Connect with us

Opinion & Analysis

What factors are most important in Strokes Gained Putting?

Published

on

As a golf statistician, I’m often asked what statistics golfers can keep to track their skills on the golf course.

For the tour players I work with, it’s easy. The PGA Tour uses a laser-measuring system called “Shot Tracker” that gives me the raw data I need to help them with their game. For amateurs, it’s much more difficult. They don’t have Shot Tracker, and many golfers who create their own metrics find that it can be a time-consuming process riddled with inaccuracies.

I am constantly looking to create new types of scoring games that are based on sound statistical information. For instance, there was a fad of people using the total distance of putts made metric. The idea was this — the longer the distance of total putts a player made, the better they putted. It does not consider, however, the golfer who putts poorly and happens to make a 60-footer. Nor does it consider the golfer who putts well and leaves himself first putts that are shorter in distance.

I started to look at some other metrics that I thought would be less cumbersome to record when it came to putting. I ran this against the past history of Strokes Gained Putting on Tour and found some interesting results.

When using putting distance metrics, the average length of all two-putts has the strongest correlation to Strokes Gained Putting on Tour.

For instance, if Brandt Snedeker two putts from 30 feet, it is recorded as a two putt from 30 feet. So, if he gets on the next hole and two putts from 50 feet, then he has an average of two putting from 40 feet.

This finding surprised me. I would have thought that the average length of a one-putt or the average length of a three-putt would have had a larger correlation to Strokes Gained Putting.

In fact, here are the mathematical correlations over the years since Strokes Gained Putting has been used by the PGA Tour:

Screen Shot 2015-02-09 at 12.08.34 PM

For those who don’t understand what the numbers mean, it is based on a mathematical formula used to tell the strength of a relationship between two variables. The correlation number ranges from +1 to -1. The closer the number is to +1, the stronger the direct relationship is between the two variables.

Let’s say I own a store and I want to see the relationship between lemonade sold and temperature. After I record the data, I find the correlation to be +0.9. That means that as the temperature increases so does the likelihood that I will sell more lemonade. A number closer to 0 means that there is no real relationship, and a number closer to -1 indicates a stronger, indirect relationship.

Say I owned the same store and wanted to look at hot soup sold and temperature. I may get a correlation of -0.85, which means that as the temperature falls it would be likely that I would sell more soup. But if I look at the correlation between temperature and bread sold and come up with a correlation of +0.003, then that shows that temperature has no real impact on how many loafs of bread will be sold.

The reason why I bring up these correlations is that we see that after the two-putt distance correlations, the one-putt distance correlations are right behind. There is a significant decline, however, in the correlation between three-putt distances and Strokes Gained Putting.

My conclusion as to why the two-putting distance metrics have a stronger correlation is that it is simply too difficult to putt great from course to course. I label putting “great” any time a Tour player averages +1 or more strokes gained per round for a tournament. Even the best putters on Tour tend to only putt great in roughly 30-to-40 percent of their events.

These were three of the best putters in 2014 and they putted great in 33-to-40 percent of their events.

Screen Shot 2015-02-09 at 12.08.47 PM

Screen Shot 2015-02-09 at 12.08.57 PM

Screen Shot 2015-02-09 at 12.09.12 PM

By looking at the graphs, we can see that great putting tends to occur at roughly the same time. Aaron Baddeley hit a stretch of great putting from The McGladreys through Riviera, as he putted great in 4 of those 6 events (notice the tall red bars, which indicate strokes gained on the field).

Freddie Jacobson hit his hot streak by putting great in 6 out of 8 events from Pebble Beach to the Zurich Classic. And Greg Chalmers started the year on fire by putting great in four events in a from The McGladreys to TPC Scottsdale.

Great putters also tend to cool off in streaks, however. They usually tend to putt better than the average (better than 0 strokes gained), but they can no longer rely on great putting for them to do well in a tournament. And that is where I believe why the average length of a player’s two-putts correlates strongest to Strokes Gained Putting.

But, I didn’t want to stop there. I wanted to look at a combination of metrics to help decide if there is more than one way to correlate to Strokes Gained Putting. And what I found was that if you combine the length of the average two putt along with the length of the average birdie putt made, you get the strongest correlation to Strokes Gained Putting in the end.

Let’s remember how Strokes Gained Putting is designed. If the field makes 40 percent of their putts from 10 feet, that will equate to roughly 1.6 strokes to the hole from 10 feet. If a player one-putts from 10 feet, then they are determined to have gained +0.6 strokes on the field. If they two-putt from 10 feet, then they have lost -0.4 strokes to the field.

The reason why birdie putts contribute to Strokes Gained Putting more is that they are by and large more difficult for players to make. In fact, only two PGA Tour players had a higher make percentage of birdie putts than par putts from 5-to-15 feet: Kevin Streelman and Luke Donald.

Anytime the make percentage goes down, there is more potential for a player to gain more strokes. If the make percentage from 10 feet went down to 20 percent and that roughly equates to 1.8 strokes from 10 feet, a player that makes the 10-foot putt now gains 0.8 strokes on the field.

Essentially, what this tells me is that even the best putters are not going to be able to take great putting from course to course, so they need to be two-putting when putts are not dropping for them. And when you combine the length of the two-putts along with the length of birdie putts made, it provides a measurement of ability to avoid three-putts and ability to make putts.

The average Tour player’s two-putt distance is roughly 22.5 feet, and their average birdie putt make distance is 10.5 feet. That comes to a combined 33 feet. I also wanted to test this with another metric that is fairly easy to measure: Greens in Regulation (GIR).

When combining GIR with the “Average All Two-Putts Distance” and the “Average Birdie Make Putt Distance,” it has an incredibly strong correlation to Adjusted Scoring Average.

Normally, I’m not a big fan of the greens in regulation metric due to its flaws and ambiguous nature, but considering that it is an easy metric to calculate and the correlation between greens hit and the two-putt and one-putt average distance is so high, I think it is a helpful way for a player to measure their skills.

And that’s how I came up with the “98 Score.”

Here are the Tour averages for each of the metrics:

  • Average All 2-Putts Distance: 22.5 feet
  • Average Birdie Putts Made Distance: 10.5 feet
  • Average Greens In Regulation: 65 percent

22.5 + 10.5 + 65 = 98

If you can score 98 or better, than you are playing quite well. If not, keep trying to improve your score so you can get the ball on the green more often and putt better when you’re on the greens.

Your Reaction?
  • 69
  • LEGIT10
  • WOW8
  • LOL2
  • IDHT1
  • FLOP11
  • OB2
  • SHANK7

Richie Hunt is a statistician whose clients include PGA Tour players, their caddies and instructors in order to more accurately assess their games. He is also the author of the recently published e-book, 2018 Pro Golf Synopsis; the Moneyball Approach to the Game of Golf. He can be reached at ProGolfSynopsis@yahoo.com or on Twitter @Richie3Jack. GolfWRX Writer of the Month: March 2014 Purchase 2017 Pro Golf Synopsis E-book for $10

16 Comments

16 Comments

  1. Trae

    Feb 12, 2015 at 7:44 pm

    This was a good read, and very interesting. And it brings back memories of stats class, and also me spending countless hours entering in data from the pga tour just to see relationships between variables and trying to create a number of statistical models to predict, etc. I’m never doing that again! Never got to a point where I could really apply it to my own game in a really effective way though. Always would end up getting caught up in trying to figure out why someone was an outlier, then later on lost interest in stats altogether. It’s interesting stuff but I think it can be very misleading in some cases. It can be tricky. Anyways, I enjoyed this article and I’m looking forward to your next one. Thanks.

  2. Jeff Watson

    Feb 11, 2015 at 4:32 pm

    Rich,

    I like it. I already pace off my putts and number of putts for a stats program I subscribe to. Any idea what the “Score” would be based on handicap?

    Jeff

  3. david

    Feb 11, 2015 at 12:54 pm

    Hi Ritch, I had to drop out of gread 11 math, so unfortunately my brain is still bleeding after reading your article. I know the tour putting stats on makeability is, from what I believe, from 6 feet to about 50%. I was a bit under this, so I worked hard on my putting last summer and lowered my cap from aprox 5 to 3.7. I know without keeping stats that not enough of my greenside chips get to inside 4 feet, so that is my focus this summer to knock off another 2 points. But I did find the article interesting, but not necessarily relevant to most amateurs, whatever the index.

  4. Dick Kusleika

    Feb 11, 2015 at 11:13 am

    I step off every first putt. I don’t know how people putt without stepping it off. They just guess or try to feel it, I suppose.

    I think the negative correlation is “inverse relationship”, not “indirect relationship”.

  5. Hudson

    Feb 10, 2015 at 2:21 pm

    I calculate the strokes gained (lost in my case, lol) and measure one pace (3 feet) to get the distance on the green from the pin. It takes a few seconds and it’s dead easy.

    I am going full speed to track/calculate my strokes gained now and I am explaining it there:

    http://golf-made-in-us.blogspot.com/2014/04/golf-stats-revolution-gained-strokes.html

    I consequently discovered that I was losing 6 strokes between 125 and 200 yards (on a 14 hcp) on average in 2014 and will work on that !

    • Rich Hunt

      Feb 10, 2015 at 2:35 pm

      Hudson, I’ve enjoyed your work. The only issue I have with using strokes gained with putting is that the make %’s at a local course are likely different then at a Tour level course, even for Tour players. As we’ve seen time and time again, slower greens mean lower make %’s for players across the board. I’ve not only seen it at Tour courses that I record the data for, but David Orr did a study on this and found the same thing as well. IMO, it is probably best to use strokes gained in comparing it to your own performance over time and see how well you improve or regress.

      • HIGHfader

        Feb 10, 2015 at 4:49 pm

        Richie, do you know where I can find that David Orr studdy?

        • Richie Hunt

          Feb 10, 2015 at 8:13 pm

          I don’t believe David has made it public. But, if you asked him that is what he would tell you. IIRC, they measured roughly 700 golfers and were on greens from the local muni to Pinehurst.

  6. jeff

    Feb 10, 2015 at 1:23 pm

    Rich, great writeup, thanks. But without data from our peers (“the field”), how is this different from simply looking at your actual score for the round as a basis for judging our “getting better”? What I mean is, what will achieving a 98 in this fashion tell us about our game that the score cannot? If the result of putting better (fewer putts per round) is a better overall score, isn’t that pretty much the same thing? Without a field to compare against, how much would we actually get out of this? Thanks again!

    • Rich Hunt

      Feb 10, 2015 at 2:18 pm

      Jeff,

      Very good point. I think we can use the 98 Score to measure how well we are striking the ball and putting. So the other factors that would cause a discrepancy in 98 Score versus Actual Score would be short game shots around the green, penalty shots and general strategy. So if the score is better than 98, but a player shoots 75, then they may have struggled with short game or hit a shot OB or in the drink.

      In my next article I will go into simple metrics I keep for my rounds that will better help determine the strengths and weaknesses of our game.

  7. myron miller

    Feb 10, 2015 at 12:40 pm

    I’m a little sceptical of your usage of the statistical correlation coefficient. Famous statistical anolmaly: Amount of ice cream eaten versus number of polio cases was fantastically high (close to .9 or something like that). Ergo if you eat a lot of Ice cream your chances of polio increased. Just because there’s a correlation doesn’t mean that there is necessarily a relationship. Usually but not necessarily. Other factors could be involved that sway the correlation coefficient.

    For the amateur to actually pace off every putt regardless does help but slows down the game a bit increasing the time for the round. Not a productive way to help the overall pace of play but it does help the individual golfer (assuming his paces are moderately accurate – which is sometimes questionable). So which is more important, pace of play or helping the individual in play? At which time?

    • Rich Hunt

      Feb 10, 2015 at 2:14 pm

      Correlation does not imply causality (unless there is a correlation of a perfect +1 or -1). The issue is that people tend to take that the wrong way as correlation is trying to define the symmetry of a relationship.

      The problem with the ice cream to polio correlation is that usually those types of analysis are done over a very short period of time so the sample size is not large enough. The other issue is that the variables (polio and ice cream consumption) have virtually nothing to do with each other.

      Correlation is not about certainty, it’s about probability. The real issue with the correlations I used is that they are not ‘strong correlations’ (+0.5 or better), so the 1:1 relationship is not exactly a confident one. But, that could be due to a myriad of factors like a golfer’s ballstriking.

      However, those correlations were fairly consistent over the seasons and on a comparative basis I think it indicates that all 2-putts length does give a better indication of how a player is putting than say 3-putt length. And given that when we combine all 2-putts and birdie putts lengths does have a strong correlation, it’s a good way to measure our true putting prowess.

      Lastly, the main point of the article was to more accurately measure our own putting prowess and do it without the measuring process being too cumbersome and time consuming. Obviously, we don’t know how well we are putting versus other golfers. But we can measure our own improvement over time with a large enough sample by seeing how well we improve compared to our performance in other rounds of golf.

    • Rich Hunt

      Feb 10, 2015 at 2:28 pm

      Myron,

      Correlation does not imply causality (unless the correlation is +1 or -1). However, many people don’t quite understand that phrase as used in the polio to ice cream consumption case.

      Correlation is designed to give a confidence level of the symmetrical relationship between 2 variables. For instance, we may see a correlation between ball speed and distance a ball travels of +0.9. That doesn’t consider launch angle, spin rate, landing angle of the ball, etc. But we have a high degree of confidence that ball speed will be a large determining factor in the distance the ball traveled.

      The problem with using the polio to ice cream consumption argument (and the countless other arguments) is that:

      1) typically the sample sizes are incredibly small so the data can be easily skewed.

      2) the variables have virtually nothing to do with each other.

      I feel the sample size of length of putt to putts gained is more than sufficient as it took place over several seasons with roughly 180 golfers per season. I also feel that the variables (length of a putt versus putts gained) are closely tied together.

      If there is a an issue with using the correlations is that none of these correlations were ‘strong’ (+0.5 or better). However, if you add all 2-putt length and birdie putt lengths, the correlation is strong.

      So, why is the correlation not at a perfect +1?

      Likely due to ballstriking. If a player hits a birdie putt to 4-feet and make the birdie putt, there’s not much they can do about it rather than the player that misses their 4-foot birdie putts, but makes a 20-footer. The same goes for 2-putting.

      Still, the main purpose of the article was to more easily determine how well you are putting and trying to improve upon your own performance. I think if one records this data over time they can use it to better gauge how well they have putted over a certain amount of time.

  8. Rob Rashell

    Feb 10, 2015 at 11:30 am

    Rich,

    This stuff is fantastic, will be sharing with the players I work with.

    Rob

  9. Rich Hunt

    Feb 10, 2015 at 10:42 am

    Steven,

    I would just recommend pacing off the initial putt. I personally pace off all of my putts except for putts inside 5-feet since I know how long they are by eye-balling it.

    It doesn’t have to be a perfect measurement because the idea is to improve your score so you can improve your putting. For instance, if you average 30-feet on your 2-putts by pacing it off and in reality it is 32 feet, that’s not as important as actually improving your 2-putt average the next time you play.

  10. steven

    Feb 10, 2015 at 10:24 am

    rich, how would you recommend us amateurs determine the distance of our 2 putt distances. As well our birdie putt distances. Of course theres no guarantee i make a birdie in a round of golf. but i sure as heck attempt and make a lot of 2 putts. without the use of a simulator, ill simply be estimating the distance i am away from the hole. thus limiting the reliability of an amateurs usage of your ’98’ theory. any tips? as the concept seems very well thought out, but the execution for an average player is really implausible. thanks.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Opinion & Analysis

An open letter to golf

Published

on

Dear golf,

I know it has been some time since we last spoke, but I need you to know I miss you, and I can’t wait to see you again.

It was just a few months ago I walked crowded isles, stood shoulder to shoulder, and talked endlessly with likeminded individuals about you and your promising future in 2020 at the PGA Show. At that time, the biggest concern in my life was whether I had packed the perfect dress-to-casual pant ratio and enough polos to get through the mayhem of six days in Orlando. Oh, how the times have changed.

On a professional level, what started with the LPGA Tour a few weeks prior progressed quickly at The Players Championship, when you ground to a complete halt within days. As much as it was a tough decision, it was the right decision, and I admire the judgment made by your leaders. Soon after, outside of the professional ranks followed suit and courses everywhere began shutting doors and asked golfers to keep away.

This is the right decision. For now and for the foreseeable future, as much as I don’t like it, I understand how important it is we let experienced health medical professionals make choices and craft policies for the wellbeing of people everywhere. Although, judging by the indoor short game trickery I have witnessed over the last 10 days, handicaps could be dropping when you finally return.

As a game, you are over 200 years old. You have survived pandemics, wars, depression, drought, and everything else that has been thrown at you. Much like the human spirit, you will continue on thanks to the stories and experiences others passed down and enjoyed.

I know you will survive because I also plan on surviving. As long as there are people willing to tend to your grounds and maintain your existence, I will also exist ready to take on your challenge.

When you are able to return in full, I will be here.

Sincerely,

Ryan Barath (on behalf of golfers everywhere)

 

Your Reaction?
  • 1
  • LEGIT0
  • WOW0
  • LOL1
  • IDHT0
  • FLOP0
  • OB0
  • SHANK2

Continue Reading

Opinion & Analysis

The Wedge Guy: Improving your short iron and wedge impact

Published

on

One of my most appreciated aspects of this nearly 40 years in the golf equipment industry is the practically endless stream of “ah ha” moments that I have experienced. One that I want to share with you today will–I hope–give you a similar “ah ha moment” and help you improve your ball striking with your high lofted short irons and wedges.

As I was growing up, we always heard the phrase, “thin to win” anytime we hit an iron shot a little on the skinny side (not a complete skull, mind you). When you caught that short iron or wedge shot a bit thin, it seemed you always got added distance, a lower trajectory and plenty of spin. It was in a testing session back in the early 2000s when this observation met with some prior learning, hence the “ah ha moment” for me.

I was in Fredericksburg, Virginia, testing some wedge prototypes with a fitter there who was one of the first to have a TrackMan to measure shot data. I had hit about two dozen full pitching wedges for him to get a base of data for me to work from. The average distance was 114 yards, with my typical higher ball flight than I like, generating an average of about 7,000 rpms of spin. What I noticed, however, was those few shots that I hit thin were launching noticeably lower, flying further and had considerably more spin. Hmmm.

So, I then started to intentionally try to pick the ball off the turf, my swing thought being to actually try to almost “blade” the shot. As I began to somewhat “perfect” this, I saw trajectories come down to where I’d really like them, distance increased to 118-120 and spin rates actually increased to about 8,000 rpms! I was taking no divot, or just brushing the grass after impact, but producing outstanding spin. On my very best couple of swings, distance with my pitching wedge was 120-122 with almost 10,000 rpms of spin! And a great trajectory.

So, I began to put two and two together, drawing on the lessons about gear effect that I had learned back in the 1980s when working with Joe Powell in the marketing of his awesome persimmon drivers. You all know that gear effect is what makes a heel hit curve/fade back toward the centerline, and a heel hit curves/draws back as well. The “ah ha” moment was realizing that this gear effect also worked vertically, so shots hit that low on the face “had no choice” but to fly lower, and take on more spin.

I had always noticed that tour players’ and better amateurs’ face wear pattern was much lower on the face than that of recreational golfers I had observed, so this helped explain the quality of ball flight and spin these elite players get with their wedges and short irons.

I share this with you because I know we all often misinterpret the snippets of advice we get from friends and other instructional content that is out there. To me, one of the most damaging is “hit down on the ball”. That is a relative truth, of course, but in my observation it has too many golfers attacking the ball with their short irons and wedges with a very steep angle of attack and gouging huge divots. The facts are that if the club is moving only slightly downward at impact, you will get the spin you want, and if the clubhead is moving on a rather shallow path, you will get a more direct blow to the back of the ball, better trajectory, more distance and improved spin. Besides, shallow divots are easier on the hands and joints.

If this is interesting to you, I suggest you go to the range and actually try to blade some wedge shots until you somewhat groove this shallower path through impact and a lower impact point on your clubface. As you learn to do this, you will be able to zero in on the proper impact that produces a very shallow divot, and a great looking shot.

[TIP: If you will focus on the front edge of the ball – the side closest to the target – it will help you achieve this kind of impact.]

It will take some time, but I believe this little “experiment” will give the same kind of “ah ha moment” it gave me.

Your Reaction?
  • 26
  • LEGIT7
  • WOW2
  • LOL3
  • IDHT0
  • FLOP0
  • OB0
  • SHANK1

Continue Reading

On Spec

On Spec: Interview with Trevor Immelman, 2008 Masters champion

Published

on

In this episode, host Ryan speaks with Trevor Immelman about his career, what it was like growing up around the game as a competitive amateur in South Africa, and what it’s like being a Masters champion.

Topics also include his experiences working with the design team at Nike Golf as well as his current “What’s in the Bag” which includes equipment from Titleist and the process he went through to get it dialed in.

Check out the full podcast on SoundCloud below, or click here to listen on iTunes or here to listen on Spotify.

Want more GolfWRX Radio? Check out our other shows (and the full archives for this show) below. 

Your Reaction?
  • 3
  • LEGIT0
  • WOW0
  • LOL0
  • IDHT0
  • FLOP0
  • OB0
  • SHANK0

Continue Reading

WITB

Facebook

Trending