Does DECADE demand an architectural response?

[David Harshbarger]

Scott Fawcet’s DECADE system uses Mark Broadie’s Strokes Gained methodology to determine optimal club selection and aim point based on individual’s shot dispersion patterns.


DECADE represents a new frontier in data-driven golf as it replaces “strategy” with a robust mathematical decision process. Architectural features such as bunkers, hazards, and rough are reduced to expected strokes values. For any club/aim point combo, the collection of expected stroke values under that shot’s dispersion pattern are summed. The optimal shot is then the club and aim point combo that has the lowest aggregate expected strokes remaining.


The methodology has seen significant success at Pro level. Adherents include Will Zalatoris and Bryson DeChambeau.


My question is does awareness of these strategy optimizing tools, which will become the norm, demand an architectural response for courses that are intended for high level play? An architectural response, for example could be where mathematically optimal play is projected on designs and the designs are altered to remove clear optimal aim points.


Interested in thoughts on this.


Dave

[Erik J. Barzeski]

Scott Fawcet’s DECADE system uses Mark Broadie’s Strokes Gained methodology to determine optimal club selection and aim point based on individual’s shot dispersion patterns.

I mean… kinda. It uses the same idea we wrote in Lowest Score Wins (which pre-dates DECADE and…) - that aiming for your shot variance to slightly safer sides of targets is the best way to create the lowest average score.

DECADE represents a new frontier in data-driven golf

Someone drank the Kool-Aid. 

My question is does awareness of these strategy optimizing tools, which will become the norm, demand an architectural response for courses that are intended for high level play?

I think we've had this discussion before here.

An architectural response, for example could be where mathematically optimal play is projected on designs and the designs are altered to remove clear optimal aim points.

Thing is… there's still going to be an optimal play even if you shift it around. It'll just be different than it was before you created a bunker here, or shifted some tall grass from over there, or whatever.

[Colin Sheehan]

As a college golf coach, I appreciate what Scott has done to brand smart, rational, decision-making. I have spent enough time scrutinizing the data to appreciate that they have simplified it in a way that is very helpful to those aspiring to improve and cut strokes, but the principles aren't anything novel. We were all subjected to years and years of Ken Venturi telling the audience that Hogan always aimed away from the flag on the 11th green at Augusta. People can learn the essence of Decade naturally from playing competitive golf...whether they are capable of grasping it intuitively is another matter. 

[Michael Felton]

I'm not sure that's a great idea. For the most part DECADE is about minimizing your expected score in a given situation. For approach play, that's likely to err on the conservative side. To tempt someone using this more, you have to lower the expected score by going for the flag. That means either you move the flag more into the green or you soften the penalty for missing. I don't think softening the penalty is a good idea and if you move the flags to the middle of the greens, then you're squashing strategy as a whole.


I'd think the best way to counter it is to increase the penalty for missing on the fat side of the green. There are a handful of greens out there where you're better off missing them than hitting them but in the wrong spot. DECADE would need to be more nuanced to deal with that. At present it's a decision tree, which is somewhat simplified to enable players to make decisions on the fly. It would be quite challenging to rework it such that it handled an awkward green well, but was still digestible for the rank and file player.


Having said that, I think you can still ask questions of a player who's using that. Tempt them with a carry that they can make 70% of the time. The other thing is to make it as firm as possible, so angles do start to matter more.

[David Harshbarger]

I'm not sure that's a great idea. For the most part DECADE is about minimizing your expected score in a given situation. For approach play, that's likely to err on the conservative side. To tempt someone using this more, you have to lower the expected score by going for the flag. That means either you move the flag more into the green or you soften the penalty for missing. I don't think softening the penalty is a good idea and if you move the flags to the middle of the greens, then you're squashing strategy as a whole.


I'd think the best way to counter it is to increase the penalty for missing on the fat side of the green. There are a handful of greens out there where you're better off missing them than hitting them but in the wrong spot. DECADE would need to be more nuanced to deal with that. At present it's a decision tree, which is somewhat simplified to enable players to make decisions on the fly. It would be quite challenging to rework it such that it handled an awkward green well, but was still digestible for the rank and file player.


Having said that, I think you can still ask questions of a player who's using that. Tempt them with a carry that they can make 70% of the time. The other thing is to make it as firm as possible, so angles do start to matter more.

This makes sense to me, to create more uncertainty on the good misses on approaches with more punitive features than their strokes gained expectation would suggest.


I wonder if anyone has experience with designs that frustrated sound decision making that either were clearer when a formulaic process was applied, or were worse? Pete Dye designs are said to frustrate many pros, does the algorithmic approach “see” his courses more clearly?

[Peter Pallotta]

Dave: a good topic, in many ways - thanks.
I've come to suspect that great golf courses have an internal logic and self-referential unity that has little to do with how we play the game.

[Kyle Harris]

The entire point of golf is to integrate with as little of the golf course as possible.


Are you really *playing golf* if any action you take is other than lower your expected score for the hole? DECADE is as much about managing those expectations based on what you can do as it is about going low.

[Tom_Doak]

I thought a lot about this when we worked on Memorial Park.


In the end, what makes the golf course tough is the combination of Bermuda rough with unpredictable lies, and greens that approach shots run away from.


Strategically, there is little that you can do to move the needle.  The one thing that these systems don’t respond well to is a hazard in the middle of the fairway - in essence they have to treat it as a cross bunker.  But some players are so sold on the system that a hazard that doesn’t compute is harshly criticized as unfair, the same way you would talk about a big cross bunker right in your landing area.  Some young players seem to believe that the right to a 60-yard target area is inviolable!  I wouldn’t make any of them a favorite at The Old Course.

[Cal Carlisle]

Some young players seem to believe that the right to a 60-yard target area is inviolable!

Great word.

[Luke Sutton]

The 70% risk will not do much with Decade players… none will take the bait. You are taught in Scott’s Decade class to never take the bait unless you have a 90% chance of pulling it off. Jack would have said 100%.


I do have some ideas for taking this out of the game thru design but it would make for terrible golf courses. Like Tom said with cross bunkers. One of my ideas was to have ZERO fairway from 220 off the tee until 50 yards to the green. You can always make the landing areas in the 290-350 range 10 yards wide with impossible bumps and tall grass (bunkers would make it easier not harder). But for the old guys who are there in two this is a terrible design. Maridoe in Dallas comes to mind when I think of theses ideas but not to the extreme I’m talking about.

[JESII]

I wouldn’t make any of them a favorite at The Old Course.

But isn't this the answer?


I wonder how the Decade system responds to abundant but random features and firm ground.

[Jeff_Brauer]

IMHO, yes, architecture should consider how top players really strategerize to be most effective, as opposed to continuing to design for how top players might have approached strategy in the 1920s.  It makes a lot of sense to design for the future, and there is really not much future in looking to the past, LOL.


I mentioned this recently, but there is supposedly a radio interview of Bobby Jones out there, where he said that the 1935 Augusta brochure touting his 4 reasons to risk a hazard on the tee shot was written by Mac and his strategy was to aim as far from any fw hazard he could and still be in play.  I really don't know too many players who do challenge fw hazards, unless their laser guides tell them that clearing it is a 99% certainty, just as they won't aim at a tight target unless it is within 90+% of their lateral dispersion pattern.


I asked one architect if the Golden Age design theories had ever really been tested and analyzed and his answer was yes.... by Scott Fawcette, LOL.


With this taking hold, (seemingly) my gca question to Fawcette or his followers would be whether a corridor width just shy of their near certain dispersion pattern would tempt them to hit the driver rather than lay up.  The few I have asked have said no.  They need 90 whatever % and never cheat on their numbers, but I have to believe if they are confident in hitting max length if the LZ is 64 yards wide, that maybe a corridor of 60-63 yards would tempt some to take chances they shouldn't, and raise their scores.  I doubt any of them would take a driver if it was, say 55 yards wide or less.   

[Blake Conant]

They need more data on the affect of topography. I’ve chatted with Kevin Moore a bit about blind shots and side hill lies and the shot penalty for each. Although the data is anecdotal, he guesses there’s roughly a 0.1 - 0.15 shot penalty for a blind shot. Think the penalty was similar for a stance on a 5-10% slope.

[Peter Pallotta]

I wouldn’t make any of them a favorite at The Old Course.

But isn't this the answer?
I wonder how the Decade system responds to abundant but random features and firm ground.

Yes.
I was thinking along the same lines when I posted about great courses having their own internal logic etc.
I wonder if the decade system works best when the architecture is most clear cut, ie not 'obvious' necessarily, but well planned and externally oriented.
PS - I'm not explaining myself well, and maybe the idea is bonkers anyway. But I think my point relates to yours: the Old Course, played in reverse, when it hasn't rained in 3 months, would serve as a fine test case

[Jeff_Brauer]

I don't know if any strategy system can take all the thinking out of a shot, even if that is the aim.  The knobs on a hole like Prairie Dunes 8 reduce the landing zone to a very small area, even if tree to tree (or native to native) is pretty small.  Fairways like Olympic would also have to be factored in by the golfer.


But I do think that the system is meant to defeat "internal course logic" as any good strategy should.  The golf design nerds hate it, but that is the holy grail for better golfers, I suspect.

[Michael Felton]

I thought a lot about this when we worked on Memorial Park.


In the end, what makes the golf course tough is the combination of Bermuda rough with unpredictable lies, and greens that approach shots run away from.


Strategically, there is little that you can do to move the needle.  The one thing that these systems don’t respond well to is a hazard in the middle of the fairway - in essence they have to treat it as a cross bunker.  But some players are so sold on the system that a hazard that doesn’t compute is harshly criticized as unfair, the same way you would talk about a big cross bunker right in your landing area.  Some young players seem to believe that the right to a 60-yard target area is inviolable!  I wouldn’t make any of them a favorite at The Old Course.

That central bunker thing it depends on how big it is. I specifically asked Scott Fawcett about the 13th at Friar's Head, which has central bunkers, but they're small. He said just aim at it. If the bunker(s) are small, then you just accept that sometimes you're going to go in them, but laying back is going to cost you too much over the long run vs going for it and hoping. In the event I was asking him about, that hole played downwind one day and into the wind the other. Downwind I went in them (because of course the only way I can hit my driver straight is by aiming at trouble) and into I couldn't reach them anyway.


I think though where that is useful is it's harder to commit to actually trying to hit it into trouble.


The other thing I think is trying to get your space between trouble to be right on the edge of where the decision tree flips. So for example, one of the decisions is do I have 40 yards between bunkers, so if you have 41 yards between bunkers, but the bunkers are nasty deep sideways play out of kind of things, that's going to nudge things towards the mathematically correct option being a little different from what you expect. Or make them a little less than 40 yards, but have them be relatively less penal. So 35 yards apart, but flat and open bunker without big lips. The other good thing this does is it enables the player who is feeling good about his/her game that day to take on a little more. If they do hit a good shot in that situation, then they can gain against the mathematically correct, but slightly inferior golfer. At the end of the day, for the tour players we watch on the TV, that's really what we want.


Having said that, architecturally speaking I don't think DECADE changes anything. Interesting holes are still interesting and they'll still differentiate who's playing the best at a given time. Trying to come up with some way to nullify the benefits of playing by some pre-determined formula seems to me like a good way to create less interesting courses.


Side note - which holes at the Old Course have less than 60 yard wide playing areas?

[David Harshbarger]

I thought a lot about this when we worked on Memorial Park.


In the end, what makes the golf course tough is the combination of Bermuda rough with unpredictable lies, and greens that approach shots run away from.


Strategically, there is little that you can do to move the needle.  The one thing that these systems don’t respond well to is a hazard in the middle of the fairway - in essence they have to treat it as a cross bunker.  But some players are so sold on the system that a hazard that doesn’t compute is harshly criticized as unfair, the same way you would talk about a big cross bunker right in your landing area.  Some young players seem to believe that the right to a 60-yard target area is inviolable!  I wouldn’t make any of them a favorite at The Old Course.

That central bunker thing it depends on how big it is. I specifically asked Scott Fawcett about the 13th at Friar's Head, which has central bunkers, but they're small. He said just aim at it. If the bunker(s) are small, then you just accept that sometimes you're going to go in them, but laying back is going to cost you too much over the long run vs going for it and hoping. In the event I was asking him about, that hole played downwind one day and into the wind the other. Downwind I went in them (because of course the only way I can hit my driver straight is by aiming at trouble) and into I couldn't reach them anyway.


I think though where that is useful is it's harder to commit to actually trying to hit it into trouble.


The other thing I think is trying to get your space between trouble to be right on the edge of where the decision tree flips. So for example, one of the decisions is do I have 40 yards between bunkers, so if you have 41 yards between bunkers, but the bunkers are nasty deep sideways play out of kind of things, that's going to nudge things towards the mathematically correct option being a little different from what you expect. Or make them a little less than 40 yards, but have them be relatively less penal. So 35 yards apart, but flat and open bunker without big lips. The other good thing this does is it enables the player who is feeling good about his/her game that day to take on a little more. If they do hit a good shot in that situation, then they can gain against the mathematically correct, but slightly inferior golfer. At the end of the day, for the tour players we watch on the TV, that's really what we want.


Having said that, architecturally speaking I don't think DECADE changes anything. Interesting holes are still interesting and they'll still differentiate who's playing the best at a given time. Trying to come up with some way to nullify the benefits of playing by some pre-determined formula seems to me like a good way to create less interesting courses.


Side note - which holes at the Old Course have less than 60 yard wide playing areas?

On your challenge avoiding the bunker playing downwind, what I’ve seen is that shot dispersions are very sensitive to wind. A helping wind reduces dispersion. The “aim at the bunker and hope” strategy concentrates more shots around your aim point, so in a tail wind the dispersion reduction may make one side of the fairway a legit target as winds pick up.

[Michael Felton]

On your challenge avoiding the bunker playing downwind, what I’ve seen is that shot dispersions are very sensitive to wind. A helping wind reduces dispersion. The “aim at the bunker and hope” strategy concentrates more shots around your aim point, so in a tail wind the dispersion reduction may make one side of the fairway a legit target as winds pick up.

Yes that's true. The downwind that day didn't feel like enough to really impact noticeably the dispersion though. It was a gentle breeze. The into day was more extreme wind wise. Those bunkers are pretty small too. There are two of them and they're both about 3 yards wide. 6 yards in a 65+ yard pattern is roughly 1 in 10 you'll go in them. If you do it costs you maybe 3/4 of a shot. If you lay up 30 yards shorter, you're looking at a roughly 0.2 shot cost each time, so over ten rounds you're giving up 2 shots for a cost of 0.75 shots. Playing to the sides brings nastiness into play on either side. The fairway is right around 67 yards wide there. Aiming 10 yards right of the bunkers is going to put maybe 15% of your shots into stuff that could cost you 0.75 shots again, so I don't think aiming to the side is optimal.


But, at the end of the day, for the three different types of architecture (strategic, penal and heroic I think we talk about), all DECADE does is tell you which strategy is right for strategic, doesn't really help with penal - you just got to hit the shot and says either yes go for it, or no don't with heroic. From a strategic bent, I think you want to try to play tricks with the mind. An angled fairway for example, where it's not obvious where to aim, even if you have done your homework.

[Jim Sherma]

From a dynamic programming viewpoint, it seems to me that the only response is to create landing areas that are smaller than the expected dispersion pattern. Augusta at Masters time seems to accomplish this via slope and green speed. The landing areas are so demanding because slight misses end up rolling out such that the scoring dispersion between just right and almost right is widened. The makes one aim at smaller targets than they would feel comfortable with. US Open rough with narrow fairways likely act the same way.


If shot patterns truly are random draws from a fixed distribution, then tournament results would be essentially randomized outcomes. The player who happens to get more draws in the "correct" parts of the distribution will win while the other player who hit all of the "same" shots but happened to get enough draws in the wrong place will lose. I'm guessing that it has always been that way. I'm sure that some great players just have much tighter distributions or can tighten them when needed (Nicklaus & Woods), while others have the inability to maintain their distribution range when pressured (e.g. Norman).


To summarize my thoughts, creating situations where there is enhanced shot outcome value differentials within the dispersion area would be the architectural response for defending par given DECADE type strategic playing. Becomes very difficult to create a course that is playable for a wide range of players given how much tighter the dispersion of top level players are than 3+ handicap players, let alone 10+.

[Tom_Doak]

To summarize my thoughts, creating situations where there is enhanced shot outcome value differentials within the dispersion area would be the architectural response for defending par given DECADE type strategic playing. Becomes very difficult to create a course that is playable for a wide range of players given how much tighter the dispersion of top level players are than 3+ handicap players, let alone 10+.

This is precisely why my dream is to build a course *without* so many forward tees. The only way to turn up the difficulty of those target areas on the greens for elite players is if the rest of us will be playing the hole like a par-5 and chipping at those targets.  Putting the white tee further forward so we can hit 8-iron at those targets is still too hard; maybe a few of the holes become 300-yard par-4’s but the rest should stay longer and play like 5’s for Colonel Bogey.


This is not a radical new idea; it’s exactly what Colt or Mackenzie were doing 100 years ago before someone decided to build 4-5-6 sets of tees, and it worked just great.

[Jason Topp]

I have never subscribed to this system but I think I know enough to get 80% of it.  Off the tee - hit driver if you have 60 yards of width and maximize your chance to have a clean shot at the green within that width given your shot pattern.  Do not worry about the angle of your approach.


Into the greens, don't chase pins and maximize your chances of hitting the green given your shot pattern with that club.


Avoid penalties and situations that impose penalties at all costs.  Over time, taking the risk does not pay off. 


It seems like the best way to play.  I have no reason to doubt the math. 


Do adherents find the game less interesting or is it just as fun to aim at the target the system suggests even though the temptation exists to be more aggressive?  Does a course that traditionally tempted players to take on a bunker or some other hazard not tempt adherents to the system?

[Jeff_Brauer]

So the motto of DECADE is, "As it turns out, I DID drive all this way to lay up!"

[Michael Felton]

Do adherents find the game less interesting or is it just as fun to aim at the target the system suggests even though the temptation exists to be more aggressive?  Does a course that traditionally tempted players to take on a bunker or some other hazard not tempt adherents to the system?

Speaking for myself, I don't find the game any less interesting. I have a numbers brain, so I enjoy figuring out the "correct" target for a given shot. My enjoyment comes from hitting the shot I was trying to hit and shooting a score. To the extent that it helps me shoot a lower score, it enhances my enjoyment. Plus I've found that getting tempted into taking on a bunker or something else more often than not results in a bad outcome and that's frustrating. So what I think the outcome is is I hit fewer dumb shots and my scores are better. What's not to enjoy about that?


DECADE is pretty aggressive off the tee. It's almost always driver off the tee, and only not driver if there is a good reason not to. It's more conservative on approach than most people would naturally play. It pretty much boils down to hit it as close to the hole as you can without taking on undue risk in the process.

[Peter Pallotta]

One thing I know, as a recreational golfer: that not all pars are created equal. Sometimes the extra challenge and variety and satisfaction is engendered by the architecture; but just as often it is engendered by my own shot selections. When you play the kind of golf courses I do, *not* routinely attempting a fade off the tee on a slight dogleg right and then a mid-iron draw to a tucked left side pin, or vice-versa, eliminates about half the fun from any given round of golf. Success and failure and to some extent even score are often secondary in importance to the attempt, since even for an average golfer the inelegantly plodding par doesn't hold all that much charm -- at least, not on a bland Doak 2-3 it doesn't.

[Thomas Dai]

How do these analytical approaches to the game take into consideration playability factors like wind, rain, firmness of the greens etc?
Atb

[Tags:]

Tags:

[Tags:]

Tags: