The Importance of Putting is Vastly Overstated (and its Contribution to Winning on the PGA Tour)

[iacas]

I made a post here that prompted this thread:

That post was in turn prompted by the age-old "Drive for Show, Putt for Dough."

I wanted to cover what I said there in a bit more detail, because I think that the importance of putting is vastly over-stated for two reasons…

#1: Proximity to the Finality of the Hole's Completion

The last thing carries the most weight. You see this all the time in day-to-day life, and even more often in sports. If you wanted to bet on the Steelers to win the Super Bowl a few weeks ago you'd have gotten 5/2 odds. They'd just defeated the Denver Broncos, and people were high on their chances. They hadn't even secured a playoff spot, yet they were the third most favorite team to win the Super Bowl behind the Patriots and the Panthers.

Then they laid an egg against Baltimore, suffered no lineup changes or injuries, and… their odds dropped like a rock. It's recency bias, pure and simple: a "what-have-you-done-for-me-lately" mentality. The same kind of thing tends to apply to one's putting.

Here's an example. A generalized player stands on the tee of a 350-yard par four which he plays in 4.0 strokes. From 12 feet he takes 1.8 strokes (he makes it 20% of the time and almost never three-putts). From 35-feet he three-putts 20% of the time and almost never makes.

So our golfer plays the hole. He hits a good drive, and hits a great approach shot to 12 feet. He misses the birdie putt, mumbles a curse word, and taps in for par. Yes, the golfer lost ground by missing the putt… but he wasn't likely to make it anyway. A PGA Tour player was not likely to make the putt. Nobody was. The strokes he gained by hitting his drive (perhaps 0.0) and hitting his approach shot (around 0.2 gained there), he simply lost with that putt… but he was likely to lose them. He only makes that putt 20% of the time.

So he plays the hole again. He toe-hooks it into the left rough, leaving himself 130 yards out. He catches the ball cleanly but it doesn't cut so it winds up in the left bunker. From there, he splashes out to twelve feet, and again misses the putt. This golfer probably blames his putter again… but this time he also blames the bunker shot. In reality the bunker shot saved him a few tenths of a stroke, because amateurs of his skill level average 3.1 strokes from the bunker, but he hit it to a position from which he will average 1.8 strokes (3.1 - (1.8 + 1) = 0.3 strokes gained for the bunker shot). If the golfer thinks about it he might even blame the approach shot, but the tee shot will be forgiven by this golfer because he only had 130 yards to the green and an open look. In reality, the tee shot likely cost the golfer the most strokes.

I see this kind of thinking time and time again.

Three times this year my college team golfers reached a par five in two (they reached more than three par fives in two…  I'm just going to talk about three occurrences):

First Time: the guy two-putted for a tap-in birdie from about 20 feet.
Second Time: the guy two-putted for a birdie from 7 feet.
Third Time: the guy three-putted for a par from about 50 feet.

In the first example, the guy missing his putt probably cost him about 0.15 strokes. Players of his skill level make 20-footers about 15% of the time. But he made a birdie… because he gained nearly a full shot on players of his skill level by hitting the green in two. If he (players of his skill level) averages 4.85 on that hole, hitting the ball to 20 feet in two is a full 1.0 stroke gain. He lost 0.15 by missing the 20-footer, but gained a full stroke with the combination of his tee shot and second shot, to net the 0.85 strokes gained over the average.

In the second example, the same thing is true, except his putting cost him 0.5 strokes… which means that his tee/approach shot saved him a massive 1.35 strokes! Half the time he plays the first two shots that well, he's walking away with eagle… yes… but he's still only 50/50 to make the putt from there. His average score after the first two shots is 3.5. Standing on the tee, it was 4.85.

In the third example, the guy three-putted from 50 feet. He hit his first putt about ten feet past the hole and lipped out the birdie attempt. The math says… from 50 feet away, he was likely to take 2.5 strokes: he'll three-putt as often as he two-putts. This means that his approach shots saved him 0.35 strokes: yeah, he was on the green, but 50 feet away. His putting? Since he's expected to take 2.5, and took 3 instead, well it cost him half a stroke.

In each case, you know what the people said? Some variant of "Man, I should have made that putt!" In the first two examples, they were talking about their eagle putts. In the third example, he was talking about his birdie putt. Yet only one of them was even money to make their putt, and he just happened to miss. The third player rightly blames his putting for making only a par, but even he doesn't really spread the blame across the first two putts. He blamed the second putt, from ten feet. From 50 feet, golfers at his level should putt to 7 feet on average (from where they average 1.5 putts), but he putted to ten feet (where they average 1.75 putts). The first and second putt are equally to blame: both cost him 0.25 strokes, resulting in a par on a hole where he gained 0.35 strokes with his first two shots.

Golfers far and wide apply increasing attention - meaning blame and credit - for the shots they hit the closer they are to finishing the hole. Yes, a missed one-footer costs golfers a full stroke loss. But so too do tee shots that you have to hit out sideways or which you can't advance your ball very far at all, and so too do tee shots and approach shots that leave you in a greenside bunker from which you average 3.1 strokes. Which of those are more likely to happen: missing a one-foot putt, or hitting into a greenside bunker?

When you fail to get up and down from that greenside bunker, do you blame the tee and approach shots, or do you blame the bunker shot (which may have actually been quite good) and the missed putt you weren't likely to make? Most golfers blame the bunker shot and the putt.

The more recently something happens, the more emphasis golfers tend to give it. And in golf, because putts are the finality - we make or miss - we tend to assign too much weight to them.

@mvmac and I were playing at Little Mountain Country Club a year or two ago with @cedrictheo and we faced two brutally long par fours on the back nine. We both had to hit hybrids (I think @mvmac may have hit 3W one of the times) from 200+ yards. All four shots we hit onto the green and to about 15-30 feet. After each of the shots, I would yell to @mvmac "Strokes Gained!" as, at even our level of play, hitting the ball to 20 feet from 215 yards is a tremendous gain in strokes for one shot! (From 215 in the fairway, we probably average about 3.25 strokes [if not a bit more], so putting the ball to 20 feet where we average 1.75 is a half stroke saved.)

You see, @mvmac and I recognize that partial strokes are saved here and there all the time. We also recognize the odds of making a putt, and that basically means that outside of seven or eight feet, we aren't likely to make a putt. We're likely below 50% from outside seven or eight feet. We likely average 2.0 putts from about 30-35 feet. If we miss an eagle putt from 20 feet, we recognize that we hit two great shots to get there, and then realize that we were likely going to make birdie. If we make an eagle putt from 20 feet, we don't give a tremendous amount of weight to the putt: we still appreciate the two shots that did most of the heavy "saving" (4.85 - 3.0 is 1.85 strokes gained. Making the putt from 20 feet only gained us 0.75 of the strokes… the first two shots gained us 1.1).

Golfers have to learn to overcome both of those things - recency bias as well as seeing things as "full strokes" lost or gained - to truly appreciate where strokes are lost or gained. It's a world of decimal points, with tenths or hundredths of a stroke gained or lost on virtually every shot, including those that happen ten minutes before your ball finds the bottom of the cup.

#2: Contribution to Winning (or Playing Well)

Again, "Drive for show, putt for dough." Gary Player thinks so. It's the phrase that started this topic, and despite what we know about how things truly play out over large sample sizes (hundreds or thousands of rounds), it's still tough to buy in completely. It still feels like putting plays a huge role in getting that "dough" (shooting low scores).

And, frankly, it's tough to buy in completely to the new way of thinking because… it's kinda true. For the rounds in which you score really well, you tend to putt really well, too.

But you also probably hit the ball really well, too. It's tough to score really well if you're missing fairways and taking penalty shots. Every aspect of your game is likely performing at better than peak levels when you play well.

As I said in the other thread…

On January 4, 2016 at 0:44 PM, iacas said:

When a player WINS on the PGA Tour, he often has a hot putting week, but getting hot is not a reproducible skill. Being good at putting increases your chances of getting "hot" but getting hot alone is not reproducible. What I mean by that is basically that ballstriking gets a player into the top 10 or 15 or so, but from there unless someone just blows everyone away with their ball striking, it becomes about putting.

In other words, if you have a great putting week, you could finish just about anywhere in the field. You might miss the cut because your ballstriking was bad. But, if you have a great ballstriking week, you are likely going to finish fairly high in the tournament. Exactly high depends a fair amount on your putting, yes, but you're basically going to make a really nice check unless you have the yips or don't make a single putt from farther than two feet away or something.

Ballstriking is the more repeatable skill. That's why the top players tend to be the top players, but occasionally, a no-name has a good week where both his ballstriking and putting get hot. Or they'll be in a few tournaments in a row where their ballstriking gets hot, rack up a few top-five finishes, but their putting never gets quite hot enough to win.

Pros tend to over-estimate putting because their ballstriking is generally fairly consistent, so the weeks their putting is "hot" relative to their standards, they perform better (even if that's just making cuts for guys who are around that skill level generally). They overlook the weeks they putt well but strike the ball poorly because more of their putts are for pars, or from longer range (a 40' putt to a foot gains a few tenths of a stroke on the field). That's all GP is doing with his re-iteration of the "drive for show, putt for dough" line.

From ESC, I think only two players have won on the PGA Tour while losing strokes to the field with their ballstriking, and they both did so in limited field events. Several players won events losing strokes to the field with their putting, but more than making up for it with their ballstriking.

The chart from ESC is this one:

Now, you'll note that only two players have overcome poor ball-striking tournaments, and both of those were in limited field events (i.e. there were significantly fewer players to beat). And… they only lost 0.26 and 0.03 strokes with all other shots outside putting.

Meanwhile, multiple players overcame shots lost to putting (and by significantly larger amounts than 0.26 and 0.03) to win, and many of those were in full-field, 140+ player events.

But, clearly in the middle of the graph, you can see a lot of players get about a 34% contribution from putting. If you average all of these numbers, you get this chart:

This chart shows that putting averages about a 35% contribution to a win. You have the exceptions, like Bill Haas in the Tour Championship with fewer than 30 other players, just as you have Vijay in 2008, Tiger in 2007, and several other players who lost strokes from putting yet still won the event.

This is to be expected, for two reasons:

1. Putting, though stable long-term, is relatively volatile in the short term.
2. There are more strokes to be lost or gained "tee to green" than from putting over the long term.

Below, I've created two charts showing 80 rounds of an "average Tour player" and Tiger Woods (not 2015 Tiger Woods, but not 2000 Tiger Woods either - just a "very good player"). I randomly generated numbers between ±5.5 strokes gained "other" (SGO, or "tee to green") and ±3.5 strokes gained putting (SGP) for the Average Tour player, and did the same with Tiger, except that I bumped Tiger's SGO limits up 1 and his SGP up 0.5. In other words, Tiger would average a full stroke gained better tee to green and half a stroke gained better with the putter. Because neither Numbers nor Excel can properly show a stacked bar chart of 1.0 for something that is +1.2 and -0.2, I added the numbers together to present a single bar. That's unfortunate, as you can't see each's contribution to the total, but it also helps to simplify the graph, too.

Now, a few caveats. First, this doesn't account for ebbs and flows, particularly tee to green - each round has completely random chances of getting any number, while in reality an average PGA Tour player and Tiger Woods could both be expected to "find something" in their full swings and putting strokes or to "lose it" for a series of consecutive rounds. What I mean is that both could have stretches of slightly better than average play and stretches of slightly worse than average play, particularly tee to green, than I was able to generate here, randomly.

Second, I came up with these numbers based on the ranges from the charts above. Given the stats from the 2015 PGA Tour (farther below), I probably over-estimated both, but more so putting. I used these numbers based on the maximum values from the chart above, and assumed that the distribution was fairly even. In other words, I assumed that if one player can be +4 strokes tee to green, another player could be -4 tee to green. This tends to be how it shakes out - we tend not to see 1/3 of the field at +x while 2/3 of the field is at -x/2. But, it's still a small assumption I wanted to note.

Here, then, are the results of 80 somewhat random rounds. Again, the Player was SGO ±5.5 and SGP ±3.5, while Tiger was SGO +1 and SGP +0.5 over the Player.

Average Tour Player

Tiger Woods

I drew a red line at 3 strokes gained. Let's imagine that's playing well enough to win a PGA Tour event (so long as someone else doesn't also have a hot week).

As you can see here, the average player plays well enough to win a PGA Tour event… maybe once, assuming someone else doesn't have an even better set of four rounds those days and that rounds 21-24 (SGT are 6.45, 5.30, 6.85, and -2.51 for a total of 16.09) occur during the same event. Those rounds break down to:

 SGO     SGP       SGT
-----   -----     -----
 4.89    1.56      6.45
 2.47    2.84      5.30
 4.79    2.06      6.85
-4.73    2.22     -2.51

That works out to about 54% contribution to the win from his putting. The player finished about 16 strokes below the average for the tournament, and nearly 8.7 of those came from putting. The player had such a big lead going into the last round that he played safe from several tees, lost a lot of strokes tee to green, but then let his "hot putting week" carry him to the possible win.

And have a hot putting week, he did… if the theoretical normal max you can gain putting is 3.5 strokes putting (14 total), this player gained an average of 2.17 per round, or 62% of what I'll call "maximum putting." Conversely, since you can gain 5.5 strokes with all other strokes (or 22 total), this player gained only 33.7% of his "maximum other." He had a great week putting, and a pretty good week tee to green.

Contrast that with Tiger Woods, who probably wins… four to eight times, depending on how the rounds fall (i.e. a stretch of four good rounds doesn't overlap two tournaments and, to a lesser extent, how his peers play. Even rounds where he treads water or loses a fraction of a stroke in total are made up for by the rounds in which he gains, for example, 5.42 SGO and 0.56 SGP. Looking at his last four rounds of the simulated "year" he has:

 SGO     SGP       SGT
-----   -----     -----
-0.91   -0.28     -1.19
 5.31    2.57      7.88
 5.42    0.56      5.98
 2.86    0.79      3.64

That's 16.31 strokes, or averaging over 4 SGT despite losing 1.19 strokes in the first round (boy what a "Tiger charge" he put on Friday, though!). His contributions? SGP: 22.3%, SGO: 77.7%. For Tiger's skillset, he could get a max of 26 tee to green, and he got about 56% of his "maximum" there. In putting, he can gain 4 strokes, or 16 total, and managed to get only 23%. Still a net positive, but not great. He had a pretty good week tee to green, and an "okay" week putting.

---

Earlier I listed these two reasons:

1. Putting, though stable long-term, is relatively volatile in the short term.
2. There are more strokes to be lost or gained "tee to green" than from putting over the long term.

The second is the easiest to address. One simply has to look at the Strokes Gained stats on the PGA Tour website for this. If they do, and if they look at the 2015 stats, they'd see these results:

http://www.pgatour.com/stats/stat.02564.2015.html
SGP: Top 5 +.65, Bottom 5 -.94.

http://www.pgatour.com/stats/stat.02674.html
SGO: Top 5 +2.4, Bottom 5 -3.0

There are between three to four times more strokes to be gained tee to green than there are putting. Now, this is all relative, remember. This has to deal with how widely separated players are. As we know, both from ESC and LSW, players are much closer in putting than they are at any other skill. The longer the shot, the wider the separation becomes. So, since the "strokes gained" stats are all compared relatively, to their peers, it makes sense that the gap between the best and worst putter is barely 1.5 strokes, while the gap between the best and worst players tee to green is almost 5.5 strokes.

The problem, though, are with the red words from up above: randomly generated. You see, while putting stats are often somewhat random, or more volatile, ball-striking and the short game is not. It's actually relatively stable, for long stretches of time. Putting, however, is not. This gets into the repeatability of the skill.

As I said above in the quote and in the other thread, putting is not a highly repeatable skill in the short term. Watch a PGA Tour player hit 20 five-irons and they'll probably all be within a pretty small area. Watch him hit 20 twenty-foot putts, though, and he'll make some and miss some. Though it is 72 holes, in comparison to the length of a season, a PGA Tour event is a rather small window, and a rather small sample size. Over the course of a long season, the better putters separate themselves, but over 72 holes, it's much, much more difficult to separate yourself due to the lack of repeatability in putting.

Tournament to tournament, more so day to day, and especially hole to hole… putting stats vary wildly. Consider a single eight-foot putt: if you make it, you gain 0.5 strokes. If you miss, you lose 0.5 strokes. Each time you face an eight-foot putt, you could be 100% or 0%. That's highly variable, with a huge swing in the results: half a stroke from the average every time. The only time you can lose or gain strokes that quickly tee to green is when you hit your ball into a penalty situation or hole out - in other words, horribly bad (super unlucky) or good (super lucky) shots that don't happen very often at all.

Because of the proximity to the hole as well as the finality of "make" or "miss," putting, in effect, has a much higher dose of randomness, of "luck," than the rest of the game. I'll illustrate this - while further addressing the red words above - with two thought exercises.

Imagine you have a magical bag with 100 slips of paper inside that say either "good" or "bad." A good putter might have 55 that say "good" and a poor putter will have only 45 "good"s. (I've chosen the words "good" or "bad" because if you have a one-footer a "bad" putt will still likely go in, and from 40 feet, a "good" putt will either go in or, more likely, leave a tap-in second putt while a "bad" putt may lead to a three-putt. If it simplifies things for you, though, consider "good" as "make" and "bad" as "miss." It's not quite that granular, though.)

So again, the good putter has 55 "good"s and the poor putter has 55 "bad"s. You ask both players to pull 20 pieces of paper from their bag. (I chose 20 because it's a nice round number, and because if you have 29 putts, we can ignore the 9 tap-ins per round.) The bag is magical because if a player draws a "good" slip, a new "good" slip is regenerated inside the bag, and the same is true of the "bad" slips. The bag maintains the 55:45 or 45:55 ratio. Over 20 tries you could easily imagine either player scoring 10-10, 13-7, or even 15-5 one way or the other. These types of results would not be uncommon.

Over the course of a season? Why, you'd expect the score to be 55-45 for the good putter and 45-55 for the poor putter. But the smaller you make the sample size, the more likely you'll get some extreme results. Again, draw one slip of paper and you'll get 100%-0%. Draw three, and you'll still get 100%-0% (1/8th of the time if there are 50 of each). And remember… these are the "best" and "worst" putters out there, too: the vast majority of putters have right around 50 "good"s in their bag. random sequences of 3, 4, 5, even 10 things in a row occur more often than you might think.

Tee to green behavior, though, is far more repeatable (or far less variable). Because there's virtually no chance of finality when you hit your tee shot on a hole, the slips of paper don't say "good" or "bad." Instead, the pieces of paper have a bunch of numbers that can range from 1 to 100, distributed not as evenly as with putting, but much like a bell curve. There may be five "50"s in a bag with very few numbers (perhaps only one each) from 1-10 and 91-100. A player pulling from that bag is likely to get a lot of average shots: a 50, a 47, a 55, and so on. A tee shot that goes 289 yards into the right rough might be slightly better than a tee shot that goes 267 yards into the fairway, but it's going to be a small fraction of a stroke: it's not going to be as final as "good" or "bad," as significant as "make" or "miss."

This means that strokes in the full swing and short game add up more slowly, yes, but also more consistently. While putting is a bit like pulling a 95 or a 5, with the full swing you're likely to pull a 58 and then a 47.

We also know from, well, everything that there's more separation to be had in the full swing. Just look again at the PGA Tour stats above: the best to the worst putter was about 1.5 strokes, while the best to the worst tee to green was nearly 5.5 strokes. And consider that while Mark Broadie tells us a scratch golfer can putt better than a PGA Tour pro 30% of the time (over 18 holes), we know they'll never hit the ball better than a PGA Tour pro tee to green over a round of golf. This gap, though narrower, still exists on the PGA Tour: the distance between the best putter and the worst putter is quite a bit smaller than the distance between the best and worst "tee to greeners."

This means that bell curve, the "SGO" is shifted to the right or left quite a bit more than we saw with putting. While putting had a range of 10 from the best to the worst putter (55/45 to 45/55), Tiger's pouch has a bell curve centered around 65 or so while the poorer "tee to greener" has a bell curve centered around 35 or so, creating a gap that's about 3x the size of the putting gap. Nice how that all works out, isn't it?

In other words, Tiger's pouch has a lot of 60s and 70s, and even a few from 91-100, while the other player's pouch has few numbers above even 70.

Again:

1. Putting, though stable long-term, is relatively volatile in the short term.
2. There are more strokes to be lost or gained "tee to green" than from putting over the long term.

I haven't really addressed the second issue directly, but I can do that now in two parts, quickly. Those parts are:

2a) There's more separation between players, long-term.
2b) While putting only offers about 20 chances, "tee to green" offers about 40 chances to lose or gain strokes.

The latter point also speaks to the variability that's possible in putting, so it also partly supports point 1.

---

Long story short, a few bullet points:

• Because putting is so close to the end of the hole, you're more likely to notice the success and failures.
• Because putting is so granular, so "make/miss," you're more likely to notice the success and failures.
• Because you hit 40 shots that allow for separation tee to green, and only about 20 putts, you're more likely to notice the putts that drop (or don't) over the smaller gains you make more often tee to green.
• Strokes gained tee to green can be incredibly small, but add up. The difference between the #1 player and the #120 player from between 150 and 200 yards on the PGA Tour over a stretch of time was three feet.*
• Putting does contribute higher to winning. Poor putting can also contribute to losing, too. But again, that's because it's highly variable, round to round, and even tournament to tournament.

To summarize even further:

• Tee-to-green play determines what quartile you're likely to finish that week.
• Putting determines where you finish in that quartile.

Thank you. And yes, I spent entirely too long on this…

---

From the asterisk above:

Spoiler

Quote

Camilo Villegas, born in Medellin, Colombia, has won three times on the PGA Tour since he joined in 2006. Until 2012, Camilo hit half of his shots starting between 150 and 200 yards from the hole to within 5.8% of the initial distance to the hole. That’s 30 feet from the hole on shots starting from 175 yards. Camilo ranks 120th in this stat. Another golfer hit half of his shots starting 175 yards from the hole to within 27 feet of the hole, three feet closer. What rank do you estimate for this other golfer?

You might be surprised that this other golfer is Tiger Woods, who ranks first in the category. On the putting green, the PGA Tour average to hole out from 27 feet is only 0.025 strokes lower than from 30 feet. The three-foot difference in proximity to the hole hardly seems significant, but in reality it’s crucial. First, a fractional gain for each approach shot adds up over an average of 18 approach shots per round. On top of that, the three-foot difference matters because of shots that finish on the green instead of in the rough or sand. Because of that, Tiger hits the green 75% of the time versus the tour average of 67% on shots starting 150 to 200 yards from the hole. More important, the three-foot difference matters for five-foot putts that become two-footers and for eight-foot putts that become five-footers. Those differences really add up. Tiger’s three-foot advantage from 175 yards, when applied to all approach shots in a round, accumulates to a 1.3-stroke advantage per round or five strokes every tournament. That’s an enormous advantage!

- Broadie, Mark (2014-03-06). Every Shot Counts: Using the Revolutionary Strokes Gained Approach to Improve Your Golf Performance and Strategy (pp. 108-109). Penguin Group US. Kindle Edition. 

 

[boogielicious]

Nice post @iacas.

[natureboy]

Excellent post. The 'putt for dough' thought probably arose primarily due to the recency bias and the granular / binary nature of make or miss.

If you don't hit to penalty it's hard for a player let alone a spectator to appreciate the small gains of a slightly better long shot, but it's memorable when an opponent drops a long putt.

The 'Strokes Gained' or lost on individual strokes sort of feel like 'statistical illusions' describing which shots on a hole helped or hurt you. You may have hit to a position that on average gains an incremental advantage over the field, but that advantage is not actually realized in your score until you follow it up by ultimately draining the putt the good shot put you in position to have a better chance to make.

Edited January 5, 2016 by natureboy

[alleztom]

An image I like is to think of the area around the hole as a gravity well. Inside 3' and you're sucked in and can't get out (you really need to stuff up to miss), then you get the sharp walls that every inch is a huge difference out to about 15', then it flattens out (to estimated 2 putts). Works on chips too.

The closer to the purple part of the well you can get the ball, the less you're relying on a 'luck' (probably better put as a less statistically likely event) to make the putt.

 

Edited January 5, 2016 by alleztom
Incorporate image

[natureboy]

14 minutes ago, alleztom said:

An image I like is to think of the area around the hole as a gravity well. Inside 3' and you're sucked in and can't get out (you really need to stuff up to miss), then you get the sharp walls that every inch is a huge difference out to about 15', then it flattens out (to estimated 2 putts). Works on chips too.

The closer to the purple part of the well you can get the ball, the less you're relying on a 'luck' (probably better put as a less statistically likely event) to make the putt.

 

Gofer1: Is that a gimme?  Golfer2: No, it's not inside the event horizon.

[iacas]

Roughly…

Note that the axes are not proportional or anything. The area under the curves is not the same. This is an illustration more than anything. And the "Playing Level" scale I'd say is not linear, except depending on how you view skill, it could be - if you realize it's easier to go from 18 to 9 than it is to go from Web.com to PGA Tour, then the scale probably makes plenty of sense. Just don't take it to be a "difficulty" scale, but rather, a playing level scale as written.

What you'll note:

• There's way more overlap in putting. Even a 90s golfer can out-putt a PGA Tour player 10% of the time. A scratch golfer can do it 30% of the time. There's less separation in putting.
• There's no overlap in the full swing motion from the scratch golfer to even the bad rounds Web.com Tour players have.
• Better players have a significantly narrower window of performance than poorer players, whose performance varies more day to day, shot to shot, etc. with the full swing. Putting, on the other hand, has similar width curves. Even Phil Mickelson has days where he three-putts from three feet a few times, and even you have days where you make a lot of long putts you don't normally make.

[natureboy]

25 minutes ago, iacas said:

Roughly…

Note that the axes are not proportional or anything. The area under the curves is not the same. This is an illustration more than anything. And the "Playing Level" scale I'd say is not linear, except depending on how you view skill, it could be - if you realize it's easier to go from 18 to 9 than it is to go from Web.com to PGA Tour, then the scale probably makes plenty of sense. Just don't take it to be a "difficulty" scale, but rather, a playing level scale as written.

What you'll note:

• There's way more overlap in putting. Even a 90s golfer can out-putt a PGA Tour player 10% of the time. A scratch golfer can do it 30% of the time. There's less separation in putting.
• There's no overlap in the full swing motion from the scratch golfer to even the bad rounds Web.com Tour players have.
• Better players have a significantly narrower window of performance than poorer players, whose performance varies more day to day, shot to shot, etc. with the full swing. Putting, on the other hand, has similar width curves. Even Phil Mickelson has days where he three-putts from three feet a few times, and even you have days where you make a lot of long putts you don't normally make.

Great illustration of your points. Those long game distributions for pros are daunting.

Was the difference in the width of the putting distributions at all intentional, or just to illustrate? If intentional, why the greater putting consistency for scratch...putting easier greens?

Edited January 5, 2016 by natureboy

[iacas]

10 minutes ago, natureboy said:

Was the difference in the width of the putting distributions at all intentional, or just to illustrate? If intentional, why the greater putting consistency for scratch...putting easier greens?

No. I'd probably change that if I was going to revise the graph. I simply tweaked the sizes and positions of the top three, then pasted in the bottom three and did the same to them, and forgot to make sure the two sets of three were consistent with each other.

I think the general point of the illustration still holds.

[Lihu]

15 hours ago, boogielicious said:

Nice post @iacas.

+1 on this. However, I simply trust the conclusion that is written in LSW and in the plots above showing the correlation to handicap rather than trying to figure out everything in that post.

[Phil McGleno]

I done wore out my page down key.-But it was worth it.

Then again we already knew all of this, kinda.

But I think the thing we did not have before is the idea of repeatability and how much that changes. You are right-Putting is not highly repeatable over a few holes or even a single round. Sometimes you get hot-Sometimes you are not.

[boogielicious]

16 minutes ago, Lihu said:

+1 on this. However, I simply trust the conclusion that is written in LSW and in the plots above showing the correlation to handicap rather than trying to figure out everything in that post.

You're an engineer are you not? I am and I understood it all. The math is easy. 

[Lihu]

4 minutes ago, boogielicious said:

You're an engineer are you not? I am and I understood it all. The math is easy. 

Yes, but I'd rather spend the time to enjoy golf!

It's like I can figure out how to make an OS from scratch and I've done simple RTOS back when they didn't have pre-canned ones, but now I use a pre-canned one. Why re-invent the wheel when there is one that I trust already made?

However, I do like that that @iacas has data to prove his theories and all, but am too busy to go through the data in the detail that is needed to re-produce the conclusions. That and I completely trust Erik's capability and knowledge.

[boogielicious]

2 minutes ago, Lihu said:

Yes, but I'd rather spend the time to enjoy golf!

It's like I can figure out how to make an OS from scratch and I've done simple RTOS back when they didn't have pre-canned ones, but now I use a pre-canned one. Why re-invent the wheel when there is one that I trust already made?

However, I do like that that @iacas has data to prove his theories and all, but am too busy to go through the data in the detail that is needed to re-produce the conclusions. That and I completely trust Erik's capability and knowledge.

It's was 9F outside this morning here in Boston. This is all I can enjoy!

[RandallT]

1 hour ago, iacas said:

Roughly…

 

I have to admit I haven't finished reading the full post, but I will!

This interesting data visualization above jumped out at me. It would be interesting to apply this style of visualization to just the PGA players during tournaments. For example, on average for the top 10 players of each tournament, how did their "playing level" that week compare full swing vs putting. And then throw in "short game" somehow.

From the scatter plot showing that 35% is that average contribution of putting to winners, we'd see more skew in the full swing for the winners. But the overlaps would be quite substantial for both full swing and putting- enough to give many people the anecdotal sense that you "putt for dough" or make your money "around the greens."  While many people might have trouble reading the scatter plot, this style of visualization might hit home more.

Since I've regained an interest in golf, the idea of variance in performance has been an area of interest for me, since we are not machines who perform at the same level round-to-round, or hole-to-hole. So the bell curve shapes tell an important story to me.

Great thread.

 

[Lihu]

Just now, boogielicious said:

It's was 9F outside this morning here in Boston. This is all I can enjoy!

Ouch, I feel for you. . .

It's raining here, but I thought I could get much more work done so I might have a chance to take a late lunch round later on?

[mvmac]

As others have said, great post @iacas.

From my own playing experience and observing the tour players at events, it makes a lot of sense to me. Tour players are the best in the world because of they're the best ball-strikers in the world. You can survive on the PGA Tour while not being a great putter (Boo Weekley John Senden), can't keep your job with a great short game and poor ball-striking.

[iacas]

15 minutes ago, RandallT said:

Since I've regained an interest in golf, the idea of variance in performance has been an area of interest for me, since we are not machines who perform at the same level round-to-round, or hole-to-hole. So the bell curve shapes tell an important story to me.

Cool. Quick note, which I think you already knew, but still worth mentioning… the plots aren't generated from anything but my history, etc. I kind of assumed bell curves because people are familiar with them, and because over a large enough sample size, you tend to get bell curves in stuff like this, but they might be bell curves with longer tails to one side (i.e. not a "centered" bell curve), they might actually have some flat spots… etc. Especially if you're talking about individual golfers over 100 rounds or something, you'd see very different looking shapes. Again, I don't think so for 1000 golfers and 100,000 rounds, really, but I can't say for certain.

2 minutes ago, mvmac said:

You can survive on the PGA Tour while not being a great putter (Boo Weekley John Senden), can't keep your job with a great short game and poor ball-striking.

Well, Boo lost his card for a bit there, but his putting was a-tro-cious. I think if my daughter putted for him he might have kept his PGA Tour card. That 30% a scratch golfer beats a PGA Tour player? They weren't using Boo in that example. I don't even want to say the number would have been flipped to 70%… because I think it would've been higher!

He was B-A-D. At putting. Like middle region pink bad.

[span3636]

I found your topic interesting.  I read a book by Bob Rotella about short game and putting a few years ago.  He talked about the chipping and pitching being more important than the putting.  His reasoning was that the closer you get the ball to the hole the better your chances of making the putt.  I'm not sure if this correlates to your article exactly, but I thought I'd share this.  I think I'd rather have great putting and chipping and be a decent putter than having to try and make long putts for par all the time.