Two's Complement - Golf for Hackers

Episode Date: June 19, 2022

Matt and Ben talk, about uh...golf? What? Is this right? Did you check this? Apparently, in this episode, Ben explains how technology and analytical advances in golf have dramatically changed the game. Matt gently prods him on.

I'm Matt Godbolt and I'm Ben Rady and this is Two's Compliment, a programming podcast. Hey Ben. Hey Matt. So you were talking to us the other day, and by us I mean the folks at work, about one of your pastimes, which isn't really very programming related but somehow because you're you you were able to bend it to fit a group of technical people so i figured we should take the opportunity to to ask you on this podcast about that so that you can you know um give more people the this uh a window into the world yeah describe so do you want to talk about it definitely yeah this is definitely off brand for us but right well i don't know there's we're gonna statistically lose seven tenths of our listener on this one but the expected return of our listener

reduces but you know well so i mean tell me about, man. And not code golf in this instance. No, not code golf or frisbee golf or there's got to be other kinds of golf. No, so I've been playing golf for a long time. I've been playing golf since I was 10, 11, 12, somewhere in that range. And lately, there's been a lot of changes to the game of golf. And it's been able to be quantified a lot more than it could in the past. And this has sort of exposed some sort of like, you know, well-trodden, well-known, you know, sort of old wisdom as being incorrect or misleading. And as a result, the game has changed pretty dramatically. And in fact, I think you could say that in the last 20 years, golf has probably changed more than it has in the first 500 years. I mean, golf has been

played since the 15th century. And to back this claim up, it's a pretty bold claim. It's a bold claim. Very bold. And to back this claim up, it's a pretty bold claim. It is a bold claim. And to back this claim up, I would maybe point to one of the most recent major tournaments in golf, the PGA Championship. The course had to be laid out in a way such that one of the tees was on the opposite side

of a green of another hole. Like the golfers had to wait for the players on the green to clear before they could tee off knocking a ball over their heads hitting it over the over the green over their heads and the reason for this is because uh the length of the course as it was originally laid out which wasn't that long ago was just too short it was too short for the modern game and so they had to lengthen it and they didn't really have any was just too short it was too short for the modern game and so they had to lengthen it and they didn't really have any space to lengthen it into so like well we're just going to put on the other side of the green and yes that kind of sucks because it means that people

are going to have to wait but this is my evidence for crazy golf yeah yeah it's like it's like through the windmill golf courses over the hill and under the thing and you got to hit it through this pipe um and and and sort of the changes that under the thing, and you've got to hit it through this pipe. And sort of the changes that have occurred in golf, and they're technological changes, but I think there are also some more sort of statistical changes, sort of the moneyball effect that baseball has had. I was going to say, it sounds very much like moneyball, but you're saying this is over and above what I might imagine is better nutrition, understanding of the physics of it, like in terms of musculature and stuff like that,

something more strategic? I mean – I think it's two things. I think there's two things. And they both have their own sort of nerdy aspects to it. But, you know, today we're probably just going to focus on one. One is just the technology, right? Like, you know, balls, golf balls that fly farther um drivers that hit the ball farther you know

distance has been sort of a main factor but also you know technology around like you know spin and control and wedges and all these other things in the last 20 years last 25 years have changed really significantly much more oh interesting i think that they have as a lay person like a golf stick is a golf stick right it's just a long tube of metal with a thing on the end of it you hit the ball with. But you're saying there's a lot more sophistication than that. Yeah, yeah. Mostly with golf balls, but also with clubs, specifically drivers. Okay.

You know, like, you know, 50 years ago, like a driver with a metal head on it was a revolutionary thing, right? Like they used to be made out of wood. They were woods. That's what I still think of them as wood, right? Because like my dad's old set would be made of wood. But yeah, I do remember that. Yeah. And then, you know, they sort of invented like,

okay, what if we made them out of steel? And now they're made of, out of carbon fiber composite, right? Wow. TaylorMade actually just came out with a new driver that is entirely carbon fiber composite. There no titanium in it which is what so these kinds of developments have have especially for the pros but also for amateur players really more for pros change the game pretty significantly

right and then the other thing aside from sort of the technology the equipment technology that changed is the stuff that I talked about yesterday, which is being able to quantify things about the game, about strategy, about swing, golf swings, and the effect that they have on the ball in ways that people were just kind of like assuming based on their observations. Right. Right. ways that um people were just kind of like assuming based on their observations right right right um i remember the sort of like cooking to completely take this off track but i remember like a friend of mine you know this is probably 10 years ago saying like you know so many things that we know know in quotes about cooking don't bear out if you actually try experimenting with it and try measuring it and try using like a heat lamp and whatever i mean you you you know about this stuff as well. Yes, yes.

But like, you know, because these things get passed down and without any way to check, you go with this sort of essentially superstition over these things. And now you're saying we have ways of measuring and saying, are we actually doing what we think we're doing? Absolutely, yeah, yeah.

So we have some of the more recent developments and I think, again, this is sort of like over the last 20 years or so. You've got launch monitors. You've got devices that you can put down when you hit a golf club, and using either high-speed cameras or radar or both, it can measure very precisely what direction the ball flies, how far it goes. You put down a little thing near where your ball is,

and you click a button or whatever, and then you take your shot, and then you get some kind of telemetry about what the heck happened. Exactly. You don't even need to click a button. It just detects if there was a golf shot. I guess, yeah. Yeah, and it'll record that, and there's different mechanisms that they use to do this, but you can get very precise information about what the face angle of the club was when it hit the ball. Was it pointing three degrees to the right or was it pointing three degrees to the left? Was it straight? What the spin of the ball was as it came off the club, how fast it was moving, the angle

that it was moving at, the angle that the club was moving down into the ball when it hit it, or was it maybe going up at a different angle? How far behind or in front of the ball did it hit? And so- That's a lot of information. Yeah, did it hit and so that's a lot of information yeah it's a ton it's a ton of information and just just take just nerding out for a second there like this is kind of pretty cool stuff if you're doing very high speed because i mean how many i mean this isn't a regular phone a phone a regular camera cameras you might have on

your phone because like i can't imagine in the small area that you're the club head and the baller in that that it spends more than a thousandth of a second in that kind of area something of that order and there's that's like not very much time to record all that telemetry and to be able to look at it how amazing and then to process it and you say you're doing spin it's doing spin recognition so presumably the dimples on the ball are enough for it to do like and the markings yeah like the logo basically yes oh my god so like there's I mean right just to get a bit of tech into this actual the dimples on the ball are enough for it to do like and the markings yeah like the logo basically yes oh my god so like there's i mean right just to get a bit of tech into this actual tech tech

into this like there's a lot of cool things going on in there that's really really clever but anyway the result is you get a ton of information about your shot that allows you to make determinations about whether right right actually made the difference and and these devices now have come down in price the technology's gotten good enough. They're very accurate, and they're not that expensive. A really good launch monitor is kind of expensive, costs about $20,000. Right, that's a lot of money.

That's sort of the high end. But you can get ones about for the price of a laptop computer. You can get one for like $500 to $2,000, $3,000. And if you're taking it you know as a hobby that's not an totally unreasonable amount of money to right right and the sort of democratization of this technology where it's you know and especially a lot of these things really took off with covid because people were you know at home not a lot to do wind up with a little bit of extra spending cash is sort of like like, well, I can't go out anymore. I can't actually go out and play golf, although they did eventually open that up. What do I do? Well, I'll go buy one of these launch monitors. And so the availability of these things and the sort of ubiquity of them has allowed people to

talk about golf in a quantifiable way where they couldn't before, right? You have instructors now that are, you know, used to say things like, oh, what you need to do is get your left arm a little bit more flat, right? Like that's the way to improve your scores. And now you have instructors that say, no, what you need is to decrease your angle of attack with the club face from 7.2 degrees down to 4.2 degrees down because you're hitting down too much on the ball and you need to get it down from seven to four and you can have somebody that just sits there and like you don't even need to tell them what to do with their arms or their body or their whatever you just say try to hit it less down and they're like oh i went from seven to six point eight good

you're improving do more of that well i'm down to 5.2 so i was gonna say like we're going with all this technology it sort of seems like it takes out the humanity of it somehow but like not at all because the skill level is still absolutely required to have that like dexterity and fine motor control at that velocity and strength that you're presumably putting into it as well so right but it lets you have a very quantum yeah i i you know i'm having a sort of flashback to my poor dad trying to teach me to play golf and standing behind me and telling me all the things that I'm doing wrong. And I'm sure he's right about almost all of them. But like, I don't know how I know I was definitely doing something wrong because I wasn't very good.

Right, right, right. And I think this has had a really positive effect on the game for lots of reasons. One is, is it creates a feedback cycle. You know, we talk about feedback on this podcast quite often yeah and it creates a feedback cycle that is is is much more um quantifiable obviously but i think also much less prone to superstition yeah right like like oh you got to keep your head down you got to keep your left arm straight you know all these kinds of things like okay well what effect does that have on the ball what does it really do and then it's do yeah what does that actually do my head up at the end of the show? And it's sort of like, well, once

you master it, it'll make your shots better. It's like, well, that's a long time away and that seems like a really long investment to make. Like, can I do something that will just make it better right now? And with these sort of like, you know, the sort of impact dynamics and like understanding how ball flight works,

which is another thing, sidebar, golfers were wrong about that for 500 years. Like, the way that ball flight works which which is another thing sidebar uh golfers were wrong about that for 500 years like the way that ball flight actually works there's if you google uh golf new ball flight laws yes you will figure out that only in the last 20 years have people actually figured out how golf balls really fly off a club and that is in in large part because of these these technology and these launch monitors. They're able to look at it and go, oh, yeah, that's not true at all. Yes, yes, yes. So now we've come into this world where you can say, like, okay, I'm not going to

give you all this advice that might maybe one day be useful. I can look at the physics of the ball. I can measure what you're actually doing on this launch monitor and i can tell you yeah the path of your club is into out by five degrees and that's too much that's what causes your hook so you need to narrow that down to two degrees and you will immediately see the the flight of your ball change now whether you can repeat that on the golf course or whether you can do other things is practice and you have to still all do those things as a as a lay person like i could see that really gamifying it for myself to be able to like i know what i'm doing wrong and i can still sort of fathom out what what i might need to change but when it's just randomly going off to the left

and my dad's saying oh no no no you need to open the club face or whatever and i'm like i'm i think i'm doing that and it's getting worse i don don't know if, am I just doing it wrong? Or is there some supposition that we both have wrong? Is that not how it really works? Right. Really interesting. Right, right. Yeah, yeah, yeah.

So that obviously makes a lot of sense, right? If you can understand the physics of the situation and what you need to change to change both the direction of the ball as it comes off the club head and also the way it's spinning and interacting with the air and all that kind of stuff. But what else can you do with numbers? So the other thing that you can do, and this is another sort of a more recent development, I think. And the pithy phrase for this is a golf club is a shotgun, not a rifle. And what that means is that when you hit a golf ball, having an expectation that is going to fly 150 yards and land and roll out five yards or something like that, it's just a flawed way to think about golf.

What you should be thinking about is this is my dispersion area. I see. I'm going to hit a ball and it's gonna be somewhere in this area and the essence of strategy really is just mapping that dispersion area for all your different shots yeah to the hole that's in front of you and trying to pick the one that has the best result the most favorable outcomes and the least unfavorable so like a monte carlo simulation of like given that you're going to be landing in this region with this sort of probability density um right what's the right place to aim for to get the best chance because yeah like you say i mean like you know i the my limit of playing golf is on the wii so you know

you know where it's gonna land because there's a dirty great big line showing you where it's going to land and it bounces beautifully every time there's never any surprises but in i have noticed that in real life that doesn't tend to happen so it doesn't really work so yeah you're saying actually so there's a strategy here over and above just get it to this point that's the nearest point you can get to without getting into the water and then go from the water to the over the water to the hole or whatever it is yeah yeah and then sort sort of quantifying what is the sort of expected value in that area is another place where both the sort of launch monitor technology and other technologies,

so there's other cool things that they have, and this is only possible because of GPS, which is only possible because of NASA, which is only possible because of, and you go all the way back in time and you see how all these wonderful things tie together. Einstein's theory of relativity, general theory of relativity. Exactly.

So thanks Einstein for the ability for us to play golf in this way, where you have devices that you can attach to your golf clubs that will track all of your shots on the course. Neat. And that will give you some interesting information it'll help you calculate these dispersion areas that we're talking about right empirically so like you know i was here with a seven iron and i doesn't matter what i did i landed in this region and i can play the same hole 20 times and go where did i land from the t every time or whatever exactly exactly which you can also do in the launch monitor and you

can compare those two results like you can hit, 10, seven irons on a launch monitor and see what your dispersion area is. And then you can go down on golf course and you can play and you can take that data and, and, you know, correlate it together and be like, am I seeing on the course what I'm seeing on my launch monitor? No, why not? Maybe wind or maybe I'm, you know, following a slightly different, uh, uh, you slightly different routine when I'm on the golf course. On a launch runway, I'm just hitting the same balls over and over again and I need to mix it up or whatever it might be. You can correlate those things. And then taking all of this information and especially the shot location information,

one of the more interesting things that the tour, the professional PGA tour did, is in 2001, they built this system that uses laser range finding to measure to just a few inches where every golf professional on every tournament hits their ball. So there's this really rich data set that is available that you can get that shows for all of these different shots. And using that information, there was a guy named Mark Brody that came up with this technique called Strokes Gained, where there used to be lots of different ways to measure sort of the quality of a golf shot, right? People would say, oh, did you hit the green in regulation? Or did you hit the fairway? Did you hit all these different things?

And these stats, like, they weren't really good. They didn't really tell you anything. And so what he wanted to do is to be able to compare from one shot to the next, from one player to the next, from one course to the next. How good was this shot? How can I quantify that? Right. Right? How do I rate a particular shot?

Yes. Yes. And he wanted to measure. The unit of measure that he wanted to use was strokes so he could make claims like that strategy that you chose was two tenths of a stroke a stroke worse than this other strategy that you could have chosen fractional strokes here as well like to be able to do a quantifiable analysis between things where right you know yeah okay right so he wanted to be able to do that so his method for doing this was basically he came up with this system where he measured the average number of strokes that it took professionals right and amateurs actually do this for amateurs too but but it was easier with professionals

because the data was there to say how many strokes on average does it take to hole out from a particular location on the course so if you're you know 420 yards away on the tee yeah on average that's like 4.1 strokes for a professional golfer to hold out. Right, because it's a par 4 or whatever. And that's what the golf course is on. Well, it doesn't even matter what the par 4 is. I guess it doesn't matter. No, no, no.

This is like an empirical determination. Yeah. It could be a par 5. It could be a par 4. It doesn't matter. We're measuring how many strokes does it take on average from golfers to hold out from this location. From this particular so there's a kind of a map over i can i'm sort of imagining a map of the golf course uh yes where i can like see the contours of from here we'd expect to be four

strokes from here we'd expect to be three two and then once you get around the whole one or whatever and that that's kind of you expected if you could play like a professional yes an average professional from this location the expected number of strokes would be 3.2, right? Or 1.6. Yep. And so the way that he measures, the way that the system that he came up with to do this was, okay, we're going to take the average from your starting location. Yep. We're going to subtract one because you're hitting a stroke. Right, you're taking that stroke.

So if you hit a stroke, you would expect to get one better. Yep. Right? And then we're going to subtract the expected number of strokes at the ending location. So if you start out at a starting location and your average is 4.1 and you hit a shot, so, okay, we're going to subtract now and go down to 3.1.

Because, like, given this system being perfect, then every time you hit a stroke, you should be one stroke nearer to the hole. One stroke nearer to the hole. Right, exactly. And then if your ending location turned out to not be very good, like you didn't hit a very good shot and your ending location wasn't 3.1,

it was actually like 3.3, then you've actually lost two-tenths of a stroke on that. I see. Whereas if you hit a better shot and your ending location was 2.8, then, you know, 3.1 minus 2.8, you actually gained three tenths of a stroke.

So you did better, right? And now he had this system that he could use to quantify and say like, okay, if you hit this shot, you gain two tenths of a stroke. If you hit this shot, you lost four tenths of a stroke. So clearly this this shot, you lost four-tenths of a stroke.

I see. So clearly this is better. So that's what you can empirically determine for any one position, like where you are, what you should get, and then hit the ball, where did you end up? And then you can just say, like, that was – I can say that it was a gained.2 of a stroke there. Yes, yes. And this system takes into account the lie that you have it takes into account how far you are from the hole it takes into account whether or not you have a clear line to the flag so if you have to like chip out you know something oh i see if you're

like behind a tree or yeah yeah yeah yeah if you're stuck behind a tree again all of my experience comes from wii sports really so we're not really talking yeah yeah um so you can imagine taking this information okay and then combining it with the shot dispersion pattern oh and now what you're basically doing is just trying to find an area with the lowest stroke or with the most strokes gained oh gosh so you have a huge data set of what actually happened and you've got this this map of where what the scoring like expected score from any location is and then you can apply the two and kind of work out see if there are patterns presumably in this data see if there are patents and so so to maybe uh make a video game out of this yes imagine you've got this little you've got this little slider and you

can just go through all your clubs and as you change from you know seven iron eight iron nine iron you know wedge whatever and as you do that you can see the your your dispersion pattern for that club changing and being overlaid on the hole yeah right now imagine furthermore that the area inside your dispersion pattern you take the uh strokes gained from that area and you and you take the average of it and you just show it on the screen right and as you sort of like like let's say that you're you're farther away from the hole you're like you're like 200 yards away from the hole right and uh you're like okay well i could i could hit like my three wood that that goes you know about 200 yards but there's a creek that runs in front of the green so as you go as you

scroll through and you go seven iron you know six iron five iron you know five wood three wood that stroke is going up and up and up right and then all of a sudden you start intersecting with the creek and then it starts going down again right it's like oh because you might hit in the creek and that'll be a one-stroke penalty right and so like and then as you move your aim left to right you know maybe you can hit it that far but take the creek out of play right and that's like the quantification of golf strategy right it's like you can run that monte carlo simulation and say like what is my expected value for this club given my abilities and my dispersion and the actual lie that i'm currently in and the actual place that i'm going to end up in

or like the likely areas and i can weight them however that's all right so now now you've definitely turned it into a video game where now and well in fact what you've turned it into is what our day job is yeah which is sort of quantitative like trying to find the the minimum or the maximum of some kind of of uh of of complicated very very complicated feature set which right exactly what this is right you know right right right wow yeah and and especially like more skilled golfers it's not like they just have like their seven iron shot right they have five or six different shots they can hit with a seven iron and each one of them has their own dispersion pattern so you know you're talking about you know potentially hundreds of different patterns that you could potentially map

on to a hole and so you know as a player or as a caddy especially on tour like you're standing there like i don't think that these guys are like you know mapping these things out quite on the course quite yet but it wouldn't surprise me if that sort of thing starts to maybe happen a little bit it's pretty high stakes stuff so i mean yeah the drive to want to do this kind of stuff so so are there any like general things that come out of an analysis analyzing this data you mentioned right at the beginning that like there are some old folk wisdom things that have been shown to be wrong so is that something in this that can that we can is there a conclusion we can draw from like this kind of data that that would help even like someone as inept as myself absolutely absolutely and you may remember some of these things from your youth being being taught golf there's there's things

about the swing that turn out not to be true and there's things about strategy that turn out not oh okay um some of the things one of the things about the swing that turns out not to be true and there's things about strategy that turn out not okay um some of the things one of the things about the swing that turns out not to be true is i i was taught as a as a young lad uh because this is golf um that the way that you shape a golf shot so the way that you control whether your ball is gonna you know curve to the right or curve to the left right right because that's sometimes important in golf you got to curve it around a tree or some around a dog leg or something yeah something, something like that. Again, Wii Sports is where I've seen this happen. Or not on purpose at the driving range when I'm doing it.

Why is it going backwards? Why did it do that? Yeah, exactly. So the old school wisdom on how to do this, the thing that was taught by teachers and by professionals forever, was that you pointed the face of the club at the place where you wanted the ball to wind up okay yes and then you swung the club along a path that represented the way that you wanted the ball to start so if you wanted the ball to curve toward a target you

would point the club at the target and then if you wanted it to curve from the left to the right you would swing the club to the left effectively opening closing the club face to to point to where you want to go but then your swing direction is where it is is the launch direction effectively got it i think i think that makes sense yep yes yes turns out that's completely wrong okay uh you can you can look at this dad it's basically the opposite of that in a way um the the the initial launch you can see this on a you know high speed camera you can you can verify it with with all these new devices that we have available to us um the initial path of the ball 85 of that initial path is determined by the face angle so the starting line is actually the face angle so So like you hit it with, yeah. Okay.

Even though the club is kind of, yeah, it's hard to describe with that because I'm gesturing wildly at you, but that's fascinating. Where the club face is pointing at impact, 85% of that initial direction is determined by that angle. Got it. Right? Where it curves is a function of the relationship between that angle and the path. Right.

And so the general rule is if you want the ball to curve to a target, neither the face nor the club path should be pointing at that target. What you want to try to do is roughly, and unfortunately it's not exactly this, but roughly you want the club face to be pointing in between the target line and the path of the club. Okay. So as a person who's trying to master this for the first time, I have these memories as a kid of trying to set myself up as a goalie. I'm going to point the club face at the target, and I'm swing along this path and i'm gonna hit it and what i would see is that the ball would start at the target and then curve away from off to the side yeah and i'd be like well i guess maybe maybe i had it wrong what did i do wrong here and it's like no that's exactly you were doing exactly what

you thought you were doing it just doesn't work that way right um how has that been so persistent for so long i it's just it's like the cooking thing it's like oh you got to sear the steak to seal the juices right thinking exactly of searing when i was bringing this up it's like that's not a thing it doesn't it's like no that doesn't work like that at all right it's just one of those things that it's like oh well if you're an expert this is what you say and then that gets passed down and passed down and passed down and then down. And then finally somebody was like, ah, guys, I got this video here. It's 20,000 frames a second.

And you can kind of see it doesn't work like that. So I don't know what you were taught, but it doesn't go. It works by accident only if it works at all. Yeah. So yeah, so that is one thing. Another thing, and this is more on the strategy side, but it used to, you know, people had this pithy saying,

drive for show, putt for dough. And what they meant by that was that putting was really important and driving was sort of seemed cool. And it's like, oh yeah, I hit the ball 300 yards, but it's really not all that. That's absolutely what I, you know, as an extremely terrible golfer, like a friend of mine said, yeah, I learned, quote, bad golf, where, you know, if you're bad at golf,

you can get by doing these things. Just learn to, you know, putt reasonably well. Everything else, you know, I basically play hockey with a seven iron down the fairway till I get into the hole and then just putt again, right? And now I'm good. But I could probably putt with a seven iron too.

I just, that's the only club that I need. They just need a seven iron. Just a seven iron, it's fine, right? As you say, you can play many shots with a seven iron. Yeah, and I mean, you i just that's the only club that i need they just need a seven just a seven iron it's fine right every as you say you can play many shots with a seven iron yeah and i mean you know that's it's honestly that's not a bad way to play golf yeah so but that's not true what how is how how is that not true and it is how do you know it's very not true how do i know it's not true is is the real question we should be asking so this is gets back to mark brody and and his strokes gained approach he wanted to try to figure out what makes

good golfers good at every level. At the PGA Tour level, if you take the top 40 golfers, the best 40 golfers in the world, what makes them better than the rest of them that are on tour? Because the rest of them on tour are so amazingly good, but they're not as good as these people.

He basically analyzed and said, okay, they perform better. I can use strokes gained to quantify how much better they perform and I can break that down by category, right? So how much does putting

actually contribute to that top 40 players additional ability? It's actually only 15%. What? So that is, to drive for show, putt for dough, doesn't really hold up. Sorry, 15% is...

Go through that again for me, because I'm being slow. So let's take the top 40 golfers in the world, and let's compare them to the other professional golfers in the world. They gain more strokes in the strokes gained methodology. Can you imagine if I had strokes gained methodology. Yeah.

Okay. If you say that strokes gained, right. So you like the top 40 golfers, let's just say I'm making these numbers up, but let's say that they gain like, you know, two and a half strokes on average versus all the other ones.

Right. Right. Where does that two and a half strokes? I see. I see. And then you can say where 15% of it comes from putting ice. Okay. So of the overall improvement,

only 15 can be attributed to putting exactly i get exactly okay and driving yep contributes to 28 so almost twice as much okay so it turns out driving is twice as important as putting which again is like so not the received wisdom because and especially you know like there's a non-linearity between accuracy and whacking it as hard as you possibly can so you know maybe yeah but so that might work for professionals but is that true generally yeah so he did the same analysis so so he didn't have the the shot link data but he actually built an app to allow people to track their own shots oh and there are other people, you know, I was talking before about those devices now that you can get that will track your shots.

And those devices have reproduced his results and show the same thing, which is that at every skill level, the thing that differentiates the higher skilled from the lesser skilled players is roughly this. So if we look at on the PGA tour so so we talked about you know putting is 15 driving is 28 driving isn't even the most important thing so the whole drive for show putt for dough it's like well guys you left out the most important thing which is actually approach play so hitting into greens right so hitting from the fairway hitting from the rough into green not on the t and not on the green but like the middle bit the middle bit right

accounts for 40 oh my gosh difference right so that is the dominating factor and it's that's true at the professional level it's also true at the amateur level so for example the difference between a golfer who usually shoots around 90 okay and a golfer that usually shoots around 80 67 percent of their strokes gain of their advantage yeah is from driving and approach play and only 33 percent is from putting and like chipping and sand shots same same kind so again that sort of yeah putting and and the sort of special special shots like there's only a third of it two thirds is is is just yeah decent play in the middle in the beginning that's right amazing that you can get that amount of information and i mean i am genuinely surprised that it holds up in the amateur set because again based on personal

experience i can reliably chip the ball forward 50 yards with a seven iron right if i don't try too hard and it goes roughly where i want it to go but as soon as i try too hard then that's when it's off into the undergrowth and i'm searching around for it so obviously a base level of skill will help you with some some amount but yeah well one explanation for this which makes a lot of sense to me at least as a golfer is that i can hit an eight foot putt you can hit an eight foot puttt. You can hit an 8-foot putt. And you can have some expectation of making that 8-foot putt, right? Yeah. A professional can hit an 8-foot putt.

And they're going to make it more often than we do, but they can do that. That's a shot that they can hit. There's no way in the world that I'm hitting a 7-iron 210 yards. I'm physically incapable of doing that. I can hit a ball 210 yards but it's not going to have that descent angle into the green and the spin rate that will make it

hold and stick on the green. And not just run off forever afterwards, right. Yes. I have to basically hit a perfect shot to hit a shot onto a green from 210 yards and I am just not physically capable of hitting this high arcing spinning shot onto a green from 210 yards that stays there. And so that's one explanation, and I think it's the one that Brody likes and a few other people have said,

as to why this phenomena happens, because that advantage that professionals have, that 40% advantage, is being able to hit approach shots, and specifically longer approach shots into greens. And that is just shots that amateurs just can't do. Like you can hit a chip from 50 yards. I can hit a chip from 50 yards. Tiger Woods can hit a chip from 50 yards. But only Tiger can hit a seven iron 200 yards. And then land and have it stop roughly.

And then land and have it stop. And then that, again, down to the splash zone. Yeah, exactly. The dispersion area. Dispersion area is much down to the splash zone. Yeah, exactly. The dispersion area. Dispersion area, as much as the splash zone, of where the shot's going to land. That's fascinating. And also with driving. You know, Tiger can hit that 320-yard, or at least he could in his prime,

hit that 320-yard, and professionals today can hit balls in 400 yards. I can't physically do that. You could give me 1,000 tries, and I just wouldn't be able to do that very long way yeah gosh yeah so all this stuff comes around to to show us that you know you need data to be able to i mean you can convince yourself anything without data right sometimes you can convince yourself anything when you do have data you know like see see for example benchmarking and micro benchmarks and other other things where you can kind of convince yourself yeah look it's much faster this way only because the branch predictor has got everything right every time or whatever whatever your folly is

there but but in order to actually have some genuine understanding of what's going on and to be able to dispel myths and um and superstitions legends superstitions yeah you do need the data and now you can get it. So is this open data, some of this stuff open data, or is it all closed? So, you know, one of the complaints that I have about all this is that the ecosystem around some of this data is very closed. Like, if you go and you buy these devices

and you record all your shots, I put that in there. That's my data. But the only way that you can really get to it is through their websites or through their apps. And if the app disappears or the company goes out of business or whatever, your data's gone.

And it's like, if I want to load this into an Excel spreadsheet or a Google Doc or a Python program, of course, which I have done. Oh, yes. And analyze it myself. I can't do that. There's one company that I, and I don't want to necessarily endorse anyone on the podcast, but there are companies out there that have more open data access. They'll let you get your data as a CSV file, which is exactly how I do that.

Exactly right. Yeah, right. Yeah. I just want a simple, open, portable format for my data. And then you can go and do your own analysis. And then I can go do my own thing. Yeah. But most of them don't do that. They have these sort of closed ecosystems. Now, it's good for them. One of the companies that is very popular and unfortunately has a closed ecosystem is called Arcos.

And they have half a billion golf shots in their database of amateur golfers that have hit real golf shots on real courses measured with GPS. And I follow the head of data analytics at Arcos on Twitter because he's got some amazing insights into golf.

But I want to do that analysis myself. I want to do that analysis myself. I want to be able to reproduce his results right now. I understand why the company isn't exactly, you know, forthcoming and sharing that. Yeah. I mean, especially aggregated data. I mean, just the privacy aspects alone, right, is a part of it. Because obviously you want to have some idea about what quality level various golfers are

and whatever, so you can partition the data. But I can't even get my own data. That seems unforgivable to me. It is. Now, the launch monitor companies actually, I think, have been pretty good about this, right? Like they have more open formats and you can generally get access to your data.

But all these companies that build like the club tracking, there's only one that I know of that lets you get your data off. And I think that's a damn shame. Well, a good friend of mine, who I'm talking to right now, has told me before now that if you have a website did you have an api so is there any way that you could uh exfiltrate your own data i have thought about that i have thought about that of

course i was like i was a little bit incensed by the idea of okay let me understand how this works so i pay you 200 for these devices and then I pay you $30 a year for a subscription fee and then I have to write the JavaScript bookmarklet that goes and extracts the data off your website and it isn't even all the data that I would want. Like I want

lat longs. I want you know, I want the you know, I want all of the data that I gave you. I just want you to give back to me. And some of that even doesn't show up on their website. So I've been a little reluctant to sort of follow that path. But I do have a system, and the system that I have does allow you to export your data. And I actually just got it a couple of weeks ago, and I'm sort of like –

You're still playing with it. In a candy store. That's great. Because I have like one round of golf in there, and I'm like picking it a couple weeks ago. I'm still playing with it. I'm in a candy store. I have one round of golf in there. I'm picking it apart and going, I wonder if I can calculate this. I wonder if I can calculate that. All of this stuff is kind of amazing.

It's really just advances in technology that have made all of this possible. It is a bold claim, but I really do think that golf has changed more in the last 25 years than it did in the first 500. You know, yeah. Okay. The golf ball changed.

We, you know, they used to play with what's called a feathery, which is basically a piece of leather with feathers stuffed into it. And that's your golf ball. And then they had the gutta percha ball, which was like a rubberized ball. And then, you know, the balada ball, which is the ball that, you know, was like a synthetic ball that golfers played with, you know, in like the 50s and 60s and 70s and 80s. And then it started to change.

And I feel like that was the beginning of both the technological and sort of the quantification change in golf that has created a situation where it's like obsoleting golf courses. Golf courses that were made 100 years ago. Now are not fit for purpose. Now are like, what do we do? We got to move the tees back, but there's nowhere to go. I see. Go ahead.

So I feel like that's a pretty significant change well i mean i'd never thought that we would sit here on a programming podcast doing essentially golf for hackers as we have done right but i mean it's really fascinating and uh i uh i might even be tempted to pick up a golf stick myself well if you ever want to play let me know okay i will do you've been listening to Two's Compliment a programming podcast by Ben Rady and Matt Godbolt

find the show transcript and notes at twoscompliment.org contact us on twitter at twoscp that's at t-w-o-S-C-P. Theme music by Inverse Phase. Inversephase.com.