difference in likelihood of a Tour pro hitting a green from 100 yards and hitting a green from 200 yards is certainly significant but the difference in likelihood of making a 20-foot putt versus a 10-foot putt is major. Making putts is all about distance, if all you know if that a putt is from on the green you don't have any idea how likely it is to go in. That's why "putts per round" doesn't really correlate with success, Tour pros putting for birdie from 50 feet aren't going to make any money competing with a Tour pro putting for birdie from 8 feet.

« **Reply #30 on:** October 01, 2004, 11:22:10 AM »

The hole is actually big enough (or close to it) for three balls to fit through at once, so the flagstick simply isn't big enough to prevent a ball from going in, really. There room on both sides and the front for the ball to drop. Besides, a ball moving so fast that it richocets off the stick and away from the hole wouldn't have gone in anyway, since the collision with the stick MUST absorb at least some of the energy of the ball. (The only exception would be a ball that struck the back of the cup, went straight up and back down into the hole, but the stick would've made that ball even more likely to go in.)

FWIW, the Rules basically recognize this beneficial effect of the flagstick by penalizing us if we putt with the flagstick in.

Pelz is just writing about physics and his research, and leaving the stick in is the clear winner.

« **Reply #31 on:** October 01, 2004, 11:29:39 AM »

But does it matter what kind of flagstick is in the hole? The skinnier, fiberglass ones tend to let more balls in than the (newer) fatter, aluminium ones do. If I see the aluminium ones in the hole I take them out. Balls bounce off of them easier.

About stats in golf: There already is one stat that is of way more importance than all others. It is easily findable in any major newspaper. It is of equal importance for all competitors. It is objective and easily measurable. It does not take a platoon of volunteers tracking every group on every hole. Just one person in the scorer's tent.

And one with a marker to write on the scoreboard.

Its called total score.

Bill James and Billy Beane et al are trying to find what kind of players to put in place to make an effective team. Teams that more often than not score more runs than they give up. Thus individual statistics are important. And finding the important stats is important, and redundant.

In golf, who really cares about stats. At the end of the day, the question is, "What'd ya shoot?" Nobody cares how you got there.

Individual players, like Annika, may find them useful to pinpoint strengths and weaknesses. But I don't think they can be used to pinpoint which ones correlate to success or failure. Golf is simpler than that.

« **Reply #32 on:** October 01, 2004, 11:39:07 AM »

Quote from: Andy Hodson on October 01, 2004, 11:29:39 AM

Bill James and Billy Beane et al are trying to find what kind of players to put in place to make an effective team. Teams that more often than not score more runs than they give up. Thus individual statistics are important. And finding the important stats is important.

In golf, who really cares about stats. At the end of the day, the question is, "What'd ya shoot?" Nobody cares how you got there.

Individual players, like Annika, may find them useful to pinpoint strengths and weaknesses. But I don't think they can be used to pinpoint which ones correlate to success or failure. Golf is simpler than that.

Agreed that a team sport is a much more complex dynamic than an individual sport, but it is still interesting to try to determine what combination of skills most often allows one to arrive at the lowest possible score. Is distance more important that accuracy, or vice versa? Or, are they co-equal? To what extent do I "drive for show, putt for dough"?

The start of the thread was the question of how the best golfer in the world could rank 145th in putting, which implies that putting isn't important. Since we know that isn't true, then what IS important, and what is wrong with the putting statistic. That is worth asking and trying to answer.

BTW, as has been noted earlier in the thread, the advocates of sabremetrics in baseball (James, Beane, et. al.) catch a ton of crap from people who say the game is simplier than that. It just gets harder and harder to argue with them every year when the A's win their division with a payroll that is less than the left side of the Yankee infield.

« **Reply #33 on:** October 01, 2004, 11:42:57 AM »

Quote from: Andy Hodson on October 01, 2004, 11:29:39 AM

In golf, who really cares about stats. At the end of the day, the question is, "What'd ya shoot?" Nobody cares how you got there.

So what if you went out and hit every green in regulation including a whole bunch of them stiff, inside six feet for birdie, but you three-putted every green and shot 90. Wouldn't you think about it for a moment and say "I've got to work on my putting"? Of course you would. It's obvious that three-putting a bunch of holes from six feet is a problem.

In that case which statement is more useful:

1) You shot 90. That sucks. Score lower next time.

2) You three putted every green. That sucks. Go work on your putting.

Obviously in that case knowing about both the score and the three-putts is more useful than knowing just the score. So why do you find it offensive that in less obvious cases a player might want to know what exactly is causing their scores to be too high? I think every Tour player probably knows that if they could score one stroke lower every round it would make them a bunch of money. It's just hard to go out and work on scoring one stroke lower every round.

« **Reply #34 on:** October 01, 2004, 11:44:37 AM »

> p.s. You do realize that I am just having fun with this
> don't you? BTW, Do you work for IBM?

As I said earlier today. Now I know where you're coming from...

No, I work at the University of South Carolina. In a former life I used to do computer stuff but it wasn't for IBM.

« **Reply #35 on:** October 01, 2004, 11:50:20 AM »

Redanman,
Agreed, of course, that the grass absorbs more of the energy of the ball than the stick as a total process, but that begs the question. Pelz is talking about the ball that is still moving at a relatively high rate of speed when it gets done with the grass, since a ball moving slowly at that point is going in anyway, regardless of the stick. The only possiblity under the laws of physics should be that the stick absorbs at least a bit more energy and MIGHT help the process of holing out, since there is plenty of room in the hole for both stick and ball.

Otherwise, we would be saying that the stick made the ball speed up, and if that were the case, all our energy problems would be over! All we'd have to do would be bounce golf balls off flagsticks to light entire cities...

« **Reply #36 on:** October 01, 2004, 12:42:45 PM »

Quote from: BCrosby on September 29, 2004, 11:08:25 AM

Why are golf statistics so unhelpful?

I'll ask this question ... Why do we need golf stats?

We aren't managing the pitching staff, setting the lineup, calling plays and neither is the golfer.

For the sport of Golf, more so then any other sport, golf stats are purely a "made-for-tv" phenomenom.

For the PGA Tour, the only stat that matters is $ $ $

« **Reply #37 on:** October 01, 2004, 01:10:52 PM »

Mike Benham,
I have found with the HS golfers that I coach that charting rounds (stats, if you will) help them to understand what is actually keeping them from scoring better. They might believe that they need more distance, or more fairways, or whatever, only to find that the problem is wedge play or mid-irons or some other surprising aspect of the game once they actually analyze their play. You have probably found the same thing to be true of your own game during a slump.

On that same premise, it could help me if I knew more about what actually matters most on Tour. (It is also possible, of course, that if the answer turns out to be "Hit it 320 off the tee", that I am screwed!

I think that was at the heart of Bob Crosby's initial question, and it is a valid one.

« **Reply #38 on:** October 01, 2004, 01:55:36 PM »

A.G.

A Tour pro should then compare his statistics to the average of the players at the level where he'd like to be an identify where the biggest variance occurs. To reduce the variance, he can then practice in that area - whether it be on hitting fairways, getting up and down more frequently, or scoring better on par 5s (which could be a result of a flawed mental approach).

I think that statistics for golf are actually pretty good. Saying someone is "145th in putting" can be flawed if we don't really know what they are measuring. Maybe that's BCrosby's point. Anyway, baseball too has meaningless statistics that don't work in a vacuum. (One example may be the player who strikes out a lot - no balls in play - and seldom walks - less chance to score runs, but has a decent batting average.)

« **Reply #39 on:** October 01, 2004, 02:08:50 PM »

John,
Adam Dunn's year for the Reds is a great example, since he just set the record for the most whiffs. 45 HR, 101 RBI, BA around .275 (I think), all of which sounds pretty good. However, he has not ONE sac fly all season. Not one! In short, he made almost 200 outs which had no chance to help produce runs for his team, which is what the game is about. My suspicion is that he is a guy the A's will NOT be trading for or looking at as a free agent, in the same way that they let Johnny Damon with his .315 BA and .335 OBP go without tears.

« **Reply #40 on:** October 01, 2004, 02:44:38 PM »

Quote from: A.G._Crockett on October 01, 2004, 01:10:52 PM

Mike Benham,
I have found with the HS golfers that I coach that charting rounds (stats, if you will) help them to understand what is actually keeping them from scoring better. They might believe that they need more distance, or more fairways, or whatever, only to find that the problem is wedge play or mid-irons or some other surprising aspect of the game once they actually analyze their play. You have probably found the same thing to be true of your own game during a slump.

On that same premise, it could help me if I knew more about what actually matters most on Tour. (It is also possible, of course, that if the answer turns out to be "Hit it 320 off the tee", that I am screwed! I think that was at the heart of Bob Crosby's initial question, and it is a valid one.

AG -

I was saying we as spectators of the sport but I do agree with you for evaluating my own game or if coaching someone else.

For coaching, the useful benefit of the stats is to get the player to monitor their own stats, be aware that the game is not just bang it and chase it (regardless of what Shivas does) and work on areas that need improvement.

Now JakaB is not going to like to hear this but for my personal stats, which I share with no one, when I hit a tee ball exactly like I intend to and it trickles into the rough, I give myself a fairway hit

... a similar situation for the greens/fringe.

To me, the PGA Stats are useless for us as spectators ...

Mike

« **Reply #41 on:** October 01, 2004, 02:51:46 PM »

Quote from: Mike Benham on October 01, 2004, 02:44:38 PM

To me, the PGA Stats are useless for us as spectators ...

I agree with that. The only thing I've seen on a TV broadcast of a Tour event is the little scatter diagrams they show with a schematic of the green and surrounding hazards and so forth overlaid with colored dots. The dots represent where approach shots ended up and the colors show eagle, birdie, par, bogey, double, other. It's kind of cool to see how a bunker that looks like a "hazard" on a Par 5 is a great place from which to make birdie if they reach the bunker in two.

But the usual TV golf stats are as useless as the "average with runners in scoring position" nonsense on baseball telecasts.

« **Reply #42 on:** October 01, 2004, 03:07:54 PM »

Mike Benham,
The 15th at my home course is a dogleg par 5 in which my only chance to get home in two is to carry the bunker on the inside of the dogleg, but I have to kill it to get that much carry. I manage it about 1 out of 3 times. The bunker shot is not a hard one in terms of getting it to around 100 yds., so I always try to carry the bunker. If I hit it in the bunker, I count that in my personal stats as a fairway hit, because I hit the shot I was attempting. Not pure, but it tells me more about how I actually hit the ball than hitting a 5 wood to short grass.

« **Reply #43 on:** October 01, 2004, 03:16:43 PM »

Quote from: A.G._Crockett on October 01, 2004, 02:08:50 PM

... he has not ONE sac fly all season. Not one! In short, he made almost 200 outs which had no chance to help produce runs for his team, which is what the game is about.

AG -

We need to dig deeper then 0 sac flies. How many times was he in a position with a runner at 3rd and less then 2 outs? It is possible (not probable) that he never batted with a runner on 3rd with less than 2 outs and therefore a stat of 0 sac flies is not meaningful.

It is also possible that he got a BASE HIT each time he came up with a runner on 3rd with less than 2 outs which is a far better outcome for the TEAM then making an out with an RBI.

The analysis of the stats and how they were achieved is far more important than the stats themselves.

« **Reply #44 on:** October 01, 2004, 03:41:15 PM »

Sac flies are a good example of a garbage stat.

Baseball has suffered from years from reification of stats that are, in fact, garbage stats. Counting sacrifice flies is a good example. Once you start counting them, you start thinking of them as a good thing. Before you know it, runners will be stepping up to the plate with a man on third and nobody out thinking "Just hit a fly ball to the outfield" when in fact they ought to be thinking about hitting the ball hard and getting a hit.

A sacrifice fly is a bad outcome unless the game is tied in the bottom of the ninth inning and the only run that counts is the next run. Otherwise a sac fly is an out, one of your team's 27 precious outs, once those are gone the game is over.

And don't even get me started on telling runners to try and hit a ground ball to the first baseman with a man on second and nobody out. Or in the first inning having a #2 hitter with OBP SLG over .400 bunting a guy from first to second (I've seen Bobby "one run at a time" Cox do that more than once).

« **Reply #45 on:** October 01, 2004, 04:17:35 PM »

Quote from: Brent Hutto on October 01, 2004, 03:41:15 PM

Sac flies are a good example of a garbage stat.

Baseball has suffered from years from reification of stats that are, in fact, garbage stats. Counting sacrifice flies is a good example. Once you start counting them, you start thinking of them as a good thing. Before you know it, runners will be stepping up to the plate with a man on third and nobody out thinking "Just hit a fly ball to the outfield" when in fact they ought to be thinking about hitting the ball hard and getting a hit.

A sacrifice fly is a bad outcome unless the game is tied in the bottom of the ninth inning and the only run that counts is the next run. Otherwise a sac fly is an out, one of your team's 27 precious outs, once those are gone the game is over.

And don't even get me started on telling runners to try and hit a ground ball to the first baseman with a man on second and nobody out. Or in the first inning having a #2 hitter with OBP SLG over .400 bunting a guy from first to second (I've seen Bobby "one run at a time" Cox do that more than once).

Brent -

I hear what you are saying but as I noted before, it is not the stat that is important but how the stat was achieved.

In your ground ball example, if you, as a right handed batter, were being pitched on the outside of the plate, then a ground ball hit to the right side that moves the runner over is an acceptable outcome of the at bat. Is it as good as a base hit to the outfield, no, but it is better than a strikeout or pop-up that doesn't move the runner over.

The situation dictates whether or not the resulting stat is good or bad and why a summary of the stats is misleading.

Mike

News:

Brent Hutto

Brent Hutto

John_Conley

Brent Hutto

Brent Hutto

A.G._Crockett

Andy Hodson

A.G._Crockett

Brent Hutto

Brent Hutto

A.G._Crockett

Mike Benham

A.G._Crockett

John_Conley

A.G._Crockett

Mike Benham

Brent Hutto

A.G._Crockett

Mike Benham

Brent Hutto

Mike Benham