Increased Randomness on Putting

[batchvt]

1 hour ago, boogielicious said:

All good science starts with looking at the problem from the simplest viewpoint. In the OP, we don’t have data to do an analysis. We can look up make percentages from various sources, LSW and Broadie, but we have no actual numbers to determine if results are in a normal distribution or other or length of putt, slope, etc.

It is best to then do a thought experiment. We are on a perfect green with a small slope (the slope doesn’t really matter in the experiment). The world best putter can putt 10 putts perfectly into the hole every time because they have the perfect speed, read and start line and their stroke is perfectly repeatable. Their ball falls into the hole at the same spot every time at the same speed which would only go 4” past the hole if it was not there.

The second putter is not very good. They don’t read puts well, cannot start the ball on line very well and cannot control speed very well. In their 10 putts, 9 miss the hole by a lot, 2-3 feet on a 10 foot putt, and miss left, right, long and short. Only one goes in because it was on the right line but not necessarily the right speed. 

Now we introduce variables into the perfect green in the form of bumps, small indentations and different grass length than can vary speed and redirect the ball to some degree. The perfect putter now, using the exact same stroke, putts again. They use the same stroke because the average variation on the green would give the same read. The perfect putter now will miss some putts due to these variables. For the thought experiment, let’s say they now miss 2/10.

The bad putter, using their bad technique now putts again. They still cannot read very well, have the same variable speed and cannot start the ball on their intended line as before. Would it make sense now with the introduced variables that they would suddenly make more putts? Probably not. They most likely would be affected the same as the good putter in terms of the speed being off and bumps redirecting the putt. But let’s say they now make 2/10 due to some randomness. 

The make percentage between the perfect putter and bad putter would get closer. Even if the bad putter now makes zero putts or 4 putts, the percentage narrows because the variation affects the perfect putter more.

This first putter who is perfect doesn't exist in human form.  Nobody can make every putt perfectly in the center of the hole at dead speed on perfectly flat ground.  There are statistics on the dispersion of tour pro trajectories and speed off the putter, and nobody hits it the same every time.

The second putter sounds like someone who started playing golf last week.  The average golfer makes over 20% of 10 footers, so you have clearly picked a very very bad putter.  

I also like that the bad putter here seems to get better because the bumps.  As if they are more likely to hit the putt just off the edge of the hole than they are to hit it on the hole.  This is a very strange assumption, and ignores how acts of skill usually get distributed in the real world.  Mark Broadie didn't just use a normal distribution when he built a model of putting for kicks.  

Now I have been accused of setting assumptions and constraints that are unreasonable, yet you have picked a putter that does not exist in real life as your example of the good putter, and a putter who beyond terrible as your bad putter.  Remind me, who is the one just making up constraints and assumptions to prove their point?

But even if we stay with your strawman thought experiment, how many of those 2-3 footers that the bad putter has left do they miss because the greens are bumpy?  Their make percentage goes down there too right?  What about the 1-footers?  They are so bad they probably miss those a decent amount, and the bumps are just going to make those putts go in less for them too.  Why didn't you include the second and thirds and fourth putts as well?  We are talking about golf right, where you have to hole the putt, and every shot counts?  

All good science tries to accurately model the world and incorporate all the important variables.  It doesn't just pick the outliers and then use those to make conclusions.  It also recognizes that even if the extreme conditions say one thing, the reasonable conditions closer to the middle may not.  

 

[StuM]

I voted “remains the same” but after reading I agree bad luck would hurt the good player more thus narrow the gap.

[boogielicious]

1 hour ago, batchvt said:

This first putter who is perfect doesn't exist in human form.  Nobody can make every putt perfectly in the center of the hole at dead speed on perfectly flat ground.  There are statistics on the dispersion of tour pro trajectories and speed off the putter, and nobody hits it the same every time.

The second putter sounds like someone who started playing golf last week.  The average golfer makes over 20% of 10 footers, so you have clearly picked a very very bad putter.  

I also like that the bad putter here seems to get better because the bumps.  As if they are more likely to hit the putt just off the edge of the hole than they are to hit it on the hole.  This is a very strange assumption, and ignores how acts of skill usually get distributed in the real world.  Mark Broadie didn't just use a normal distribution when he built a model of putting for kicks.  

Now I have been accused of setting assumptions and constraints that are unreasonable, yet you have picked a putter that does not exist in real life as your example of the good putter, and a putter who beyond terrible as your bad putter.  Remind me, who is the one just making up constraints and assumptions to prove their point?

But even if we stay with your strawman thought experiment, how many of those 2-3 footers that the bad putter has left do they miss because the greens are bumpy?  Their make percentage goes down there too right?  What about the 1-footers?  They are so bad they probably miss those a decent amount, and the bumps are just going to make those putts go in less for them too.  Why didn't you include the second and thirds and fourth putts as well?  We are talking about golf right, where you have to hole the putt, and every shot counts?  

All good science tries to accurately model the world and incorporate all the important variables.  It doesn't just pick the outliers and then use those to make conclusions.  It also recognizes that even if the extreme conditions say one thing, the reasonable conditions closer to the middle may not.  

 

Apparently you don’t understand the concept of thought experiment. We don’t have trains that move at the speed of light either, which Einstein used in his thought experiments. You are missing the point entirely. You are vastly overcomplicating your thought process with data analysis that you cannot do because you do not have the data. You cannot model the world without data, which you do not have. You instead make assumptions which don’t have validity or add more putts. The OP did not talk about making three putts or even Ernie Els Masters four putts.

In very simple math terms, if a person who has a higher percentage for making putts misses more, their make percent will get closer to a person who has a lower make percent who misses more putts.

 

[batchvt]

34 minutes ago, boogielicious said:

Apparently you don’t understand the concept of thought experiment. We don’t have trains that move at the speed of light either, which Einstein used in his thought experiments. You are missing the point entirely. You are vastly overcomplicating your thought process with data analysis that you cannot do because you do not have the data. You cannot model the world without data, which you do not have. You instead make assumptions which don’t have validity or add more putts. The OP did not talk about making three putts or even Ernie Els Masters four putts.

In very simple math terms, if a person who has a higher percentage for making putts misses more, their make percent will get closer to a person who has a lower make percent who misses more putts.

 

Really loved this reply. 

I like how your argument is now a thought experiment.  This is a good admission, because thought experiments are useful, but they don't prove anything, and there are lots of statements here that "bumpy greens close the gap between good putters and bad putters" as some kind of proven indisputable fact.  Thought experiments don't prove things.  Especially thought experiments without any data.  

I also like that you are allowed to use grand generalizations for your thought experiment, but if I try and state an assumption for my model, I don't have the data and I'm making it too complicated?  Really? 

You say "you can not model the world without data, which you do not have".  Yet I did use data.  I used The statistics of make rates for professionals and amateurs.  I used a normal distribution, which has been proven to map to many many process in the real world, and its what Mark Broadie used for his study.  I used the size of the hole and reduced it for the effective size of the hole based on actual data. I used statistics to map the normal distribution to the players shot pattern so that I wasn't just imagining a player, but basing the model on data from collected thousands of actual players.

What data did you use to model the world?  It doesn't look like any to me, but please correct this if I missed the data you are using.  What assumptions are you making with your thought experiment?  

On the point on the OP's original question, he askes.  "would that narrow or increase the gap (or keep it the same) between the good putters and the bad putters? " Was I wrong to think that putting included the second putt and third putt?  Isn't that part of what every shot counts is about?  You have to think about every shot, not just the first one.  Shouldn't a thought experiment be created to address the full question, and not just half of it?

[phillyk]

2 hours ago, batchvt said:

This first putter who is perfect doesn't exist in human form.

 

I don’t know, Tiger had some tournaments where he made nearly every putt inside 8ft. There are really good putters out there. Now imagine they play Chambers Bay in the 2015 us open. No one came close to making every putt inside 8ft. An average golfer is going to still miss some but his percentage make would likely not change as much. 

[batchvt]

Nearly every reply to me on here:

"You are making up your own assumptions without any data to prove your point.  You don't have any data or proof those assumptions and constraints are correct, so you conclusions are wrong and unfounded.  Thats not science or logical.  

To show you why your logic is so bad, here are my assumptions and constraints that I just made up and that don't have any data to support them either, but they prove my original point so this is science and they show that I'm correct......if you can't see this you don't know what you are talking about."

Good times.  

 

Edited October 4, 2021 by batchvt

[boogielicious]

6 minutes ago, batchvt said:

Really loved this reply. 

I like how your argument is now a thought experiment.  This is a good admission, because thought experiments are useful, but they don't prove anything, and there are lots of statements here that "bumpy greens close the gap between good putters and bad putters" as some kind of proven indisputable fact.  Thought experiments don't prove things.  Especially thought experiments without any data.  

I also like that you are allowed to use grand generalizations for your thought experiment, but if I try and state an assumption for my model, I don't have the data and I'm making it too complicated?  Really? 

You say "you can not model the world without data, which you do not have".  Yet I did use data.  I used The statistics of make rates for professionals and amateurs.  I used a normal distribution, which has been proven to map to many many process in the real world, and its what Mark Broadie used for his study.  I used the size of the hole and reduced it for the effective size of the hole based on actual data. I used statistics to map the normal distribution to the players shot pattern so that I wasn't just imagining a player, but basing the model on data from collected thousands of actual players.

What data did you use to model the world?  It doesn't look like any to me, but please correct this if I missed the data you are using.  What assumptions are you making with your thought experiment?  

On the point on the OP's original question, he askes.  "would that narrow or increase the gap (or keep it the same) between the good putters and the bad putters? " Was I wrong to think that putting included the second putt and third putt?  Isn't that part of what every shot counts is about?  You have to think about every shot, not just the first one.  Shouldn't a thought experiment be created to address the full question, and not just half of it?

I didn’t claim to model the real world, you did. I was simplifying the concept for you so you would better understand. But you still do not. Show us the data for all putters, pro and amateur alike that proves misses are a normal distribution? I don’t have that data and therefore did not try. You do not have that data either. You cite Broadie, but there is no actual reference to the proof of normal distribution. Just your word.  So your first assumption is suspect. Then you add all sorts of scenarios, without data or proof to support your assertion.

Thought experiments are just that, experiments to visualize a situation. Try one using your assertions, like where  every putt that is bumped 1” will magically still go in. You confound your thought process by adding what ifs and are not looking at the question in its simplest form. If you introduce randomness, both putters will putt worse and mathematically it is more likely the gap will narrow.

[batchvt]

21 minutes ago, phillyk said:

I don’t know, Tiger had some tournaments where he made nearly every putt inside 8ft. There are really good putters out there. Now imagine they play Chambers Bay in the 2015 us open. No one came close to making every putt inside 8ft. An average golfer is going to still miss some but his percentage make would likely not change as much. 

It may be a coincidence that the best putter from the mid 10's won that tournament on notorious bumpy greens.  Or maybe it's not a coincidence.  

[iacas]

2 minutes ago, batchvt said:

It may be a coincidence that the best putter from the mid 10's won that tournament on notorious bumpy greens.  Or maybe it's not a coincidence.  

Aaron Baddeley won at Chambers Bay?

[batchvt]

20 minutes ago, iacas said:

Aaron Baddeley won at Chambers Bay?

He wasn't in the field. 

You don't think Jordan Spieth was one of the best, if not the best putter in the mid 10's?

36 minutes ago, boogielicious said:

You cite Broadie, but there is no actual reference to the proof of normal distribution. Just your word.  So your first assumption is suspect. Then you add all sorts of scenarios, without data or proof to support your assertion.

 

I linked to the Boadie paper here. He states he uses a normal distribution for directional errors in strike.  If you think a normal distribution is a bad assumption, what distribution do you think is correct?

15 hours ago, batchvt said:

Mark Broadie used a normal distribution for the trajectory of the putt in this paper.   He was trying to figure how how many more putts someone would make on a larger hole.  

(PDF) A simulation model to analyze the impact of hole size on putting in golf

---

PDF | We develop a model of golfer putting skill and combine it with physics-based putt trajectory and holeout models to study the impact of doubling... | Find, read and cite all the research you need on ResearchGate

The normal distribution shows up in a lot of places, and it does seem somewhat reasonable to use here.   Plinko, interestingly enough creates a normal distribution, so at a minimum it addresses @iacas plinko quip.  

 

41 minutes ago, boogielicious said:

Show us the data for all putters, pro and amateur alike that proves misses are a normal distribution? I don’t have that data and therefore did not try.

Except you did try.  You assumed the pro putter didn't miss the line at all from 10 ft, and had no distribution.  Then you assumed the bad player only made 1 out of ever 10 putts from 10ft, and often missed by 2-3ft. Those are both distributions. Did you have data for those distribution, or a reputable source that has used a distribution like those in the past to model putting?

 

[woodzie264]

14 minutes ago, batchvt said:

He wasn't in the field. 

You don't think Jordan Spieth was one of the best, if not the best putter in the mid 10's?

Speith was #9 one year and #2 one year but was outside the Top 10 in SGP the rest of the mid-10’s. No, he was not the best putter in the mid-10’s.

[Vinsk]

20 minutes ago, batchvt said:

You don't think Jordan Spieth was one of the best, if not the best putter in the mid 10's?

Strange comment from someone so indulged in statistics. Surely you know how easy that is to look up. What you see is what you see. Numbers don’t lie.

[boogielicious]

11 minutes ago, batchvt said:

He wasn't in the field. 

You don't think Jordan Spieth was one of the best, if not the best putter in the mid 10's?

I linked to the Boadie paper here. He states he uses a normal distribution for directional errors in strike.  If you think a normal distribution is a bad assumption, what distribution do you think is correct?

 

Except you did try.  You assumed the pro putter didn't miss the line at all from 10 ft, and had no distribution.  Then you assumed the bad player only made 1 out of ever 10 putts from 10ft, and often missed by 2-3ft. Those are both distributions. Did you have data for those distribution, or a reputable source that has used a distribution like those in the past to model putting?

 

Broadie’s paper is about strike directional errors for capture speed WRT hole size. No where does he state that the ball will end up in a normal distribution from the hole due to green reading errors, speed calculations or technique for good versus bad putters. The directional errors are in a normal distribution, not the resulting end location of the putt. He also does not discuss increase randomness of the putting surface. So it does not apply to this particular discussion. 

You again are missing the point. I was not claiming in any way that what you are stating I said. I did not “try” to create false data. I gave you opposite ends of a particular situation to demonstrate the simple math. You are purposely confounding what others are telling you. You keep making incorrect statements as absolutes and trying to do data analysis where you have no actual data for this scenario.

This is very tiring because you are not discussing, your are just being contrary.

[KMP]

Interesting that Broadie chose a normal distribution, which is understandable if data are not available. How would you handle the golfers that may miss more often to one side or another? This might be especially true on a sidehill putt. The distribution in this case would seem to be skewed to the side of the predominant miss, much like a "pull hitter" in baseball. Perhaps better putters would tend towards a tighter, more "normal" distribution; while for really bad putters the distribution could vary with the individual and differ from normal. Even if normal the "poorer putter curve" would be much wider, with an low peak. 

[batchvt]

1 hour ago, KMP said:

Interesting that Broadie chose a normal distribution, which is understandable if data are not available. How would you handle the golfers that may miss more often to one side or another? This might be especially true on a sidehill putt. The distribution in this case would seem to be skewed to the side of the predominant miss, much like a "pull hitter" in baseball. Perhaps better putters would tend towards a tighter, more "normal" distribution; while for really bad putters the distribution could vary with the individual and differ from normal. Even if normal the "poorer putter curve" would be much wider, with an low peak. 

Thank you for saying that assuming a normal distribution is understandable.  

I think you bring up a good point about the skewedness of the distribution.  Some players don't have symmetrical distributions.  Now across a group of players this might get canceled out, but for individual players, it wouldn't.  I've said before that the answer to this question is "it depends" and certainly at a singular level, a certain bad players could be expected to outperform good players on bumpy greens, and certain bad players could be expected to underperform on bumpy greens, all depending on the typical distribution of thier putts.

I agree that better players would tend towards a tighter more normal distribution and the poorer putters would have a wider lower peaked curve.  Thats how I tried to model it.

Interestingly, Broadie didn't assume the player always aims at the hole and tried to optimize making the putt. He assumed the players were trying to putt optimally for total score, including 3 putts. With slopes, this could mean just what you said, the player doesn't aim at the hole but aims above the hole.  It feels like many on this thread have oversimplified putting to the point that they don't even think about times when players don't aim at the hole, since every example here seems to assume the player is trying to make every putt.

Its a very complicated problem. By my count, Broadie used 14 variables to model how putting would change with a larger hole in is paper.  14.  Complex problems don't lend themselves to the correct answer when simplified down to a single variable.  

 

[batchvt]

1 hour ago, Vinsk said:

Strange comment from someone so indulged in statistics. Surely you know how easy that is to look up. What you see is what you see. Numbers don’t lie.

If I had said, "statistically a top 10 putter that year that many think was the best putter of the era", would that have changed the likely hood of a coincidence that this player won on unusually bumpy greens? 

"Great putter wins on controversially bumpy greens." vs best putter probably would have been a better choice.

[woodzie264]

15 minutes ago, batchvt said:

If I had said, "statistically a top 10 putter that year that many think was the best putter of the era", would that have changed the likely hood of a coincidence that this player won on unusually bumpy greens? 

 

It would’ve only demonstrated that those peoples’ opinions did not align with fact

[saevel25]

50 minutes ago, batchvt said:

I agree that better players would tend towards a tighter more normal distribution and the poorer putters would have a wider lower peaked curve.  Thats how I tried to model it.

It would be hard to assume this. 

1. Does reading a green follow a normal distribution curve? I highly doubt it.
2. Does hitting putts a certain distance follow a normal distribution curve, maybe. 
3. Does hitting putts on the line you want follow a normal distribution curve, maybe.  
4. Does lining up to the line you want follow a normal distribution curve, not entirely sure. 

Ok, so all these things are things a person can control, are skill based. 

So now, take all of these, put them into an equation, and ask does that create a normal distribution curve. This is the issue I have with this line of thought on normal distribution curves. There is a lot that go into putting, and not all of it is normally distributed. 

THEN! add in green imperfections, which basically throws the normal distribution out the window.