DGCR Scratch Scoring Estimate

[Stud Muffin]

Originally Posted by Dave242

Maybe we should change the name to Pro Scoring Estimate (PSE)?

^ That would make more sense, because scratch relates to 'par' and 'recreational par' and 'professional par' are 2 completely different scores. The formula made here reflects what a 1000 rated player would shoot, and the number is going to be distorted from recreational tees. As I said, at Engler, from our tournament tees, Bunnell shot -2, and won the tournament. Other open players shot over par from those tees.

It is a good scoring estimate for a 1000 rated player, so (PSE) would make more sense, my 2 cents.

[Dave242]

Originally Posted by Stud Muffin

^ That would make more sense, because scratch relates to 'par' and 'recreational par' and 'professional par' are 2 completely different scores. The formula made here reflects what a 1000 rated player would shoot, and the number is going to be distorted from recreational tees.

I am not sold on this reasoning. While "Par" is not nearly as standardized in DG as in BG (we are lightyears behind), you have the same confusion in BG where the term Scratch comes from......just like you mention with us, you need to know what tees the scratch BGolfer is teeing from.

The official definition of "Scratch" does not include anything to do with par:
the USGA defines a scratch golfer thusly: "An amateur player who plays to the standard of the stroke play qualifiers competing in the United States Amateur Championship. The male scratch golfer hits his tee shots an average of 250 yards and can reach a 470-yard hole in two shots. The female scratch golfer can hit her tee shots an average of 210 yards and can reach a 400-yard hole in two shots."

[Stud Muffin]

Originally Posted by Dave242

I am not sold on this reasoning. While "Par" is not nearly as standardized in DG as in BG (we are lightyears behind), you have the same confusion in BG where the term Scratch comes from......just like you mention with us, you need to know what tees the scratch BGolfer is teeing from.

The official definition of "Scratch" does not include anything to do with par:
the USGA defines a scratch golfer thusly: "An amateur player who plays to the standard of the stroke play qualifiers competing in the United States Amateur Championship. The male scratch golfer hits his tee shots an average of 250 yards and can reach a 470-yard hole in two shots. The female scratch golfer can hit her tee shots an average of 210 yards and can reach a 400-yard hole in two shots."

Yes, the tees mean everything.

In Bolf, there is one great 'par' tournament every year, the U.S. Open. The goal is for the winning score to be 'par'. They set back the tees to incredible lengths, and make the hazards as extreme as possible. The winning score was +1 this year. Imagine that, a course so tough, the best in the world could not par the course.

But scratch is defined "Golf a handicap of zero, indicating that a player is good enough to achieve par on a course." <~I got the definition from my Apple dictionary, have no link.

Still, it is common for bolf, for par 5's to average under par, and par 3's to average over par in tournaments.

All this random info just to say, par is nt set in stone, it is just a general rule of thumb. I believe your formula is more accurate as to judge your game against professional level play.

[smyith]

course par is different from pro par.
personally i think that the SSE should be a true course par (what a golfer would shoot if they can throw ___ distance with high accuracy and makes a majority of their putts inside the circle. Or about what a 900 rated golfer would shoot on average.) and this is what is think the SSE should be
pro par is SSA to me. What a 1000 rated player would shoot on average. and what as a competitive player i am striving to shoot on any course.
tournaments need to move away from using the course par mentality and using the SSA as par. i think it makes the support look "easy" or "dumb" if pros are finishing 2 round tournaments at -15 or better.
on the whole "tee/ pin placement" discussion. does anyone remember the old SSA pages on pDGA? they would indicate what tee and pin placements were used for that round. I miss those wonderfully informative pages. its too bad the pDGA felt the need to remove them for bs reasons (some one might try to steal the info and use it to compete against us....lame!)

[jenb]

IF SSE isn't intended to mimic the PDGA SSE, then what's the point? That's all people really care about. Anyone who cares about what their "round rating" would be is really interested in what it would be in the context of PDGA round ratings.

So if anything should have weight in the courtroom of what would be a good SSE calculation, its the PDGA app's calculation using real round ratings for courses. If we can't call the DGCR SSE round ratings to task for failing to correlate with PDGA app generated round ratings, then you might as well ditch the DGCR SSE ratings entirely. It seems like a bunch of "locals" are too entrenched in defending the system they engendered to recognize that reality, or take any criticism, no matter how constructive.

[Dave242]

Originally Posted by jenb

IF SSE isn't intended to mimic the PDGA SSE, then what's the point? That's all people really care about.

I for one care almost exclusively about PDGA SSAs and round ratings. The truth is however, that most people here are not PDGA members....so something more general like "what would a 'Pro' score on this course?" is also a very meaningful question to the majority.

Originally Posted by jenb

So if anything should have weight in the courtroom of what would be a good SSE calculation, its the PDGA app's calculation using real round ratings for courses.

Agreed. When they port it to Android, I will will be even more happy because I will be able to use it.

Originally Posted by jenb

If we can't call the DGCR SSE round ratings to task for failing to correlate with PDGA app generated round ratings, then you might as well ditch the DGCR SSE ratings entirely.

DGCR SSE actually tracks PDGA SSA quite well. As has been discussed recently, things do fall apart for short courses. How does the PDGA app handle really short courses? I do not have it so I am very curious.

I imagine that if it is based on round ratings like you say, it will never have data for any 9ers or courses with SSAs <43 (or whatever the cutoff is). So, for all the short-comings of the SSE on short courses, does the PDGA app do better?

If it is indeed PDGA-minded folks who care about this stuff, do they really care about the SSAs and SSEs on 9ers and short courses?

Originally Posted by jenb

It seems like a bunch of "locals" are too entrenched in defending the system they engendered to recognize that reality, or take any criticism, no matter how constructive.

I'm not sure if this is aimed at me, but I assume partly so since I have been one of the contributors to the conversation and one of several collaborators on the SSE algorithm (you were one of them too).

The short course issue is indeed a place where SSE fails. What other constructive criticism has been offered? Your example of the Cherry Creek SSE error I believe was fixed when you updated the course correctly to "moderately wooded".

The SSE was implemented by timg after a lot of collaboration, arguing/criticizing, and comparing with reality (you were part of all that). I'm sure if you have great input that makes things better, timg in his normal fashion will implement it.

[Dave242]

Originally Posted by Dave242

The truth is however, that most people here are not PDGA members....so something more general like "what would a 'Pro' score on this course?" is also a very meaningful question to the majority.

Oh, and I meant to add this: Calling it PSE rather than SSE also makes it a little more non-specific so as to account for the variations that exist between SSE and SSA that will 100% certainly be there no matter what algorithm is used and no matter what manual methods of making adjustments to courses' SSE to match their SSA.

BTW, I would really not be happy for my own sake if it changed to Pro Scoring Average (PSE) since I like the similarity of SSE and SSA.....but my opinion is that there is valid reason to do so.

[jeverett]

Originally Posted by jenb

IF SSE isn't intended to mimic the PDGA SSE, then what's the point? That's all people really care about. Anyone who cares about what their "round rating" would be is really interested in what it would be in the context of PDGA round ratings.

So if anything should have weight in the courtroom of what would be a good SSE calculation, its the PDGA app's calculation using real round ratings for courses. If we can't call the DGCR SSE round ratings to task for failing to correlate with PDGA app generated round ratings, then you might as well ditch the DGCR SSE ratings entirely. It seems like a bunch of "locals" are too entrenched in defending the system they engendered to recognize that reality, or take any criticism, no matter how constructive.

Hi Jenb,

The overall problem is lack of inputs into the formula. In the original thread discussing the best implementation of the SSE formulas, we extensively discussed using 'known' PDGA SSA and course length pairs to calibrate the formulas against. I wrote up a formula that would do that, but Timg didn't want to use it (for a number of reasons). There is also definitely the matter of these *very* short courses not actually having events on them at all.. so we're estimating what an SSA for a course *might* be, if it actually were to have one.

The only other thing I can think of, and I hesitate a little to suggest it as it would mean going back and looking at a ton more SSA data, would be to switch the linear formulation based on course length to an exponential one. I think it would take quite a lot of data to make it accurate, and again it would still only be as accurate as 'lightly wooded', 'moderately wooded', and 'heavily wooded' labels are in actually describing the technicality of a course. I suggested a number of other methods for increasing the accuracy of that system, all of which would require more data (and fields) in the DGCR course pages.

[mashnut]

On the low end, a curve would make more logical sense, but I'm not sure it's worth messing with since it's not all that important or useful at that extreme anyway. Once you dip below a certain threshold it gets much more difficult to shoot one stroke better and the linear relationship breaks down.

[jenb]

Well, at the very least, how about putting in a floor function on SSE that makes the minimum SSE be (((#holes * 2) + (#holes * 1/3)) +1). At least that would more closely track the PDGA SSAs for short courses.