We recently discussed the Doak "0" rating.  A far more interesting discussion can be had discussing the Doak "6" rating:

6  -  A very good course, definitely worth a game if you're in town, but not necessarily worth a special trip to see.  It shouldn't disappoint you.

Tom's verbal description of the rating system places a lot of emphasis on the distance one should travel to see/play a course.  In general, I believe his goal is to develop a 0-10 scale with a normal distribution, centered around 3, so that the median golf course is a 3.

So let's take out the travel descriptions for a minute:

5 - A good course
6 - A very good course
7 - An excellent course

Everything over 5 is recommended.  So here are a few comments to get things started.

1.  Everybody believes their home course is at least a 6.  Tom gave the two Pumpkin Ridge courses a 6 and a 5, whereas I see them as 6s.

2.  It seems there are a lot of solid (very good) courses being built these days.  Are there a lot of 6s out there now?  Do we move the curve so an average course is closer to a 4 now?  Or do we toughen our grading process so we (he) give 6s a little less generously?

3.  The big jump in Doak ratings happens when a course is rated 7.  Once a course is rated 7, you know it's special.  What are the reasons behind a 7 rating?  I say a big difference with an excellent golf course is the decision making the player faces, how much he has to consider his shots.

OK.  Hope this generates interest.

John,

I like your idea of thinking about this statistically, but that means I have to get mathematical on you. Excuse me while I achieve a high dork factor with the following comments.

I can't make a normal distribution work.  If you have a mean of 3 you need a large standard deviation to have values up to 10.  Then you will also have values below 0, and we know the Doak scale is all positive numbers.  

Ideally, we should analyze all the data in The Confidential Guide to devise a custom distribution that is the best fit for The Critical Mind of Tom Doak.  I'll get right on that.  

Better than the normal is a triangular distribution, with a most likely value of 3, a minimum of 0 and a maximum of 10.  

My calculations indicate:
*Both the mean and the median of that distribution would be about 4.  
*Less than 2% of values at 9 or better, less than 10/1000 courses are 10s.  That seems about right.  
*1 of 8 courses would be 7 or better, and 1 of 4 courses would be 6 or better.  Does that work for you?  

For me in addition to the factors you mention, I'd say a "special"  golf course is one I can't wait to play again.  They draw you in for another trip around.  

Playing at Bandon Dunes with one of my pals on his first visit, my second, I pressed him to name his favorite course after we had played all three.  He said, "The one I just played."

I would guess the number of courses Tom has rated "7" is much closer to 150 than 500.  Maybe I'll go ahead and count, or see if he has that listed in his book.

It gives some perspective on the nature of great courses when you realize we're talking about the best 1-2%.

John,

I like your idea of thinking about this statistically, but that means I have to get mathematical on you. Excuse me while I achieve a high dork factor with the following comments.

I can't make a normal distribution work.  If you have a mean of 3 you need a large standard deviation to have values up to 10.  Then you will also have values below 0, and we know the Doak scale is all positive numbers.  

Ideally, we should analyze all the data in The Confidential Guide to devise a custom distribution that is the best fit for The Critical Mind of Tom Doak.  I'll get right on that.  

Better than the normal is a triangular distribution, with a most likely value of 3, a minimum of 0 and a maximum of 10.  

My calculations indicate:
*Both the mean and the median of that distribution would be about 4.  
*Less than 2% of values at 9 or better, less than 10/1000 courses are 10s.  That seems about right.  
*1 of 8 courses would be 7 or better, and 1 of 4 courses would be 6 or better.  Does that work for you?  

12.5% of all courses are 7 or better?  If there are 26,000 U.S. courses, that means 3250 courses are at least 7's.  Sounds way high to me.  

If the same distribution applies globally, there are at least 7500 courses rated 7 or better in the world.  Again sounds way high.  

On your other points: I know very little about statistics.  That said, I think you are not using the math correctly.

Doak's scale, as I understand it, is really just a ranking.  It says a 10 is better than a 9.  It does not say by how much.  A 10 might be 30 times better, or 0.1% better.  There is no way of knowing or calculating.  

And Doak's scale does not really say 10 = 10, or 9 = 9.  Does Tom think all the courses he rated as 10's are equally good?  What about all the 9's, or 8's, or 5's or 3's?  

There is no mathematical rigor in the scale.  It's just Tom's subjective way of putting courses in a kind of ranked order.  So it doesn't seem to me that calculating an arithmetic average, or standard deviations, means anything in this case.  Similar to how they mean nothing for Golf Magazine's and Golfweek's rankings.  

In Golfweek, e.g., you get the apparent anomaly that the 3rd highest ranked modern course, Whistling Straits, gets an average score of 8.59.  That score should disqualify it from the top 3, the top 5 and the top 15.  This is what happens when you try to apply arithmetic means to what is really just a ranking of order.    

Like I said, I know next to nothing about statistics, so maybe I'm way off base about all this.  

Statistics aside, I think average folks have a mental block using a scale of 0-10 that doesn't use smootheven distribution.   That may be where the Doak scale struggles a little bit.  I think we all want to believe that our home club is at least in the top 10 or 20% of the courses in our respective countries.  So putting the emphasis on a large group of courses at 6 is hard.  

We all want to believe our girlfriends/fiances/wives are at least a 9, why not our golf courses?  Might not be reality, but...

OK, so my triangular distribution model doesn't work.  But one of the great things about golfclubatlas.com is it's easy to find out how wrong you can be!  

Jim N,

For every set of numbers that you suspect describes random variation you can try to fit a distribution to the observations and then do some statistical calculations.  There are several possibilities but none of them may fit.  When you say those statistics are meaningless, you may not be analyzing a random variable.  In Doak's work for the CG, the inputs are not randomly selected, as Doak chose the courses he evaluated, he didn't pick them out of a hat. The output is certainly not random, as the observer has (at least) some biases that figure into the scoring.  He may also be determined not to give more than a few courses 9 or 10.  His distribution might look completely different than yours or mine, but I thought around here we accepted his as the gold standard  and John's comment about his results being a normal distribution sparked my just-for-the-hell-of-it alternative statistical evaluation.  Maybe among the 28,232 (?) courses Tom hasn't seen, we could, based on statistical analysis of his accumulated data, predict that there might be one or three more 10s out there.  But he might say he's definitely seen all the 10s already.

So if it's not random and all subjective, that's fine with me, and if my answer to the question "what makes a 7 stand out" apply to me only, that's fine too.  What's your answer?  

(I carefully wrote this before Tom D. responded, so I'll post as written.  I see he has refuted some of what's here)

As advertised, I flipped through the pages of the Confidential Guide, counting 7s.  There are about 100 in the book.  It was an interesting exercise.  Tom Fazio is the modern architect with the most courses ranked 7 or better, though the courses Tom Doak selected as 7s (Wild Dunes, Wade Hampton, Pelican Hill, for example) are not the current darlings of the "best of" lists. You would either conclude that Fazio builds a lot of 7s, or that much of his his modern work would be regarded as 6s.  Nevertheless, there's a great deal of respect there.

Let's say that Tom missed a few in his travels.  Add in the second Golden Age courses from 1995-2006, plus some sympathetic restorations of classic courses, and maybe there are 150 courses worthy of a 7, with about 80-100 in the U.S..  That's 0.5% of all courses.  Eric, I think we're still looking at a bell curve centered around 3 or 4, truncated at 0.

There was a recent thread that the "best of" lists experience a significant drop in quality about halfway down the list, that any of 300-500 courses could represent the 75th best modern course.  This would support Doak's delineation between 6 and 7, that once you get two thirds of the way down the list, you run out of excellent courses.  I think it shows that raters need to figure out a little better what the next best courses really are.  Where does "very good" actually become "excellent"?

I'd like to think there are quite a few 7s being built these days.  Let's take an example.  I played Colorado GC once last year, and though it did not thrill me, it is an excellent course, with a great deal of variety and challenge, with a reasonably interesting walk.  It is also artistically built and shaped into the terrain.  I thought it was about a 6 when I played it, but a strong case can be made to call it a 7.  Anyway, this is me making my usual case for the modern architects.  They don't get the best pieces of property, but a lot has been learned in the last 20 years, and excellent courses are being built.

All of this analysis of Tom's method may drive him crazy.  It's his own personal rating system which became popular with his friends, so he published it.  On the other hand, the Golfweek rating system is similarly designed, so the method has been adopted.  
 


 

 I'd like to think the typical course by the greatest architects are worth seeing, regardless of travel distance.

John, I want to agree, but last time I was in Traverse City with some golf enthusiasts, I didn't find much enthusiasm for traveling to High Pointe, especially in comparison to the (admittedly spectacular) available alternatives.  

Any discussion of distributions around all 32,000 courses is tough to gauge from just counting the courses rated in The Confidential Guide.  Despite his best attempts, Tom could not quite play or rate all of them.  There are also many new courses and many that have been altered since he saw them.  Most importantly, even though he hits great spectrum of courses, the rankings probably skew toward better (or at least better known) courses, so that the average in the book is almost certainly higher than the average of the whole population.  It reminds me of slope ratings.  The average slope is supposed to be 113, but I don't think I have ever played a course that has a slope below that.  There must be a lot out there, but they don't exactly draw a crowd.

Despite all of that, I agree that 7 seems to be an important step from "good course that is worth playing if you happen to have the chance" up to "very good course that is worth going out of your way to play."  That distinction seems to be the real test of what is "special".  However, even if the highest rankings are more subjective, it feels like it starts to get very selective above 7 (as it should).  Based on how few reach each subsequent level, the ratings seem to rise more in line with the Richter Scale.  A 9 is not just a little better than an 8.  It is multiples better (although "better" as always, is in the eye of the beholder).  

http://www.linksmagazine.com/best_of_golf/features/the_doak_scale.aspx

7. An excellent course, worth checking out if you get anywhere within 100 miles. You can expect to find soundly designed, interesting holes, good course conditioning and a pretty setting, if not necessarily anything unique to the world of golf.

I have always looked at it as a basis of travel. 100 miles from my apartment are:

* North - Misquamicut - a unique Ross is a special setting on the Rhode Island ocean/sound.

* East - Atlantic - a Rees Jones top 2-3 in his portfolio?

* South - Huntington Valley CC - The best 100% Flynn? (never saw Lancaster).

Quote from Tom Doak.

To me, the point is moot: The West is a 10 on the Doak Scale and the Composite is slightly better.
The link to the whole article
http://www.linksmagazine.com/golf_travel/international/australia/royal_melbourne_golf_club.aspx

If some thing is slightly better then a 10. I guess that would make it a 10.5 or a 11.So maybe there room for even better then "Nearly Perfect".


10. Nearly perfect; if you skipped even one hole, you would miss something worth seeing. If you haven’t seen all the courses in this category, you don’t know how good golf architecture can get. Call your travel agent—immediately

I think that the bulk of modern courses (if not the majority) may rate a 6 on the Doak scale.  They're all well constructed, impeccably groomed, and few of them have really stupid features.  (Indeed, some of the courses which I rated a 0 would be a 6 for others.)  There are a lot more 6's in the world than there were when I wrote The Confidential Guide, and my own associates say with some confidence that we ought to be able to build a 6 on any site, if we just go out and do a good routing and build a good set of greens.

If it's a below-average example, handicapped by the setting or by housing or by a mailed-in design, then it's probably a 4 or 5 instead.

So then what's a 7?  A 7 is a course that sets itself beyond the typical Tom Fazio or Jeff Brauer or Tom Doak course.  If it's just an average example of our work, then it isn't worth traveling to play, any more than any other of our courses.  A 7 has to have something about it that's really special -- a unique setting, a better set of greens, a couple of truly great holes, etc.  Almost by definition, that would make it a legitimate contender for Best New Course of the Year (as opposed to all the 5's and 6's which advertise that they were "nominated" for the award).

Helps to post the quote

I think I agree with Jeff S.
Is this about what you said?:

The Doak scale is best in determining if you want to come to Houston, not which course to play.

I live in Houston.
Currently there is no course higher than a 6 in the city.
There are many 6s that I enjoy playing.
There is definately a heirachy to the 6s.

If someone was visiting me I wouldn't just say they were all the same - play any of the 6s, it doesn't matter.  I'd take them to play my favorite which I think is significanly better than the other 6s.