Putting Evaluation

[Howling Coyote]

About 6 months ago it seemed like my putting was getting worse.  I began 3-putting, 4-putting and even 5-putt on occasion.  I went over some old score sheets and could see that I used to be a much better putter.

 

So what happened?

 

Before my tee shot was really bad, forcing a pitch up to the green and resulting in very short putts (of all the elements in my game, short pitching is very strong).  However, as my tee shots got better I would hit the greens but then would be left with longer putts.  Since I did not have experience with longer putts I would 3-putt.  Then I would psych myself out, convince myself that I was a lousy putter, and this lead to even more bad putts—a vicious cycle.

 

Well before I figured the above out, I was seriously thinking about buying a new putter—to fix my putting woes.  But then I realized the big question, “How would I know if my putting was getting better?”  I looked at strokes gained putting, the stat used by the PGA.  When I finally understood what that was about it was disappointing, because that stat really doesn’t tell the whole story.  So I invented my own putting evaluation technique, which today I want to share.

 

Method

For each hole measure the distance to the pin after the ball first lands on the green.  Don’t make additional measurements as only this initial distance matters.  Then count how many putts are made before you hole out.  Repeat the above for each hole.  At the end of the round total the distances (Dt) and total the putts (Pt).  Plot these on a graph (see figure 1).

Figure 1 shows data from playing on a 9-hole course (using Microsoft Excel which has a fit feature).  There are 2 points, one point for each round played.

Assuming one is not trying to alter their putting technique, not trying out a new putter, and one plays the same course day after day, then the points should roughly fall on a line* (+/- some delta value).  If your putting is improving the points should fall below the line.

One caveat, if one aces a hole or pitches the ball into the hole without any putting, then this is like playing a different course one with 8 holes (or 17 holes).  One can perform simple linear interpolation to correct this.  However, be aware if the data lost was on the courses most difficult or easiest green, then that might not be the most realistic comparison.

Consider using this method to evaluate different putting techniques, a new putter, etc.  Any new improvement should make the next point land well below the line.

*When I first began doing this, the assumption was the points would fall on a curve but over many games the equation turned out to be a line.  The data always had a curve-fit with an offset that approached the minimum number of putts.

Edited March 11, 2016 by Howling Coyote
Move figure inline

[Mike Boatright]

Hmm well I'll be honest sir I didn't read the entire post most the graph party sorry. Putting is an art you have to evaluate the green speed slope and gravitational forces some how they apply and it's a mystery. Anyway putting is truly an art you have a clubhead with an arc that only travels a small distance so line shouldn't really be an issue so it comes down to applied forces in relation to the line slope grain and again gravity...

[iacas]

Why not use actual strokes gained putting stats and calculate it that way? The way you have it, you could hole a 35-foot putt and three-putt from five feet and that would show up the same as two-putting twice from 20 feet.

[Howling Coyote]

iacas

As I said in my original post, I do not feel that tells the whole story.

This next chart is from data taken from R3 of the Phoenix Waste Management Open (2016).  I have data from 6 players and of these 6, Rickie Fowler had the best strokes gained during (being +1.4) putting while Bubba Watson had a negative strokes gained during putting (I think it was like -0.4).  But watching the event, just my lay opinion, it was Phil Mickelson whose putting was awesome.  So I plotted the data using my method ( see figure).

 

H. Matsuyama ended winning the event.  I also notice that most players were very consistently hitting about 400 feet of putting.  But unusual outliers  were Brandt Snedeker and Matsuyama.  This graph clearly shows that Phil's putting was superior and it was not just my imagination.  Also that Watson and Fowler's putting were about the same.

 

Edited March 14, 2016 by Howling Coyote
Added another paragraph

[iacas]

1 hour ago, Howling Coyote said:

As I said in my original post, I do not feel that tells the whole story.

What doesn't? What is "that" in the sentence quoted here?

1 hour ago, Howling Coyote said:

This next chart is from data taken from R3 of the Phoenix Waste Management Open (2016).  I have data from 6 players and of these 6, Rickie Fowler had the best strokes gained during (being +1.4) putting while Bubba Watson had a negative strokes gained during putting (I think it was like -0.4).  But watching the event, just my lay opinion, it was Phil Mickelson whose putting was awesome.  So I plotted the data using my method ( see figure).

Since I know how Strokes Gained Putting is calculated (it's pretty simple), I'm going to go along and say that I think Rickie putted much better than Bubba during that round, and if phil was in the middle, he was in the middle.

Again, from 35' and 5' if a pro 3 putts and 1-putts, but another pro two-putts from 20' twice… the strokes gained would be:

35'@3/5'@1: 2.02 + 1.23 = 3.25 putts, so 0.75 putts lost.
20'@2 x 2: 1.87 + 1.87 = 3.74, so 0.26 putts lost.

In other words, you're making something which is not really all that linear (total putt distance divided by the number of putts), when the simple two-putt example I gave you here shows a whopping half-stroke difference despite both being 40 feet and 4 putts.

Given the choice between your eyes lying to you and this particular statistic lying to you, I'm going with the former.

[Mike Boatright]

This looks like an over complicated mathematical equation just saying.. 

 

 

[iacas]

10 hours ago, Mike Boatright said:

This looks like an over complicated mathematical equation just saying.. 

What's your point?

[ZappyAd]

 

15 hours ago, iacas said:

… the strokes gained would be:

35'@3/5'@1: 2.02 + 1.23 = 3.25 putts, so 0.75 putts lost.

Is there an up to date source of strokes gained putting?  I have seen some data from 2010 but wondered if it is updated annually by the PGA?

[iacas]

16 minutes ago, ZappyAd said:

Is there an up to date source of strokes gained putting?  I have seen some data from 2010 but wondered if it is updated annually by the PGA?

I extrapolated from ESC, which is more recent than 2010, yeah.

[Howling Coyote]

21 hours ago, iacas said:

In other words, you're making something which is not really all that linear (total putt distance divided by the number of putts), when the simple two-putt example I gave you here shows a whopping half-stroke difference despite both being 40 feet and 4 putts.

I am not making it linear.  I have about 18 points now on my own plot and decided to let the data tell me what is going on.  As I stated in the first post, I thought it would be a curve, but as I collect more points, for me, for my personal data it is looking like a line.  I don't know why this is so.  So that is why I choose to do a curve fit to a line.

[Howling Coyote]

6 hours ago, iacas said:

I see the value in using this, but I have 2 comments.

(1) First I thought these numbers were unique to each golf course depending on the green speed.  That is part of why I didn't use it. I do not have numbers for the golf course I play on.

(2) I am very good with my short pitching.  Let's just say a player like me frequently misses the green on a certain par 3 course, but then pitches and always lands the ball within 5 feet of the hole.  This player then 1 putts maybe 90% of the time, resulting a slightly positive strokes gained in putting.  Then that player takes lessons from a tour pro and gets more distance and more accuracy from the tee.  He now hits the green in regulation, but now is always about 20 feet (or more) from the pin.  He then 2 putts these resulting in a slight negative strokes gained in putting.  So on paper, it looks like his putting got worse, but really his tee shot got better.  In both case he was mostly par'ing the holes, so his overall score is the same.

[saevel25]

7 minutes ago, Howling Coyote said:

I see the value in using this, but I have 2 comments.

(1) First I thought these numbers were unique to each golf course depending on the green speed.  That is part of why I didn't use it. I do not have numbers for the golf course I play on.

Not really. Distance is the primary in reality the sole indicator of putts probability. Speed of the greens has little to do with it unless you are on extremely slow greens, like in the realm of 7ish. If you are on very fast greens like 13+ 

I would say proximity to the hole is 99% the determining factor on putts made, besides the situations where you are on tier and the slope and speed make the putt impossible to make. That is a very rare occurrence when you consider how many putts you hit throughout a season. 

10 minutes ago, Howling Coyote said:

So on paper, it looks like his putting got worse, but really his tee shot got better.  In both case he was mostly par'ing the holes, so his overall score is the same.

Doesn't really work that way. The odds of having a 90% short game is absurd. PGA tour players maybe get near 75% on the high end. Most of the time they are in the 60% range. A 50% short game is pretty solid for most people. 

To average 90% from 5 feet would be better than a PGA tour player, who one putt 77% of the time. It's an unlikely scenario. 

GIR is king. In the end you are giving yourself more opportunities to one putt. You have to look at this in terms of averages. Even if you average 30 feet to the hole on approach shots. You one putt at a higher rate than you would chip in from that distance. So you get more birdies. You're also less likely to three putt than let's say leave an extra shot in the bunker or duff a chip. You are gaining strokes there. Also, there will be situations where you might stick one close. That alone can give you a huge strokes advantage. 

Just ask yourself this. I get 100 putts from 35 feet. You get 100 short game shots from 10 yards off the green. Who do you think wins in lowest scores? The guy who hit's it to 35 feet. 

Could the strokes gained putting lower, yea. It could also mean it gets a dramatic boost because if you're putting gets hot with a lot of high GIR's then you are making more putts at longer distance. That quickly adds up. 

One putt made from 20 feet gives you. +0.87 strokes gained. 

[iacas]

19 minutes ago, Howling Coyote said:

(2) I am very good with my short pitching.  Let's just say a player like me frequently misses the green on a certain par 3 course, but then pitches and always lands the ball within 5 feet of the hole.  This player then 1 putts maybe 90% of the time, resulting a slightly positive strokes gained in putting.  Then that player takes lessons from a tour pro and gets more distance and more accuracy from the tee.  He now hits the green in regulation, but now is always about 20 feet (or more) from the pin.  He then 2 putts these resulting in a slight negative strokes gained in putting.  So on paper, it looks like his putting got worse, but really his tee shot got better.  In both case he was mostly par'ing the holes, so his overall score is the same.

That player's putting did get worse.

If he makes 90% of his five footers (i.e. he averages 1.1 putts from five feet), and he has 18 of them per round, he's gaining 0.13 * 18 = 2.34 strokes per round making 90% of the five-footers per round. If he always two-putts from 20 feet, he's losing 0.13 strokes per hole, or the same 2.34 strokes per round.

He should make about 13% of those twenty footers, or perhaps more since he's SO good from five feet, he must be reasonably good at hitting putts online, reading putts, and controlling the distance.

My point remains that you're converting putt distance by feet and that doesn't really work. The chart above (which isn't really going to change much based on green speed) looks a bit like this:

[onthehunt526]

18 minutes ago, iacas said:

That player's putting did get worse.

If he makes 90% of his five footers (i.e. he averages 1.1 putts from five feet), and he has 18 of them per round, he's gaining 0.13 * 18 = 2.34 strokes per round making 90% of the five-footers per round. If he always two-putts from 20 feet, he's losing 0.13 strokes per hole, or the same 2.34 strokes per round.

He should make about 13% of those twenty footers, or perhaps more since he's SO good from five feet, he must be reasonably good at hitting putts online, reading putts, and controlling the distance.

My point remains that you're converting putt distance by feet and that doesn't really work. The chart above (which isn't really going to change much based on green speed) looks a bit like this:

I have a question, @iacas. Is that why when the tour introduced Strokes Gained Putting, someone could have 26 putts for the round and lose strokes on the greens and someone could have 32 putts and actually gain strokes putting?

Because I'm led to believe strokes gained putting has nothing to do with the total number of putts in a round, but the number of putts on a given hole from a certain distance vs tour average.

Edited March 16, 2016 by onthehunt526

[iacas]

16 minutes ago, onthehunt526 said:

I have a question, @iacas. Is that why when the tour introduced Strokes Gained Putting, someone could have 26 putts for the round and lose strokes on the greens and someone could have 32 putts and actually gain strokes putting?

Because I'm led to believe strokes gained putting has nothing to do with the total number of putts in a round, but the number of putts on a given hole from a certain distance vs tour average.

Yes.

Here's something from ESC:

Quote

In the opening round of the 2011 Frys.com Open, Tiger Woods carded a 73, taking 27 putts. To put this into perspective, PGA Tour pros average 29 putts per round. The 2011 leader in putts per round was Kevin Na, with an average of 27.8 putts per round. Tiger’s 27 was nearly one less than the tour-leading average. Tiger had no three-putts on that day. He one-putted nine times and two-putted nine times. After the round, Tiger rated his putting performance in an interview. “That’s probably one of the worst putting rounds I’ve ever had,” he said. “I can’t putt the ball any worse than I did today. I just had a hard time hitting my stroke, and then I started altering it. And my stroke got worse . . . and I started losing confidence in it because I wasn’t hitting my line. So it was just a downward spiral.”

He lost 1.3 strokes because his putts came from a distance of 11 feet while the PGA Tour average is 20 feet (or something like that).

[onthehunt526]

So I'm gathering the guys who shoot 69 have 23 putts, but scramble well are probably not gaining very many strokes putting? And then there are guys who shoot 69 hit every GIR have 33 putts and might be gaining strokes putting?

[iacas]

17 minutes ago, onthehunt526 said:

So I'm gathering the guys who shoot 69 have 23 putts, but scramble well are probably not gaining very many strokes putting? And then there are guys who shoot 69 hit every GIR have 33 putts and might be gaining strokes putting?

Depends.

You can scramble well by hitting your short game shots to ten feet and making a lot of them or by hitting them to three feet and making most of them. You'd get very different strokes gained putting in those cases.

[Howling Coyote]

hmmm

First, iacas, maybe I don't have enough data at short distances.  Maybe it would be a curve if I had the data.

Second, Tiger had only 207 feet of putting which is extremely small.  He also 2 putted both a 3 footer and a 4 footer, something only I would do.  So even plotting that on my pro chart, his putting would be worse than Brandt Snedeker.  Matsuyama on the other hand had one 145 foot putt which he 3-putted, but he still par'ed the hole.

In spite of all that if 3 golfer's played 2 holes

player 1, 2 putts a 20 footer and then 2 putts another 20ft.

player 2, 3 putts a 35 footer and then 1 putts a 5ft.

player 3, 2 putts a 10 footer and then 2 putts a 30ft.

They all covered 40 feet in putt distance in the same number of strokes.  I might call player 2 inconsistent or streaky, but they all covered the same distance.

My main issue is statistical (mathematical).  For example at 8 feet it is suppose to take 1.5 strokes, but I can't sink it into the hole in 1.5 strokes.  It will either take me 2 putts or 1 putt.  Therefore on any given day, just by pure luck, the putts could break against me on one day and in my favor another day, and perhaps 18 holes is just too small a sample size.