During Worlds, there was some discussion on this board about how things would be different if the top pros weren't playing some of their rounds on shorter courses.

This topic was touched on again in the Worlds review issue of DGWN.

For 3 of the 9 rounds (8 plus semi-finals) at Worlds, the MPO division played on courses with an SSA of around 49. This was 2 rounds at ISU, and 1 round at Grandview. For the other 6 rounds, the SSA was in the range of 55 to 57.

I have removed these 3 rounds from the final results to compare the scores across the other 6 rounds.

The results are shown below.

<table border="1"><tr><td>Place</td><td>Name</td><td>Strokes</td><td></td><td>||</td><td></td><td></td><td>Place</td><td>Name</td><td>Without IS & GV
</td></tr><tr><td>1</td><td>Barry Schultz</td><td></td><td>||</td><td></td><td></td><td>1</td><td>Brian Schweberger</td><td>---
</td></tr><tr><td>2</td><td>Ken Climo</td><td>+11</td><td></td><td>||</td><td></td><td></td><td>1</td><td>Steve Rico</td><td>---
</td></tr><tr><td>3</td><td>Brian Schweberger</td><td>+16</td><td></td><td>||</td><td></td><td></td><td>3</td><td>Barry Schultz</td><td>+1
</td></tr><tr><td>3</td><td>Steve Rico</td><td>+16</td><td></td><td>||</td><td></td><td></td><td>4</td><td>Ken Climo</td><td>+4
</td></tr><tr><td>5</td><td>Cameron Todd</td><td>+18</td><td></td><td>||</td><td></td><td></td><td>4</td><td>Keith Warren</td><td>+4
</td></tr><tr><td>5</td><td>David Feldberg</td><td>+18</td><td></td><td>||</td><td></td><td></td><td>6</td><td>David Feldberg</td><td>+5
</td></tr><tr><td>5</td><td>Keith Warren</td><td>+18</td><td></td><td>||</td><td></td><td></td><td>7</td><td>Cameron Todd</td><td>+6
</td></tr><tr><td></tr></td></table>

Rodney,

I think you might have lost some information there in the formatting, but it looks like Schweberger and Rico would have tied, with Schultz one stroke back, Climo and Warren another three strokes back, and Feldberg and Todd also nipping right on their heels.

With such tight races, one can only imagine how the outcomes might have changed if you take out Barry's 41 and 42 on the lower SSA ISU course and the 43 recorded by Schultz/Climo on the Grandview course. How differently would everyone have played if Barry hadn't jumped out to an early 5 stroke lead? As impressive as that 1088 round of 41 was, should our world championships be decided by deuce or die?

If you are going to be a world champ you should be able to shoot well on ANY course which is why they try and have a mix of courses to play on. IMO a world champ should be able to shoot well on ANY course that he has to play.

I mean if i got to take out the technical course of Walnut Ridge that we had to play twice in the ADV division. I could shave 21 strokes off my score and I could have jumped up about 10 places. The thing is we DID have to play that course and i didnt fare well on it because that part of my game isnt very good yet. There for i am not the well rounded player that i want to be. I did on the other had shoot very well at DMACC and Ewing because those are more my style of courses at the moment. So if we played those all week i would have fair alot better.

The point is all the World Champs and everyone in the top 10 of all the big divisions played well on ALL the course which is the true test of a good player. Plays well on deuce or die. Plays well on Long and Open. Plays well on long and technical. Plays well EVERYWHERE.

It's seems like the title should be decided by playing a selection of courses with a variance in length, foliage and elevation if possible. Some courses have all of these and others with some elements need to be balanced with courses having the other elements. One thing to consider is that you're comparing one set of scores with another set with three fewer rounds. Barry might still have been hot that day with a 1088 round on Walnut or Ewing in place of ISU?

I still think categorizing ISU or Grandview (or any course for that matter) as SSA 49 imparts no meaning. Like y'all and Houck always preach, if the holes are fair and create a good spread, they should be acceptable holes for a major tournament. In other words, the holes themselves are important, not the SSA.

There's scoring spread on each hole which is good (2s, 3s, 4s, 5s) and scoring spread among holes which is also good (par 3s, 4s and 5s). If most holes at a major are Gold par 3s resulting in mostly SSA 46-49 courses, we're missing the 'among holes' spread.

A) Rookie formatting mistake on my part. Sorry. Jeff interprets correctly.

B) James, I agree with you, and I would have much rather listed the p*r with the SSA to give it more context (along with same for the other courses), but I didn't want to let loose the P*r Crazies. Maybe I'll sneak that in later. Also, though I didn't mention it, I think ISU was a fine course for MPO, especially to use for only 2 of the 9 rounds. It did well at spreading scores, with only 4 questionable holes, and only 2 of those were very bad. Point is, those two courses created lower scores, regardless of SSA, terrain, foliage, or hole design, and some people wondered how those lower-scoring courses impacted the competition.

short courses are a part of the game - its not all about how far you can throw or how many threes you can make sometime.

Those courses were an awesome blend of golf and the best overall golfer won.

I think Schweb would even tell you exactually what I said

I'd comment on the issue at hand if I understood the criticism behind using shorter courses. Is it that people feel that the holes are short and reachable and conclude that it allows a person with less distance or skills to compete? or that shorter courses are somehow easier?

In my experience, there is little difference between a well designed low SSA or a well designed high SSA course. I am a short arm and I feel that Pickard/Ewing/WR were as fair as ISU. If there was a difference, it is only that ISU creates a little more pressure if I didn't start out hot. I might press more than I would if I started slow at Pickard. However, this is purely a psychological weakness that I should be punished for having.

Anyways, dealing with the pressure or pace of the short courses as well as the physical demands of the longer courses should always be present at worlds. They are all qualities that a true champ should have.

... this is purely a psychological weakness that I should be punished for having.




In a just world, psychological weakness would be merely the first of many things for which you would be punished.

Below are some stats on the courses, based on one MPO round each.

Note that ISU and Grandview had 40% and 36% 2's respectively, while none of the other courses had more than 25% 2's. This might lead some to call these "deuce or die" courses.

Ideally, I think the shots provided by a short course could be provided on long courses by way of the approach shot into 2-shot and 3-shot holes. Like Chuck was saying, it's probably a good thing to have courses that produce a good mix of 2's, 3's, and 4's overall. But 2-shot and 3-shot holes are still far from the norm in disc golf.

So anyway, if you have the luxury of whatever courses you want, the question might be whether you *want* to mix in a couple rounds where there are 14-16 reachable holes on the course. That type of course is certainly representative of the vast majority of courses in the world today.

As it was, Iowa didn't have much choice. ISU, Carroll Marty, and Grandview each have a high number of reachable holes. Grandview was to be played by everyone. That left one course for the Pros to play twice, and one for the Ams to play twice.

And finally, Chuck made another good point. Namely, you can't just take away three rounds and say "look who would have been ahead without those short courses". You would really need to replace those rounds with 3 others, and who knows what might have happened in those 3 replacement rounds. What you *can* say is "here's how everyone did on the 5 rounds on the longer courses".

<table border="1"><tr><td> Course</td><td>1-shot</td><td>Tween</td><td>2-shot</td><td>Tween</td><td>3-shot</td><td>Par</td><td>Round</td><td>SSA</td><td>%2\'s</td><td>%3\'s</td><td>%4\'s</td><td>%5\'s</td><td>%6\'s</td><td>%7\'s
</td></tr><tr><td>ISU</td><td>16</td><td>0</td><td>2</td><td>0</td><td>0</td><td>56</td><td>A Pool Round 5</td><td>48.7</td><td>40%</td><td>47%</td><td>10%</td><td>2%</td><td>0%</td><td>0%
</td></tr><tr><td>Grandview</td><td>14</td><td>2</td><td>2</td><td>0</td><td>0</td><td>58</td><td>A Pool Round 7</td><td>48.7</td><td>36%</td><td>55%</td><td>8%</td><td>0%</td><td>0%</td><td>0%
</td></tr><tr><td>Ewing</td><td>10</td><td>2</td><td>6</td><td>0</td><td>0</td><td>62</td><td>A Pool Round 8</td><td>54.9</td><td>22%</td><td>54%</td><td>18%</td><td>4%</td><td>1%</td><td>0%
</td></tr><tr><td>Pickard</td><td>10</td><td>0</td><td>6</td><td>1</td><td>1</td><td>64</td><td>A Pool Semis</td><td>55.4</td><td>25%</td><td>47%</td><td>22%</td><td>5%</td><td>1%</td><td>0%
</td></tr><tr><td>Big Creek</td><td>11</td><td>0</td><td>6</td><td>0</td><td>1</td><td>62</td><td>A Pool Round 4</td><td>56.7</td><td>21%</td><td>51%</td><td>18%</td><td>7%</td><td>2%</td><td>1%
</td></tr><tr><td>Walnut Ridge</td><td>10</td><td>0</td><td>7</td><td>0</td><td>1</td><td>63</td><td>A Pool Round 3</td><td>57.3</td><td>17%</td><td>54%</td><td>23%</td><td>5%</td><td>1%</td><td>0%
</td></tr><tr><td> </tr></td></table>

Notes:
- ISU - Second time MPO played here. Great weather.
- Grandview - Great weather.
- Ewing - Great weather.
- Pickard - After cut to Semis. 3rd round here. Great weather.
- Big Creek - Windy.
- Walnut Ridge - Pre-shuffle. A bit windy?

And now for additional fun and discussion, I present the deuce percentage for the holes that were clearly reachable in one shot.

At ISU, where Barry had a 41 and a 42, only 5 of the 16 reachable holes had over half the field getting a birdie on them.

At Grandview, that number is 6 of 14 reachable holes. But on the other end, there were 4 holes with deuce percentages below 30%.

<table border="1"><tr><td> ISU Hole</td><td>Deuce %</td><td></td><td>Grandview Hole</td><td>Deuce %
</td></tr><tr><td>6</td><td>8%</td><td></td><td>18</td><td>14%
</td></tr><tr><td>3</td><td>33%</td><td></td><td>13</td><td>26%
</td></tr><tr><td>12</td><td>33%</td><td></td><td>5</td><td>28%
</td></tr><tr><td>13</td><td>35%</td><td></td><td>9</td><td>28%
</td></tr><tr><td>17</td><td>35%</td><td></td><td>3</td><td>35%
</td></tr><tr><td>11</td><td>38%</td><td></td><td>17</td><td>35%
</td></tr><tr><td>4</td><td>40%</td><td></td><td>8</td><td>43%
</td></tr><tr><td>10</td><td>44%</td><td></td><td>10</td><td>49%
</td></tr><tr><td>7</td><td>46%</td><td></td><td>16</td><td>51%
</td></tr><tr><td>5</td><td>47%</td><td></td><td>2</td><td>60%
</td></tr><tr><td>14</td><td>47%</td><td></td><td>11</td><td>61%
</td></tr><tr><td>8</td><td>56%</td><td></td><td>15</td><td>63%
</td></tr><tr><td>2</td><td>61%</td><td></td><td>7</td><td>67%
</td></tr><tr><td>1</td><td>63%</td><td></td><td>14</td><td>71%
</td></tr><tr><td>15</td><td>68%
</td></tr><tr><td>9</td><td>74%
</td></tr><tr><td> </tr></td></table>

Rodney,

I appreciate you pulling all this data together, but I'm not sure what point you are trying to make about the ISU course. Again, I am convinced that Barry's 1088 round was equally as impressive as if he had shot a round with that rating at a high SSA course.

Maybe it's just me, and I guess I really dodn't know how to say it, but I just feel that world championships should be contended on world championship caliber layouts. It just seemed strange to me that during the very first round of the Pro Worlds tournament, more than half the Pro field shot in the mid to high 40s.

Think about the USDGC. Why is that event so enormously popular? Why do people pay a very high entry fee to compete in that tournament? I'm willing to bet that it's the world championship caliber and high SSA of that course. People love the challenge. It features many multi-shot holes and more opportunities to display golf knowledge. It's got the 'among holes' that Chuck talked about. It's also got the drive/putt holes, as well as the other types of holes. Would that event be nearly as popular if they played the regular Lakefront Withrop layout with a much lower SSA? I think not. Or how about if the USDGC tournament also included several rounds at Boyd Hill? I don't think so. The USDGC and higher SSA courses test the overall skills of the golfer much more so than drive/putt courses, in my opinion, and just seem to me like the best courses to be used in world championships.

Hey Jeff, I'm not trying to make a point. I'm just providing data. I'll let anyone else work out any points to be made.

I think it's probably healthy for future tournaments (future Worlds) to discuss such topics and learn as much as can be learned from the past so future endeavors have clear objectives.

You've made your thoughts pretty clear. I think others might disagree with some or all of them.

Given the statements in your last post, the data I'm providing might help you (or others) sort out whether a course like ISU is of "world championship caliber", as you put it. In other words, by your observation, over half the field was scoring in the mid- to upper-40's, and this seemed "strange" to you. By providing data, I'm perhaps helping you (or others) to understand why this might be, and also to help you shape opinions on whether this is a good thing or a bad thing.

I know that you know that Iowa didn't have much choice on courses, as they pretty much had to use them all. But in the future, some hosts might have choices to make, and I'm just thinking those hosts might be ill-advised to make those choices based on "hey, that course has a bunch of scores in the upper 40's, but this course has a bunch of scores in the upper 50's, so this 2nd course is obviously much better for Worlds than the first one".

Besides, what good is typing in all the scorecards if I don't get to play with the numbers?

The initial post on scores without the short courses was just for curiosity. The subsequent data was just stuff that sprang to mind based on some of the other posts. Guys like me have fun looking at the numbers when we know the specific courses and holes. I realize you haven't seen any of those courses, and maybe you're not a geek, so perhaps the numbers don't have any fun value for you.

So geekmeister Rodney, how about two more lists. One would rank the courses in descending order based on the standard deviation of the scoring averages for the 18 holes on a course. The other would be the average of the standard deviation of the scoring distributions on each hole. The course ranking higher on both would provide the most shot and scoring variety which, in theory, is better for those who feel these are important elements for championship courses (which includes me).

In a just world, psychological weakness would be merely the first of many things for which you would be punished.


You should make that your tag-line/sig. :)

higher SSA courses test the overall skills of the golfer much more so than drive/putt courses



I think I know what you are saying, but this statement is such BS and people blindly following it has screwed up many a good hole and course. I can increase SSA merely by lengthening well designed holes to just the right length to make them poorly designed. I have increased the SSA and made the course a poorer test of overall skills.

Chuck, I'll bet you that ISU's overall spread (even if measured by standard deviation) fits nicely within the data from the other courses. Like holes, my idea of a good course at worlds is that it is fair and that I can make a nice move if I play well. I always felt that ISU fit that bill.

While a higher SSA course (in the normal range from 45-65) isn't necessarily better than a lower one for a particular skill level, if two courses are well designed for the same skill level, the higher SSA is probably better. However, the two lists I asked Rodney to prepare don't directly relate to SSA. Scoring spread on a hole and among holes is somewhat independent of SSA. However, it's pretty difficult to have scoring spread among holes on a low SSA course where you mostly have par 3s (and 2s) like ISU.

higher SSA courses test the overall skills of the golfer much more so than drive/putt courses



I think I know what you are saying, but this statement is such BS and people blindly following it has screwed up many a good hole and course. I can increase SSA merely by lengthening well designed holes to just the right length to make them poorly designed. I have increased the SSA and made the course a poorer test of overall skills.




I agree with James. Oh wait, I just report data. Nevermind.

I'm not sure about the measures (they may end up measuring the same thing), but having a large spread for holes may not be the same as having a large spread for the round totals. My belief is simply this: If I was 4 strokes from the money with one round to go, I feel that I had as good a chance at ISU as I did at the other courses, not because it was a deuce or die course, but because it was a well designed deuce or die course where many would be "dieing" rather than "deucing."

One would rank the courses in descending order based on the standard deviation of the scoring averages for the 18 holes on a course. The other would be the average of the standard deviation of the scoring distributions on each hole.



<table border="1"><tr><td>Course</td><td>Stdev of Hole Averages
</td></tr><tr><td>BC</td><td>0.707
</td></tr><tr><td>PK</td><td>0.685
</td></tr><tr><td>EW</td><td>0.588
</td></tr><tr><td>WR</td><td>0.544
</td></tr><tr><td>IS</td><td>0.492
</td></tr><tr><td>GV</td><td>0.335
</td></tr><tr><td> </tr></td></table>


<table border="1"><tr><td> Course</td><td>Ave of StDev of scores
</td></tr><tr><td>BC</td><td>0.705
</td></tr><tr><td>WR</td><td>0.651
</td></tr><tr><td>PK</td><td>0.581
</td></tr><tr><td>EW</td><td>0.569
</td></tr><tr><td>IS</td><td>0.548
</td></tr><tr><td>GV</td><td>0.546
</td></tr><tr><td> </tr></td></table>

HOWEVER, there was a rogue '7' on hole 11 at Grandview. I'm guessing this is due to late arrival. This '7' takes the StDev for that hole from 0.520 to 0.749. That, in turn, raises Grandview's average to 0.546 from 0.533. There may be other rogue scores as well. If they all move the StDev by that much, I'm not sure the 2nd list is all that instructive.

For those that played the holes, here are the highest StDev for each course, along with a reminder of the hole:
<table border="1"><tr><td> Hole</td><td>GV</td><td>Comment
</td></tr><tr><td>11</td><td>0.749</td><td>0.520 without a rogue 7
</td></tr><tr><td>17</td><td>0.639</td><td>Uphill anhyzer
</td></tr><tr><td>6</td><td>0.612</td><td>Backstop hole.
</td></tr><tr><td>
</td></tr><tr><td>Hole</td><td>IS</td><td>Comment
</td></tr><tr><td>18</td><td>0.751</td><td>Long 2-shot
</td></tr><tr><td>14</td><td>0.678</td><td>Longish 1-shot
</td></tr><tr><td>13</td><td>0.655</td><td>Tricky turnover
</td></tr><tr><td>
</td></tr><tr><td>Hole</td><td>BC</td><td>Comment
</td></tr><tr><td>11</td><td>0.907</td><td>The Freak, 3-shot and nasty
</td></tr><tr><td>7</td><td>0.888</td><td>Short 2-shot but tight
</td></tr><tr><td>12</td><td>0.869</td><td>Dogleg left 2-shot
</td></tr><tr><td>
</td></tr><tr><td>Hole</td><td>WR</td><td>Comment
</td></tr><tr><td>15</td><td>0.926</td><td>The monster tunnel
</td></tr><tr><td>7</td><td>0.906</td><td>Downhill beauty, 5 6\'s?
</td></tr><tr><td>9</td><td>0.785</td><td>Long 2-shot dogleg right
</td></tr><tr><td>
</td></tr><tr><td>Hole</td><td>PK</td><td>Comment
</td></tr><tr><td>18</td><td>0.730</td><td>Longish 3-shot
</td></tr><tr><td>14</td><td>0.699</td><td>2-shot gapper 2nd
</td></tr><tr><td>12</td><td>0.691</td><td>2-shot after ski slope 11th
</td></tr><tr><td>
</td></tr><tr><td>Hole</td><td>EW</td><td>Comment
</td></tr><tr><td>12</td><td>1.080</td><td>Long 2-shot, OB left
</td></tr><tr><td>18</td><td>0.712</td><td>Longish 2-shot
</td></tr><tr><td>6</td><td>0.650</td><td>Downhill dogleg right 2-shot
</td></tr><tr><td> </tr></td></table>

I owe myself a beer.

Thanks, Rodney. What do you think of these two calculations as ways to analyze courses? These calculations still don't indicate whether the courses were designed for Gold level players nor does it indicate how much terrain variety there is on the courses. But from a statistical standpoint, I believe it shows it would have been better if pros played BC, WR or EW twice instead of ISU. PK was good as expected (by me) in both categories.

I remember all those holes....those holes were very very tough. I know I got behind that stupid backstop at Grandview.....

Rodney, is the average standard deviation of scores based on the round total scores? Please explain, I'm not really a stats guy but .5 to .7 seems kind of low for a measure of spread for final scores.

The calculation first determines the standard deviation of the scoring spread on each hole, not total round scores. That's why they should all be less than 1 on each hole. Then, you average those 18 numbers for each course to get the number shown in the ranking list.

Yes, in that case, I would like to see a standard deviation for final scores. I suppose you would expect the same ranking of courses in a somewhat proportional way. I am curious to see if that relative ranking and proportion holds up.

I am also curious why you think Ewing should have been used twice and Pickard was good (but ISU presumably was not) when they all had virtually the same average spread. In fact, when you factor in the total SSA of Pickard and Ewing (read longer holes) and realize that their average spread is virtually the same as ISUs, its hard to make an argument (statistical at least) for their superiority over ISU.

I said those courses: WR, EW, PK and BC were better statistically than ISU and GV strictly on the basis of the two lists Rodney prepared. I believe those are important tools but certainly other factors come into play. As Houck has pointed out, high standard deviation for scores on a hole doesn't mean it was done well if the deviation is primarily due to luck (OB 10 feet away from pin or pinball corridors) not skill elements. I also think that GV would have been a better lower SSA course than ISU to play twice. ISU could eventually match GV but it was still a little raw and not worked in as you'd expect for a course just completed less than a month earlier.

[QUOTE]


Ideally, I think the shots provided by a short course could be provided on long courses by way of the approach shot into 2-shot and 3-shot holes. [QUOTE]


Amen.

[QUOTE]
Like Chuck was saying, it's probably a good thing to have courses that produce a good mix of 2's, 3's, and 4's overall. But 2-shot and 3-shot holes are still far from the norm in disc golf. [QUOTE]


True dat.

[QUOTE]
So anyway, if you have the luxury of whatever courses you want, the question might be whether you *want* to mix in a couple rounds where there are 14-16 reachable holes on the course. That type of course is certainly representative of the vast majority of courses in the world today. [QUOTE]


I've always thought that the traditional Worlds setup of four courses should include one "traditional" birdie course. That kind of course is still the norm, and it's where we came from.

And while longer courses can provide similar shot challenges, birdie courses -- as I think someone has already pointed out (maybe even James) -- present a different psychological challenge.

And, Jeff, the USDGC course is awesome. But when you're looking to explain the popularity of the event, you need to account for the very professional way in which it's run, the uniqueness of the format... and, oh yeah... the massive amounts of cash offered.

I guarantee you that there are touring players there who don't care for all the yellow rope, but they go to make money. Sad, but true.

And, Rodney, thanks again for doing all that work. It's great to finally have all that data available. Should I send you the cards from '94 and '95 Worlds? Maybe for Christmas?

Well, it will be interesting at Allentown next year where I doubt any course will be under SSA 52 and one will be in the mid-60s. All four of MN PW2001 courses were about SSA 54. So no birdie course but several holes marked and played as Par 2s on each one.

Is this the
best way
to post?
:p :p

Well, it will be interesting at Allentown next year where I doubt any course will be under SSA 52 and one will be in the mid-60s. All four of MN PW2001 courses were about SSA 54. So no birdie course but several holes marked and played as Par 2s on each one.



You can count on lots of par 2 holes at the four courses that will be used for Pro Worlds 2005, while the courses will still have an SSA near 54 or higher. That's part of what I'm trying to say - is having an entire deuce or die course at a Pro Worlds necessary?

James, stick to short jokes.

You're going to hurt yourself trying to be funny.

See you and your cigar Saturday.

The calculation first determines the standard deviation of the scoring spread on each hole, not total round scores. That's why they should all be less than 1 on each hole.



I don't understand the second sentence. And I'm not clear on the first.

For each hole, I calculated the stdev of all the scores recorded on that hole. I'm not sure if that's the same as what is said in the first sentence above. And I don't know why they "should" all be less than 1 on each hole.

Anyway, then I averaged those 18 stdev's.

Thanks, Rodney. What do you think of these two calculations as ways to analyze courses?



Not sure. I think it's additional information, and it seems like it would be meaningful. Though I'm leary of the fact that a single bad score by one player can pretty drastically impact that second number (the average of the hole stdevs).

Not having any formal statistics training, I don't know if this is really a good use of stdev. For a hole like #11 at Grandview, we have 44 2's, 26 3's, and 1 '5'. That's not anywhere near a classic normal distribution. Then you add 1 '7' and the stdev goes all crazy. Again, I'm not familiar with the theories behind using stdev on datasets that are as far from normal as this one.

Anyway, I haven't looked much at using stdev. I'm not sure it tells you much more than looking at the scoring spreads (% of each score for a hole), though it does allow you to collapse it to a single number and do further aggregation from there, which is nice.

I also think that GV would have been a better lower SSA course than ISU to play twice.



After looking at the numbers, I disagree with that pretty strongly.

Grandview had 6 holes on which over 70% of the A Pool got the same score.
ISU had 2 holes on which over 70% of the A Pool got the same score.

Both courses had two legitimate 2-shot holes. The % of 3's/4's/5's on those holes were:
- GV#6 : 65/31/1
- GV#12 : 67/31/1
- IS#16 : 24/69/7
- IS#18 : 18/50/29
In other words, the 2-shot holes were much more demanding at ISU.

ISU scored "better" on the two stdev measures you suggested earlier. (Though without the 2-shot holes, GV appears to show more variance of hole scoring averages.)

Overall spread of individual holes scores (% of 2's/3's/4's/5's) was like this:
- GV = 36/55/8/0 (only 4 5's out of 1296 holes played)
- IS = 40/47/10/2

GV had two "tweener" holes (between 1-shot and 2-shot) that gave up 4 deuces and 5 deuces, respectively, in the field of 72 A Pool players (both holes had 74% 3's).
By contrast, ISU's only "tweener" hole was reachable but hard to get to, and thus gave up only 6 deuces and had 86% 3's.

GV had 6 holes where 18 of the top 20 men got the same score on the hole (including both 2-shot holes).
ISU had 0 holes where 18 of the top 20 men got the same score on the hole (and only 2 holes where 16 or 17 of the top 20 men got the same score on the hole).

GV requires very few recovery shots.

GV is arguably less demanding for accuracy.

GV arguably requires fewer shot types off the tee.

GV has lots of artificial OB (though it doesn't come into play much) and other unnatural distractions or obstacles (backstop, restrooms).

Yes, in that case, I would like to see a standard deviation for final scores.



Here are the stdevs of the scores (for the one round per course that I currently have data on).

<table border="1"><tr><td> Course</td><td>StDev
</td></tr><tr><td>BC</td><td>4.5
</td></tr><tr><td>WR</td><td>3.4
</td></tr><tr><td>PK</td><td>3.3
</td></tr><tr><td>GV</td><td>3.1
</td></tr><tr><td>EW</td><td>3.0
</td></tr><tr><td>IS</td><td>2.8
</td></tr><tr><td> </tr></td></table>

For fun, here is another look at score distributions for GV vs. IS:

IS - average score = 49.3
<table border="1"><tr><td> 42</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>45</td><td>45</td><td>45</td><td>45</td><td>45</td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>46</td><td>46</td><td>46</td><td>46</td><td>46</td><td>46</td><td>46</td><td>46</td><td></td><td></td><td></td><td>
</td></tr><tr><td>47</td><td>47</td><td>47</td><td>47</td><td>47</td><td>47</td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td></td><td></td><td></td><td>
</td></tr><tr><td>49</td><td>49</td><td>49</td><td>49</td><td>49</td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50
</td></tr><tr><td>51</td><td>51</td><td>51</td><td>51</td><td>51</td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>52</td><td>52</td><td>52</td><td>52</td><td>52</td><td>52</td><td>52</td><td>52</td><td>52</td><td>52</td><td></td><td></td><td>
</td></tr><tr><td>53</td><td>53</td><td>53</td><td>53</td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>54</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>55</td><td>55</td><td></td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>56</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td> </tr></td></table>


GV - average score = 49.2
<table border="1"><tr><td> 43</td><td>43</td><td>43</td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>45</td><td>45</td><td>45</td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>46</td><td>46</td><td>46</td><td>46</td><td>46</td><td>46</td><td>46</td><td>46</td><td></td><td></td><td>
</td></tr><tr><td>47</td><td>47</td><td>47</td><td>47</td><td>47</td><td>47</td><td>47</td><td>47</td><td>47</td><td></td><td>
</td></tr><tr><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48</td><td>48
</td></tr><tr><td>49</td><td>49</td><td>49</td><td>49</td><td>49</td><td>49</td><td>49</td><td>49</td><td></td><td></td><td>
</td></tr><tr><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td>50</td><td></td><td></td><td>
</td></tr><tr><td>51</td><td>51</td><td>51</td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>52</td><td>52</td><td>52</td><td>52</td><td>52</td><td>52</td><td>52</td><td></td><td></td><td>
</td></tr><tr><td>53</td><td>53</td><td>53</td><td>53</td><td>53</td><td>53</td><td></td><td></td><td></td><td>
</td></tr><tr><td>54</td><td>54</td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>55</td><td>55</td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>56</td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td>58</td><td></td><td></td><td></td><td></td><td></td><td>
</td></tr><tr><td> </tr></td></table>

One more. Something that was posted made me think of the StDev of average scores, grouped by hole type (1-shot vs. 2-shot).

Here are those stats.

1-shot holes
<table border="1"><tr><td> Course</td><td>Number</td><td>StDev
</td></tr><tr><td>EW</td><td>10</td><td>0.208
</td></tr><tr><td>PK</td><td>10</td><td>0.207
</td></tr><tr><td>GV</td><td>14</td><td>0.197
</td></tr><tr><td>BC</td><td>11</td><td>0.187
</td></tr><tr><td>IS</td><td>16</td><td>0.179
</td></tr><tr><td>WR</td><td>10</td><td>0.114
</td></tr><tr><td> </tr></td></table>

2-shot holes
<table border="1"><tr><td> Course</td><td>Number</td><td>StDev
</td></tr><tr><td>EW</td><td>6</td><td>0.406
</td></tr><tr><td>WR</td><td>7</td><td>0.293
</td></tr><tr><td>BC</td><td>6</td><td>0.264
</td></tr><tr><td>IS</td><td>2</td><td>0.236
</td></tr><tr><td>PK</td><td>6</td><td>0.218
</td></tr><tr><td>GV</td><td>2</td><td>0.029
</td></tr><tr><td> </tr></td></table>

The thing that really jumped out at me here was the incredibly low stdev for the 1-shot holes at WR. Here at the scoring averages of the 1-shot holes at Walnut Ridge:
<table border="1"><tr><td> Hole</td><td>Ave
</td></tr><tr><td>1</td><td>2.72
</td></tr><tr><td>6</td><td>2.72
</td></tr><tr><td>4</td><td>2.73
</td></tr><tr><td>8</td><td>2.73
</td></tr><tr><td>3</td><td>2.75
</td></tr><tr><td>11</td><td>2.75
</td></tr><tr><td>18</td><td>2.79
</td></tr><tr><td>2</td><td>2.82
</td></tr><tr><td>5</td><td>2.99
</td></tr><tr><td>14</td><td>3.03
</td></tr><tr><td> </tr></td></table>

As you can see, there are 6 holes with a scoring average between 2.72 and 2.75. Wow.

--------------------------------------------------------------------------------

The calculation first determines the standard deviation of the scoring spread on each hole, not total round scores. That's why they should all be less than 1 on each hole.


--------------------------------------------------------------------------------



I don't understand the second sentence. And I'm not clear on the first.

For each hole, I calculated the stdev of all the scores recorded on that hole. I'm not sure if that's the same as what is said in the first sentence above. And I don't know why they "should" all be less than 1 on each hole.

Anyway, then I averaged those 18 stdev's.



That's what I said. We didn't have any holes like Winthrop Gold 12 or 13 which might be some of the few where the standard deviation could go above 1. I would think holes probably have to have scoring averages over 4 and lots of OB to potentially get a scoring spread resulting in a standard deviation over 1.

I would think holes probably have to have scoring averages over 4 and lots of OB to potentially get a scoring spread resulting in a standard deviation over 1.



Okay, cool. And you are correct. The only hole with a StDev over 1 was #12 at Ewing, which had a scoring average of 4.45 and gratuitous OB left. This hole had 9 3's, 36 4's, 16 5's, 7 6's, and 3 "other".

I also think that GV would have been a better lower SSA course than ISU to play twice.


--------------------------------------------------------------------------------



After looking at the numbers, I disagree with that pretty strongly.



In this case, it was not the numbers I was looking at to make the comment. Just that ISU was new and not worked in so there were more 'lucky' aspects, versus GV which had been around and popular.

Thanks for all the work Rodney. That 42 at ISU sure sticks out on the visual presentation.

Chuck, I pretty much agree with a lot of what you are advocating regarding spread, but I find it kind of interesting that even with much higher SSAs, only Big Creek really produced a scoring spread much greater than the two short courses.

A little off-topic, but related. Is the PDGA considering ways of educating TDs about some of the design issues y'all work on. Specifically, the issue of holes that produce little spread and encouraging TDs to address them. I'm imagining a little primer handed to TDs that educates them on design principles, shows them how to do a statistical analysis, and suggests ways for them to improve their course and make it "tournament worthy."

One thing I've done when analyzing scoring spreads related to determining par on a hole is to eliminate any numbers that were more than 3 from the current average, then recalculate it. That takes care of fluky numbers like a 12 that a player took on a par 4 hole when we were setting pars for PW2001. For example, if your current average is 4.3, any scores of 8 or higher would be dropped.

Much of this scoring dialog is being incorporated in materials developed for and used by members of the Disc Golf Course Designers (DGCD) group. There are several TDs among our 70 members but obviously there are many more TDs out there who are not members. Houck's design articles have touched on these concepts so it's not like a lot of TDs haven't heard of these ideas. But as you can see, it takes a fair amount of effort and knowledge to do these calculations. I think Rodney, Nick and myself (who also happen to be members of the Ratings Committee) are the only ones I'm aware of who have done this publicly. We're planning our first ever DGCD convention next year with some seminars so this will likely be one of the topics. We're using these concepts to develop and analyze the courses being built at Augusta. So we'll be able to use this center to spread the knowledge to a wider audience which will hopefully rub off on the TDs.

I wonder if Robert Trent Jones was a statistician?

Ball golf course design has been more mathematically structured than disc golf course design, and they have less flexibility due to the difference between the available routes for balls versus discs. So, that's both a benefit and challenge for us.

I finally got around to reading the Fall 2004 Special Worlds Edition of Disc Golf World News.

I found it highly ironic that Mark Ellis made the following comments on page 42:


If you compare Barry's 25 down at ISU with what the rest of the top-ten finishers recorded, you might conclude that Barry won his second consecutive World title, in large, due to ISU.



This is the exact same sentiment that I was trying to express on this thread. While at the same time I have no doubts that because of Barry's solid and steady play during the other eight rounds he still would have prevailed, I think there is something wrong when one course was so hugely influential in deciding our world champion. I still believe strongly that efforts should be made to try to prevent this sort of anomaly at future world championships.

I think the anomoly is Barry rather than the course.

The short, technical course is justified in a Worlds by being a test of technical skill off the tee and putting. You might argue that luck would play a large part on such a course and I would tend to agree that a player could get hot and make up a lot of strokes over the course of a round, skewing the curve a bit. That's why you have a 2nd round at that course to even the odds. The fact that Barry dominated the field a 2nd time indicates that he was just the more skillful player, not the luckiest. Barry can thank his own skills for his title, not ISU.

One viewpoint that was PMed to me over the weekend by a member of the Wisconsin crew is the following:


ISU course is just like Barry's home course Vollrath Park in Sheboygan, Wisconsin. Trust me all of us from Wisconsin that were at this years Iowa Worlds were saying "oh my god Barry is going to eat up ISU, it's just like Vollrath".



This is something I hadn't considered at all.