[james soper]

Doral Blue Monster
Pinetree
Burning Tree

[Andrew Hastie]

How about "De Pan".

[Sean_A]

1.  It makes NO sense to nominate a course unless you yourself would vote for it as one of the top 100 in the world ... otherwise it's just wasting everyone else's time.  Do all the nominations so far really qualify by this standard?

Tom

How do we know we are voting for a top 100 course until the results are in?  Just because you or anybody gives a course an 8-10 doesn't necessarily mean anything until folks back it up.  Isn't that the point of a panel VS the Wardian Wizard System?

Ciao

[Carl Nichols]

Has Gilette Silver voted?

[John Kirk]

I've seen 22 of Golf Magazines's top 100.  Not all of us have seen 100 great courses.  I thought we should include all courses that potentially deserve a top 100 rating.  I don't know what the cut line is.  I suppose if you think a course deserves a 7 it should be on the list.

I haven't see Alotian or Madison, but I'm also very curious how our members would rate them.

By the way, this is how the list will become skewed.  People who haven't seen a broad cross section of courses will give a ranking of 7 to courses that well traveled raters would give a 5 or 6.

[Tom_Doak]

Sean / John:

If you list every course which two people think deserves a "7", the ballot would be over 1,000 courses.  Do you really need to do that?

If you have seen 22 of the top 100 courses in the world, as rated by anybody, it is probably fair to assume that you shouldn't give an "8" to more than 25 or 30, max.  If you've seen 90 of the top 100, it is ridiculous to assume that you should vote for more than 90-95, although there's always somebody who will vote for 125 in the top 100.

It should take an average vote of between 7.2 and 7.5 to make 100th place on the final tally.  Below that, you are right, the list will be skewed by "7" votes from panelists who haven't seen too much ... but that's why the #200 course should not be taken too seriously.  Most of the top 100 will have been seen by enough people that they won't be propped up by voters with less broad exposure to the world's best.

[John Kavanaugh]

I can't see why nominations should be limited anymore than Golfweek which has over 1400 nominated courses for the United States alone.  Gillette Silver is not a member of GCA and has no interest in becoming so...Gillette is having a difficult time dealing with the recent death of his beloved Mother and as has been said on this site...Is getting his priorities in order.

[Sean_A]

Sean / John:

If you list every course which two people think deserves a "7", the ballot would be over 1,000 courses.  Do you really need to do that?

If you have seen 22 of the top 100 courses in the world, as rated by anybody, it is probably fair to assume that you shouldn't give an "8" to more than 25 or 30, max.  If you've seen 90 of the top 100, it is ridiculous to assume that you should vote for more than 90-95, although there's always somebody who will vote for 125 in the top 100.

It should take an average vote of between 7.2 and 7.5 to make 100th place on the final tally.  Below that, you are right, the list will be skewed by "7" votes from panelists who haven't seen too much ... but that's why the #200 course should not be taken too seriously.  Most of the top 100 will have been seen by enough people that they won't be propped up by voters with less broad exposure to the world's best.

Tom

I don't think there is much fear of reaching 1000 candidates.  I personally was using 7 as the cutoff for nomination because of the average from the 1st poll.  I think your list essentially lists 8s as top 100, but there is plenty of room for debate between numbers.  IE I could think a course is an 8 and you may think it a 9 etc. 

There is no fear of me giving out too many 8s and 9s - they can probably be counted on one hand.  I spose I am the teacher fighting grade inflation - not because I am mean, but because I really do think most top course are over-rated.

You are right that folks should use some common sense based on the number of top courses they have seen in relation to scores that would qualify a course for top 100. 

There is no fear of me giving out too many 8s and 9s - they can probably be counted on one hand.  I spose I am the teacher fighting grade inflation - not because I am mean, but because I really do think most top course are over-rated.  Meaning, they aren't that much better than 7s - not enough to be clear on why they are better from any sort of objective analysis.  They all tick the right boxes, but 8s and above (based on personal opinion and bias) nay have a few bold ticks and a few extra ticks.  While many might feel a 7 has those same attributes. 

Ciao

[Adrian_Stiff]

The best way to do this type of thing is for people to vote regionally. Most people here are familiar with the courses in their state or county and would be knowledgeable enough to be able to grade the best 20. If you have only played 20 (which is lots) out of a top 100 you are not qualified to judge. I would rather see a GCA list of courses 'Doak scaled' by the members on this forum and perhaps as soon as a course gets rated by 3 GCA'errs its a solid rating.

[henrye]

So what happens if I think some courses have no business being on the list and I would only rate them as a 5 or 6?  Should I give them those scores or will someone think I'm being too harsh?  I can't be the only one with this issue.

[Mark Pearce]

If you list every course which two people think deserves a "7", the ballot would be over 1,000 courses.  Do you really need to do that?

If you have seen 22 of the top 100 courses in the world, as rated by anybody, it is probably fair to assume that you shouldn't give an "8" to more than 25 or 30, max.  If you've seen 90 of the top 100, it is ridiculous to assume that you should vote for more than 90-95, although there's always somebody who will vote for 125 in the top 100.

It should take an average vote of between 7.2 and 7.5 to make 100th place on the final tally.  Below that, you are right, the list will be skewed by "7" votes from panelists who haven't seen too much ... but that's why the #200 course should not be taken too seriously.  Most of the top 100 will have been seen by enough people that they won't be propped up by voters with less broad exposure to the world's best.

Are you now abandoning the original definitions of the Doak scale in favour of a definition linked to World Rankings?  Because if you're not then the number of 7s, 8s or whatever will be subjective, not objective, because the Doak Scale is subjective not objective.

There's also a circular argument in the second part of what you write.  How do I know how many of the World's Top 100 I have played?  If I calculate that from an existing list, I'm assuming that list is right but then this whole exercise is redundant.  Better not to work on the basis of an existing categorisation.  Surely that's the magic of the Doak Scale?

I'm happy that Silloth is UK Top 50, possibly Top 30, so that nomination, at least, is quite serious.  Others may well disagree but that's what this thread is about, surely.

The fact is that panel ratings (whether for golf courses or wine) are flawed because of the subjective nature of our assesment of both.  I'd rather find a critic whose palate (or taste in golf courses) seems to calibrate with mine and rely on his (or her) judgment than work off a panel ranking.  Equally, if Parker marks a wine very highly and uses certain words to describe it (gobs of fruit!) I'm likely to dislike it, or at least rate it rather lower.  I'm sure the same thing will apply with golf courses, though I haven't yet found a reviewer whose opinion strongly correlates against mine.  I suspect though, from what I have read, that I'd be better off following an Arble rating, for instance, than a Ward rating.

[Adrian_Stiff]

I suspect a World top 100 would only consist of courses with a '9' rating possibly a few '8's. I love Boat of Garten but its probably nowhere near in the best hundred golf courses in the world. Its not even top 100 in the UK. I would think 20-25 courses from GB & I would get in a 100. From the rest of Europe it may be a small handfull. Whtever happens., no ones going to agree.

[Anthony Fowler]

1.  It makes NO sense to nominate a course unless you yourself would vote for it as one of the top 100 in the world ... otherwise it's just wasting everyone else's time.  Do all the nominations so far really qualify by this standard?

Saticoy and San Diego CC seem highly unlikely choices.

Tom, I appreciate and understand your comments to some degree.  Unfortunately, I have not played enough courses to say whether I would rate San Diego and Saticoy inside the world top 100.  However, I do like these courses over several courses that are widely considered to be within the world top 100, so I thought it would be worthwhile to nominate them here.

I really don't think there could be 100 courses worthy of the top 100 that I've never gone to look at.

I don't think this is a good argument for anything.  Why shouldn't we let the rating process determine how many of the top 100 you have seen?  I would rather take a little extra time by considering an extra 50 courses than thinking that 1 worthy course might have been ignored.



A lot of concerns about bias have come up.  For instance, some raters may just rate all the courses they have played really highly.  Other raters might rate strategically, giving one course a 9 or 10 that they really think is a 7 or 8 because they think it is underrated by others.  The opposite is also a serious concern.  All of these concerns and others could be eliminated with a more clever system which has never been used by any of the ranking lists out there.  It is adapted from the Condorcet method of voting, and would work as follows:

Everyone would rank order all the courses they have played (with the possibility of ties).  Then, each course would only be evaluated based on its performance in head-to-head matches.  For example, if Pine Valley is to be ranked #1, it must be true that more people rank Pine Valley over course X than rate course X over Pine Valley.  Thus, Pine Valley would have to beat every other course in a head-to-head match.  Then, after the #1 course is removed, the #2 course would have to beat all remaining courses in a head-to-head match, etc.

One issue with this method is that we would need a lot of raters who have played a lot of different courses.  There would have to be a good number of people (say 10) who have each possible pair of courses.  For instance at least 10 people would have to have played both Chicago Golf Club and Hirono to effectively evaluate them.  Given the number of people that have played a ton of these courses, this may not be a serious problem.

The other problem that may arise is that there could be a "Condorcet Cycle" where X beats Y, Y beats Z, and Z beats X.  Under this case, the Condorcet method does not provide a clear rank order of these 3 courses.  I propose that in such a case, we should simply break the tie by counting only first place votes.  So for all raters who played all three courses, how many raters put X ahead of both Y and Z?

Note that under this system, strategic rating is no longer an issue.  Imagine that I think that ANGC is not the 3rd or 4th best course but more like the 15th best.  Under the currently proposed system, I could simply rate ANGC a 1 and knock it down in the rankings.  Under the Condorcet method, I would simply provide my true rankings.  Rating it really poorly doesn't increase the chances that it will lose to Shinnecock, it only increase the chances that it will lose to worse courses, which is not desirable. 

If we could get enough raters who have played enough courses, and a good programmer willing to put in a little time to design this system, we would get some results that would be pretty difficult to argue with.

[Ian_L]

JK, you bring up two good points.  First, I agree this ranking should be renamed (for different reasons than you).  But you'll have to come up with something better than "Linford and Friends." (I do like the "List of Greatest Courses" part though)

Concerning your second interesting point, I'd like to hear from more people about whether to publish names along with results.  A user name will be required to vote, but I was not planning on making these results public unless the member himself requests it.  If you are truly interested in the opinions of established GCAers, you could start this yourself by publicizing your own answers to get others to do so as well.  I don't think abstaining will benefit anyone.


Once again, I'd just like to remind you that you are voting for a course that you think qualifies for TOP 100 IN THE WORLD.  That means a course could be better than 200 courses on the current ballot, but still not be worthy of nomination.  I will not be including nominations from posters that say something like "I don't think this is really a top 100 course, but..."

It's my fault for not making this clear enough from the outset.  If you really think a course should be considered as top 100 in the world, by all means nominate it.  Just remember, this is for a top 100 list, not a top 500 list.

Tom,

"3.  I will second Capilano (even though I wouldn't vote for it myself) because I swear it was on my original list, or should have been." -Probably my mistake, I was a bit hasty cutting down your list to asterisk-only courses.

"4.  Are you going to let architects vote -- assuming we don't vote for our own courses?"
-Yes, I think the experience architects on this sight outweighs any bias they might have.

"5.  On a quick read through the list there are 99 courses I haven't seen.  (I guess that leaves 290 I have seen.)  So, please don't add to the ballot, because I really don't think there could be 100 courses worthy of the top 100 that I've never gone to look at."
-True, but there may be courses you have seen (especially in the US and UK) that are not yet on this ballot.  I'd like to make sure these courses get consideration.  Hopefully we can keep the number of nominated courses to a minimum.

Anthony, I suggest you look at the previous thread if you are interested in our discussion on keeping voters honest.  Your idea is an interesting one and perhaps could be a future project (probably with fewer courses...).  However, I would argue that your method is not immune, as a rater who would give Augusta National a 1 would probably purposely rank it below inferior courses to drive down its ranking.


I will be updating the nomination this evening.

[Sean_A]

Once again, I'd just like to remind you that you are voting for a course that you think qualifies for TOP 100 IN THE WORLD.  That means a course could be better than 200 courses on the current ballot, but still not be worthy of nomination.  I will not be including nominations from posters that say something like "I don't think this is really a top 100 course, but..."

Ian

Like I told Tom, I don't know if my nominations deserve to be top 100 as I haven't seen all the courses which are candidates.  I can say with absolute sincerity that I believe my nominations are better than many top 100 picks and many more other candidates, but I am only one person.  IMO they may have a shot, but that is down to others to back up or not.  Like you said before, posting the number of voters who rated the courses is a good idea so we can sort for ourselves if some courses are looney tunes choices or if they are really where they belong.  Personally, I think the chips will fall pretty accurately and there will be very few surprises.

Ciao

[Ian_L]

Sean, understood.  If you think they should be considered as top 100 courses, then by all means I will include them.

So what happens if I think some courses have no business being on the list and I would only rate them as a 5 or 6?  Should I give them those scores or will someone think I'm being too harsh?  I can't be the only one with this issue.

Henry, simply give your honest assessment (although there will likely be a minimum of 5 to keep people from intentionally throwing courses they don't like under the bus).  I'm sure there will be courses here that average around 6.  Now, if you give Cypress Point a 5, then your vote may be tossed through statistical screening Jonathan has suggested.

[John Kirk]

I'm sorry, just from the standpoint I thought we were also preparing a top 100 U.S. list.  Otherwsie, I probably wouldn't have nominated or seconded the courses I named.

[TX Golf]

I'll third California Golf Club and  The Alotian Club.

[Brian_Ewen]

The best way to do this type of thing is for people to vote regionally. Most people here are familiar with the courses in their state or county and would be knowledgeable enough to be able to grade the best 20.

Or even COUNTRY Adrian !!!

I thought it was a WORLD Top 100 , not USA ?

[Jonathan Cummings]

Ian - all I can say is that I tip my hat to you.  You are embarking on a thankless pursuit which if taken to its limit could yield the best list out there.  I think (and have thought for years) that the potential exists within the gca.com family to compile a better top 100 list than published by any of the magazines.  The knowledge, compassion and committment of the majority of the gca.com "panelists" is clearly deeper than any of the magazine panels.

Ignore the naysayers, I think this is a worthwhile and interesting effort.

JC   

[Pete_Pittock]

Nominations:
Blackhawk (Edmonton, Alta)
Kooyonga

Q: Which iteration of Nefyn & District

[Matt_Ward]

I think it's a great idea that "sacred cow" courses can never receive a five (5) or less rating.

Nothing like making sure they are protected from honest feelings.

Ah, but then again -- we have to protect against those who might throw certain courses "under the bus" -- geeze, what about those sacred cows that have been milking their preferred guaranteed place in line for years?

This is a wonderful exercise but when straight jackets are applied, no matter how well intentioned, you have issues.

Ditto the preposterous idea that all raters are equal.

They are not -- and never will be. This isn't like political voting which follows the dictum of "one man -- one vote."

No doubt it would be difficult to weigh certain raters and what they offer but for people to ignorantly believe that all people should be treated the same is not true. Consensus lists fail because they too often protect the sacred cows and minimize the unique elements of a great many new or failr new layouts that have not been permitted to shine because of their relative newness.

Nothing against Ian as I've said a few times -- best of luck no matter how you set things up.

[Anthony Fowler]

Anthony, I suggest you look at the previous thread if you are interested in our discussion on keeping voters honest.  Your idea is an interesting one and perhaps could be a future project (probably with fewer courses...).  However, I would argue that your method is not immune, as a rater who would give Augusta National a 1 would probably purposely rank it below inferior courses to drive down its ranking.

Hi Ian, thanks for all the effort.  This is a little bit tricky to understand and convince yourself of, but bear with me for a moment.  Under this Condorcet method, raters have no incentive to misrepresent their actual views as they do under your Doak scale method.  Let me try to demonstrate with an example. 

Say that we are only rating 4 courses: Shinny, National, Sebonack, and Southampton.  Imagine that I am a rater and I prefer the courses in the following order: Shinny > Sebonack > National > Southampton, but I think that most of the other raters will vote Shinny > National > Sebonack > Southampton, so essentially I think National is overrated compared to Sebonack.

Under the Doak scale method, I could misrepresent my views by giving Sebonack a 10 and National a 1, when I really think they are a 9 and 8 respectively.  This would dramatically shift the results toward my preferred ratings.

However, under the Condocet method, the best thing I could do would be to provide my true order.  When the results are calculated, only head-to-head matches are considered.  For instance if I vote Shinny > Sebonack > National > Southampton, it would be recorded as 6 different votes in 6 different head-to-head matchups.  In the contest between Shinny and Sebonack, I vote for Shinny.  In the contest between Sebonack and National, I vote for Sebonack; etc.  Since I think National is overrated, I might think that I should misrepresent my views and rate Southampton ahead of National to compensate, but this doesn't help.  Doing this does not affect National's chances against Shinny or Sebonack in any way.  It only would increase the chances that Southampton would be rated ahead of National, which I would not prefer. 

I hope that this is clear, but I would be happy to discuss it at more length.  It is possible that people would still rank courses irrationally and misrepresent their true view, but this problem could be solved by educating the raters on this aspect of the voting system.

[Ian_L]

John, no worries, my fault for not making this clear enough.

Jonathan, thanks for the kind words.  I've actually been quite happy so far that nearly all the criticism has been constructive and not destructive.  I'm glad this activity has sparked debate.

Pete, I don't know.  I got the course from this page: http://www.todaysgolfer.co.uk/Golf/courses/Top-100-Golf-Courses/Top-UK-Golf-at-Druids-Glen-Nefyn-and-more/ .  I hope this helps, and if you know a way to make this less ambiguous I'd greatly appreciate it.

Matt, I think we've been through these before but let's give it another shot:

No list will be perfect. However, the truth is (and I witnessed it firsthand in the previous activity) that the number of people maliciously giving 2's and 3's out will be much greater than those who honestly believe this.  I know this because the patterns of the voters were (1) repeated almost precisely which means the same person was voting multiple times, and (2) these votes were often going against a specific architect (Tom Doak was the one I witnessed).  I find it difficult to believe someone would give a place like Friar's Head a 9 but still think all Doak courses are 2's (not to mention travel the world to Tasmania to play all Doak courses).  Thus, I've concluded that the harm from allowing low scores outweighs the good that comes from one person honestly thinking Pebble Beach is a 4, or Muirfield Village is a 2 (in fact, we have no evidence that such a person exists as no one has spoken out as of yet).  To be honest, I would say if anything not allowing less than a 5 is unfair to the better courses, as some of the OK courses on this list might honestly be deserving a 4 from some raters.  Another reason not to post the entire list.

As for giving more weight to voters who have played more courses, I think (1) it provides an incentive for voters to add more courses than they have honestly played, (2) it would be fairly complicated to implement, and (3) I think it conflicts with the concept of this poll (I can't think of a better way to explain this point).

I think the solution to both your objections is to ask a very knowledgeable GCAer for his Top 100 rather than count on this poll.  We both agree on that.  This poll will not unequivocally rank the top 100 courses in the world (even if such a thing could be done), and as you have alluded to in previous posts, a personal list from a knowledgeable source might provide a better list.  However, that is not what this poll is about.  BTW, thanks for your contributions, they've definitely made me stop and think.

Anthony, I see what you mean there (nice explanation).  However, wouldn't you end up with a mess of non-linear rankings, as you mentioned earlier? (A > B > C > A)  Perhaps not.  While I don't see this specific project going down that route.  I would definitely be interested in such an activity.

P.S. I didn't get a chance to proofread this, so my apologies if some points aren't clear.

[Peter Pratt]

Ian,

I'm glad you're doing this, and I appreciate all the refinements you've made to improve the list. I love this kind of thing and think it will be very interesting. Thanks!

[Tags:]

Tags:

[Tags:]

Tags:

Call to undefined function theme_linktree()