Par Talk

[Jukeshoe]

It would be extremely difficult to pin down which particular throws are errorless. However, I think it is a lot easier to identify the score the player expects with errorless play.

When players come back after a round and say "I left four throws out there today", that means errorless play would have been four throws better. They might think it was the four putts they missed on those holes, but it might actually have been the result of bad upshots, or being a tiny bit off on three throws in a row, or any combination of full or partial errors.

They might not identify which throws had errors, but they know the score they should have gotten on those holes.

Fortunately, there is no need to look at or define anything at a more detailed level than the score.

I definitely understand what you mean; however, isn't this a circular argument?

If errorless play is defined by expected score, then how do you know what score to expect if you can't define errorless play?

Also, I think what score players expect coming off a round isn't infallible truth. Saying "I left 4 shots out there" is a matter of opinion. Perhaps someone is super arrogant and non-introspective and claims he left 4 shots out there when really he played worse than he thought and left 6.

Etc.

 

[DavidSauls]

Perhaps it's more of a dismissive argument than circular.

We don't expect experts to make errors, or at least not one per hole. The expected score already assumes errorless throws.

 

[Cgkdisc]

Perhaps it's more of a dismissive argument than circular.

We don't expect experts to make errors, or at least not one per hole. The expected score already assumes errorless throws.

Is an OB penalty a 1-stroke error even if on some holes you can still "par" the hole? Is an OB penalty that also includes a re-throw the equivalent of 2 errors?

 

[DavidSauls]

Is an OB penalty a 1-stroke error even if on some holes you can still "par" the hole? Is an OB penalty that also includes a re-throw the equivalent of 2 errors?

Is O.B. such that you expect the expert to go OB?

If not, in the first case, are you expecting a birdie (without the OB)? Or does it change the error if a great save shot can be made?

 

[Steve West]

I definitely understand what you mean; however, isn't this a circular argument?

If errorless play is defined by expected score, then how do you know what score to expect if you can't define errorless play?

Also, I think what score players expect coming off a round isn't infallible truth. Saying "I left 4 shots out there" is a matter of opinion. Perhaps someone is super arrogant and non-introspective and claims he left 4 shots out there when really he played worse than he thought and left 6.

Etc.

Not quite circular. Maybe recursive with convergence at the second iteration.

As others point out, errors should be infrequent. (Instead of "errors" we could that say an accumulation of minor flaws which result in a higher score should be infrequent, and that big mistakes which cause an increase in score should be infrequent.)

Yep, not infallible. More likely he left 6 throws out there and didn't recognize the two throws that luck gave him. The point is that the expected errorless score is a lot more recognizable than identifying whether each throw was errorless or not.

 

[Steve West]

Is an OB penalty a 1-stroke error even if on some holes you can still "par" the hole? Is an OB penalty that also includes a re-throw the equivalent of 2 errors?

You can still par a hole with good par after an OB penalty. It takes an unusually high level of play, like one would normally use to get a birdie.

For OB with re-throw; in the context of par talk, I would put it this way: Will an OB penalty that also includes a re-throw result in a score that is two higher than the expected score with errorless play?

Yes, sometimes. The question is whether we can identify the score they would have gotten if they had not thrown OB and been forced to re-throw. Often, that will be two throws lower than the score they got. Not always, as sometimes they'll play birdie-level skill to recover a throw. Other times the re-throw might not cost as much as a full throw.

 

[Cgkdisc]

My point regarding holes with OB and OB with re-throw being significantly increased on tour has blurred the line on determining errors and whether some of those errors correlate with skill level or skew toward random distribution.

 

[Steve West]

My point regarding holes with OB and OB with re-throw being significantly increased on tour has blurred the line on determining errors and whether some of those errors correlate with skill level or skew toward random distribution.

OK, but as long as most of the 1000-rated players can avoid the OB, you could have four-throw penalties and still not hide par. Unless you use average score to set par, but as everyone knows by know, par is not average.

 

[Steve West]

BC Open 2019 Raptors Knoll Disc Golf Course Gold; 8102 Feet

Par is good for Intermediate players.

 

[teemkey]

Steve, I looked at the PDGA report of BC Open MPO scores and a 1000 rating was given to 60 (r2=1001, r2=999, r3=997).

RPS (Rating Points per Throw) values are r1=7.8, r2=7.9, r3=7.9.

If your analysis is that a 64 should be 900 rated, then the RPS for a 1000 rated round would be roughly 300% of the PDGA value.

Your methodology may be a little too aggressive.

 

[Steve West]

Steve, I looked at the PDGA report of BC Open MPO scores and a 1000 rating was given to 60 (r2=1001, r2=999, r3=997).

RPS (Rating Points per Throw) values are r1=7.8, r2=7.9, r3=7.9.

If your analysis is that a 64 should be 900 rated, then the RPS for a 1000 rated round would be roughly 300% of the PDGA value.

Your methodology may be a little too aggressive.

I didn't say 64 should be 900 rated. Nor did I say 900 rated players average 64.

I said 64 was a good par for 900 rated players. There is a difference.

Ratings are based on average scores. I'm not computing average score because:

Par is not average.
Par is not average.
Par is not average.​

I'm computing the score that an expert disc golfer would be expected to make on a given hole with errorless play under ordinary weather conditions. At the intermediate level, players make a significant number of errors. Take all those errors (and the unexpectedly good throws) out of the 71.295 average score, and you are left with the par of 64.

Here is the scoring distribution relative to course pars. As you can see (orange bars), par was most common, but players often got scores that were not only one throw higher than par, but sometimes up to five throws higher than par. (Maybe more, I stop counting at 9.)

The blue bars show the impact on the average score relative to par. (This is simply the frequency of the score times the difference from par.) The total size of the blue bars going up is much larger than the size of the blue bars going down. This indicates the average score is much higher than par.

Because course par is to be the most expected score, it indicates course par is a good par for 900-rated players.

 

[teemkey]

I didn't say 64 should be 900 rated. Nor did I say 900 rated players average 64.

I said 64 was a good par for 900 rated players. There is a difference.

If 64 is a good par for 900 rated players, then a round of 64 should be rated 900. What, exactly, are you parsing here?

Ratings are based on average scores. I'm not computing average score because:

Par is not average.
Par is not average.
Par is not average.​

I only cited round ratings. What are you fixating on?

I'm computing the score that an expert disc golfer would be expected to make on a given hole with errorless play under ordinary weather conditions. At the intermediate level, players make a significant number of errors. Take all those errors (and the unexpectedly good throws) out of the 71.295 average score, and you are left with the par of 64.

So you're saying that you expect a 900 rated player to shoot 71, but that 64 would be a "good par" for that player? You're even saying that you expect errors, but don't include them in the computation of par.

Maybe you have trouble seeing what I'm seeing -- trees vs forest. When you publish these charts for FPO, you change your definition of 'expert' don't you? So my natural thought is that this chart might be for 900 rated 'experts.'

Now it seems you are warping your own interpretation of the PDGA definition of 'par' to produce a par for non-expert levels, that players at the selected non-expert level are not expected to match (unless they play above that level). Is that what you're saying? Why?

Here is the scoring distribution relative to course pars. As you can see (orange bars), par was most common, but players often got scores that were not only one throw higher than par, but sometimes up to five throws higher than par. (Maybe more, I stop counting at 9.)
...
The blue bars show the impact on the average score relative to par. (This is simply the frequency of the score times the difference from par.) The total size of the blue bars going up is much larger than the size of the blue bars going down. This indicates the average score is much higher than par.

Because course par is to be the most expected score, it indicates course par is a good par for 900-rated players.

The 'most expected' score in a population of expected scores is the median, but there is no reason to say the median should be par among your arguments scattered throughout the nearly 4,000 posts in this thread. (Until now, of course.)

I think a lot of your analyses are helpful. Laying out the scoring separation, correlation to rating, the star diagrams and performance charts are pretty cool. As I've said before, it would be nice to see comparisons from year-to-year when the same course is played, and/or when a hole is tweaked or a new hole added.

I understand you're the only one doing these analyses (at least on DGCR), and sometimes your descriptions include unusual uses of commonly used words. Maybe it's a argot specific to disc golf statisticians, but you shouldn't be surprised when you're misunderstood (nor when posters fail to remember, or refuse to use your definitions in their posts).

 

[ChrisWoj]

If 64 is a good par for 900 rated players, then a round of 64 should be rated 900. What, exactly, are you parsing here?



I only cited round ratings. What are you fixating on?



So you're saying that you expect a 900 rated player to shoot 71, but that 64 would be a "good par" for that player? You're even saying that you expect errors, but don't include them in the computation of par.

Maybe you have trouble seeing what I'm seeing -- trees vs forest. When you publish these charts for FPO, you change your definition of 'expert' don't you? So my natural thought is that this chart might be for 900 rated 'experts.'

Now it seems you are warping your own interpretation of the PDGA definition of 'par' to produce a par for non-expert levels, that players at the selected non-expert level are not expected to match (unless they play above that level). Is that what you're saying? Why?



The 'most expected' score in a population of expected scores is the median, but there is no reason to say the median should be par among your arguments scattered throughout the nearly 4,000 posts in this thread. (Until now, of course.)

I think a lot of your analyses are helpful. Laying out the scoring separation, correlation to rating, the star diagrams and performance charts are pretty cool. As I've said before, it would be nice to see comparisons from year-to-year when the same course is played, and/or when a hole is tweaked or a new hole added.

I understand you're the only one doing these analyses (at least on DGCR), and sometimes your descriptions include unusual uses of commonly used words. Maybe it's a argot specific to disc golf statisticians, but you shouldn't be surprised when you're misunderstood (nor when posters fail to remember, or refuse to use your definitions in their posts).

What he's saying makes sense to me. An average score doesn't have to be the par for the course. If that layout was designed for 900 rated players, it was designed specifically as a track that would play difficult for 900 rated players. A course can be appropriate for a skill level and play above its appropriate par for those players, that's how it comes together intuitively for me.

 

[DavidSauls]

Among other things, the expected score for the course isn't the sum of the expected scores of the holes.

 

[Steve West]

...
I understand you're the only one doing these analyses (at least on DGCR), and sometimes your descriptions include unusual uses of commonly used words. Maybe it's a argot specific to disc golf statisticians, but you shouldn't be surprised when you're misunderstood (nor when posters fail to remember, or refuse to use your definitions in their posts).

I'm no longer surprised when I'm misunderstood. Befuddled, maybe. Anyway, thanks for asking for clarification. I'll try.

1. Good catch, for the chart with White bars I should have said that "I'm computing the score that an Intermediate/900-rated disc golfer would be expected to make on a given hole with errorless play under ordinary weather conditions."
Whether that should be called "par" for Intermediate players - or whether Gold par is the only par - is an unresolved question. My point was that par is too high for Open players by a couple of skill levels.

2. Par is based on errorless play. There is a difference between errorless play and average score. Round ratings are based on average scores. So, a score equal to "Intermediate par" will not generally be rated at 900; usually it will be rated higher.

3. I did not mean to imply that the median score is the one that should be chosen for par. (It's an OK choice, not my favorite.) Only, that the heap at par indicates the course par would appear to work well for these Intermediate players - if there was such a thing as Intermediate par.

 

[TAFL]

So you're saying that you expect a 900 rated player to shoot 71, but that 64 would be a "good par" for that player? You're even saying that you expect errors, but don't include them in the computation of par.

I think this may be where you're getting lost.

The definition of par expressly calls for errorless play. So, no, par won't be set with the expectation of strokes due to error.

Par would be set with the assumption of errorless play--and nobody expects 900-rated players to play without error, so a good score for such a player is likely to be well over par. If those players were to be expected to play without error--or be able to recover from errors with heroics--then they'd be 1000-rated experts.

 

[Steve West]

I think this may be where you're getting lost.

The definition of par expressly calls for errorless play. So, no, par won't be set with the expectation of strokes due to error.

Par would be set with the assumption of errorless play--and nobody expects 900-rated players to play without error, so a good score for such a player is likely to be well over par.

Right.

If those players were to be expected to play without error--or be able to recover from errors with heroics--then they'd be 1000-rated experts.

Not quite.

"Errorless play" for an Intermediate player will not usually be 1000-rated.

It is not an error for a 900-rated player to fail to throw a full 400 feet, like it could be for a 1000-rated player. So, errorless play for an Intermediate player would usually still be worse than average play by 1000-rated players.

At this event, errorless play by a 900-rated player would have been rated 967. However, I suspect this course had a lot of punishment for errors. That would inflate the errorless rating. So, 967 might be untypically high.

 

[Steve West]

ET#7 - Estonian Open 2019 powered by Prodigy Disc, Kõrvemaa Prodigy Disc Golf Park, Modified Layout; 18 holes; Par 59; 2,246 m

Apparently, they used Advanced pars for Open players. Which makes FPO come out OK. Hole 1 maybe could be higher par for FPO, but there weren't many near-930 rated FPO players and if you also mix in the 930-rated MPOs, it stays par 3.

 

[TAFL]

Not quite.

"Errorless play" for an Intermediate player will not usually be 1000-rated.

It is not an error for a 900-rated player to fail to throw a full 400 feet, like it could be for a 1000-rated player. So, errorless play for an Intermediate player would usually still be worse than average play by 1000-rated players.

True. Mea culpa for not covering that angle.

Even my erorr-free rounds wouldn't warrant a 1000 rating because I don't throw near as far as experts.

 

[Steve West]

European Open 2019, The Beast, European Open 2019; 18 holes; Par 64; 2,738 m

Par might have been a couple too high for MPO and a couple too low for FPO.

MPO had a birdie-rich environment, while FPO had lots of holes aspiring to a higher par.