New World Handicap System

[DaveLeeNC]

Casey, I am aware of it but that is about it. I would assume that the App's approach is similar to what I do. I started doing what I do back in around 2003. When Broadie's work became known I was surprised at the similarities. The big difference is that I compare my results to some arbitrary standard that makes sense for me personally, as opposed to how the field is performing in some event.

It was interesting and on occasion even provided some practical insights. But mostly it simply fed the 'number guy' in me.

dave

ps. I drastically deprioritized golf in my life back in 2015, and quit doing any kind of stats work at that time. You can see that point in my handicap history -:)

[antip]

Perhaps, but that may need a change to 5.6a.

[James the Hogan Fan]

If I have to carry my own index card, so be it. Frankly one of my favorite things about playing a new course as a kid was collecting the pencils and scorecards. (and I do it to this day)

[DaveLeeNC]

Now you collect XML files - pretty much the same thing -:)

dave

[reidsou]

The 2020 "Rules of Handicapping" PDF was published this week. Rule 5.6 discusses the method behind the Playing Conditions Calculation (PCC). However, does not give details of how this is actually computed. I have been told by one golf professional that this will remain a secret. Not sure why.

Researching, I found that PCC is derived from (copied from?) Golf Australia's DSR (Daily Scratch Rating). And I did find a formula for that, but without explanation, so not that helpful.

Does anyone have more information about how the PCC is actually computed? Or clues on how to find out?

[Newby]

Does it really matter? They are not going to change it soon.

[reidsou]

It's for the curious.

Prior to this change I was able to follow along and completely replicate handicap calculations on my own, if I chose to. Now there will be an added black box adjustment. It's a significant philosophical departure. (IMO!)

[Sawgrass]

I agree that one's potential inability to calculate their handicap manually is a big change, but I think it's a change for the better given the perversity of playing in terrible conditions.

 

And if you fail to post on the same day as you play, you'll still be able to do the manual calculation, though that would seem to be manipulation.

[reidsou]

Page 56 of the 2020 Rules of Handicapping: "The PCC is applied to all acceptable scores that are submitted on a day of play and retrospectively for scores played on that day but submitted at a later date." FYI.

[reidsou]

I agree it's a change for the better that differentials will be adjusted for playing conditions. I don't think it's a change for the better that the method behind calculating handicaps may now have a private component.

But maybe the details will be forthcoming. I sent a question to the USGA.

[Newby]

It's only 'added' for the US. In effect, it has been around in the rest of the world for years

Have a look at thishttp://www.golfwa.org.au/cms/daily-scratch-rating/

[DaveLeeNC]

One value that the WHS offers WRT the PCC adjustment that has not been often discussed is that it will tend to normalize court ratings across all courses. Those courses that truly are 'too easy' (or 'too hard') vs their ratings, will see the appropriate adjustments made more frequently. I like that aspect of the new system.

dave

ps. I think of this as more of a 'CRA' (course rating adjustment) vs a conditions adjustment

[Newby]

If you read the link above you will see it has very little to do with how the CR worked out

The DSR system will establish each of the following: (for DSR read modified PCC and for Golf Link read WHS)

• An average net score for the field.

• An average handicap for the field.

• The field size.

• The type of competition (Stableford, Par, or Stroke).

• The gender of the competitors.

Once it has established each of these factors, GOLF Link will compare the average net score it has calculated from the scores on the day, with the average net score it EXPECTS for this precise field composition. 

(The EXPECTED average is determined by GOLF Link from millions of prior rounds.)

GOLF Link will then determine the DSR by using the difference between what happened on the day and what was EXPECTED to happen.

[Newby]

Don't you think that the CR itself is, to virtually all players, a 'private' component unless they have access to the rating team's data?

[reidsou]

Yes, that's a good point. However, the methods for determining the course rating are public. Also, this is a static component, in comparison.

[Sawgrass]

Thanks, you’re right of course. I had mixed it up with the fact that later-posted scores will not be part of the calculation for adjustment — just subject to that adjustment. (All of which is off your original, valid point.)

[sui generis]

Suppose you knew the formula, what would you do with all the data? Say there were a hundred rounds posted at the course on the day you played. They might be played from different tees and by men and women playing with different course handicaps. Would you crunch all of that? To what end? Would you like to double check GHIN's arithmetic?

[DaveLeeNC]

I had assumed that 'expected results' would relate to the 'standard result' (derived from Rating/Slope) that is the entire basis for this whole thing. If that is not the case then what I said is wrong.

 

dave

[antip]

Having lived with this daily rating adjustment in Australia for a few years now, I'll make the following observations.

1) Yes it does produce some minor variation in the final course rating for any given day of competition and that GENERALLY is an intuitive change - that is tougher conditions see it raised and easier conditions can see it lowered.

2) But most days it does not change.

3) And the scale of the change (from the scratch rating), IMO, is notably less than the actual change in difficulty of the course. That is not a criticism, just an observation.

4) The more surprising feature is the frequency with which the daily rating changes but there is absolutely no 'conditions' driver. An example: we recently completed club championships of 72 holes stroke, Saturday and Sunday of succeeding weekends. Course set up varied little, tough all the way but very slightly tougher for rnds 3 and 4. Playing conditions were toughish 25mph winds on day 2 but a horrendous 35mph and gusting to 45mph higher on day 3 (this on a course of over 7000 yards with no dense wind protecting trees). Yet the daily course rating came out lowest on day 3. The explanation: the field (of 150) simply played better than their playing history suggested - past statistics simply didn't apply. Another example is there are days when conditions are absolutely benign, but the daily rating gets moved away from the scratch rating - again, the field has performed inconsistently with their playing history.

5) So, you never quite know what you are going to get.

6) But because the final course rating for a day's play is finalised and published overnight in the handicap database, by that following day you can still replicate the calculations manually if you wish.

A PS Edit for context: in Australia the whole field plays from only one set of tees in the vast majority of competitions.

[Newby]

The method is given in the DSR link that I posted. The exact formula and underlying figures are not. Much the same as the CR.

[DaveLeeNC]

Here is an 'interesting question' that is not important in the long run, but will provide comparative data (depending on the answer).

As I understand things, the new system (when it changes over - supposedly 1/2020 in the US) will be back-calculating 2019 indexes (in the US) to the new system to properly deal with 'soft caps', hard caps, etc. So if you have your old history recorded you then have a years worth of comparative data (although one side of the comparison won't have the soft cap stuff). And will there be an indicator for the case when an index has been affected by either a soft or hard cap adjustment?

dave

[Roadking2003]

For me, entering scores hole by hole while playing would definitely increase playing time. And also add new weight to my bag.

[reidsou]

Yes, at least for a representative sample of scores from our tournaments. It will be interesting to know the algorithm to get my own sense of how scores correspond to course difficulty. Also, for the statistically inclined, a matter of intellectual curiosity.

However, I contacted the USGA and they confirmed the method for calculating PCC will remain a secret, because it's "proprietary". (Why?)

For the first time, a component of handicap calculation will be unknown to players.

Wonder what Dean Knuth will say about this...

[Roadking2003]

Does anybody have a link to the PDF version of this booklet?

https://www.usgapublications.com/products/the-rules-of-handicapping-effective-january-2020?variant=30267885158480

[Roadking2003]

Found it:

https://msga.org/wp-content/uploads/2019/11/Rules-of-Handicapping_USGA_Final.pdf

[Shilgy]

At the WHS seminar I attended on Monday they stated the PCC would likely only come into play a couple percent of the time. Expected maybe 4-5/times per year here in Arizona.

Remember it is only whole number adjustments so there must be statistical evidence of at least one stroke in either direction necessary of change.

 

[Halebopp]

We were given similar information, the CBA should've affected competition scores about 20% of the time and with the PCC it should be even rarer.

[ThinkingPlus]

Seems silly to restrict the adjustment to whole numbers given that differentials are computed to a tenth of a stroke.

[Newby]

Given that playing, course handicaps, scores and differentials are whole numbers, it would seem reasonable to have the adjustment as a whole number.

[antip]

The mathematician in me says surely it would be simplest to have every adjustment process conducted in tenths and then have one single rounding process at the last stage.