What in the world is "Pope of Slope" talking about here? (new WHS par adjustment)

Den Knuth has an article up in Golf Digest. LINK

He mentions that in the new format, your course handicap is no longer just [index X (Slope/113)] but is now [index X (Slope/113) + (CR - Par)]. Ok, fine. I like this. I actually play a fair amount of competitions where players play from different tees and this is going to help some people recognize why we adjust from different tees. It gets weird when someone goes, "what do you mean I'm an 8? I play from these tees all the time and I'm a 10." In other words, in the current system, your COURSE HANDICAP changes simply because someone is playing another set of tees. This is very difficult to explain to people.

In this system, that person is going to be an 8 from those tees, all the time, regardless of whether people are playing from other tees or not. What's different is that MY course handicap will now change as I change tees (it used to by a little, but for a lot of common slopes, it wouldnt'). This is much more intuitive. "Oh, I get 8 strokes from the whites and 10 strokes from the blues?" That makes sense. And it DOES. My scoring average from two sets of tees at my course is almost exactly the difference in course ratings, but my course handicap is the SAME from both sets of tees.

So, that's it. But, this apparently is driving Knuth crazy.

Knuth : "Let’s start with the fact that par is hardly the most reliable measure of course difficulty (that would be course rating). Almost any golfer can list two courses that are both par 72s but vary greatly in how tough they play."

Me : Right, but we ARE abandoning par here, effectively. We're just using it as a baseline to adjust TO. Whether par is 65 or 85, it's just going to scale all the course handicaps. We just use it to "centralize"; we're not getting rid of CR. If you're playing an "easy" set of tees, you're going to have strokes taken off; if you're playing a "tough" set of tees, you're going to have strokes added on. There's almost always a set of tees that has a CR near par, and that's the baseline. (as a person who runs tournaments where people play from different sets of tees, picking one as the "baseline" can actually be tricky.

Knuth : "For example, where once a course handicap was a 12 from the back and middle tees, and an 11 from the front, under the new WHS calculations there will be much larger variations—as many as 18 shots in some instances—between tees. Part of the reason for this is that during the calculation, an approximation is being approximated again by adding Course Rating minus Par, creating an imperfect “over-spreading” of the course handicaps. Golfers moving to longer tees will think this is a logical change (they’ll be getting more strokes). Golfers playing shorter tees won’t be so happy."

Me : what is he talking about here? Under the new WHS, there will be as much as 18 shots in your CH between tees? He's drunk. Is he talking about a 35 index playing par 71 Bethpage Black at 78.1/150 and then turning around and playing par 71 "Field of Daisies" from the red tees at 65.1/100? There is no world in which the bulk of golfers (say between 10 and 24 indexes) playing "normal" tees are going to be seeing differences in their CH's of 18 strokes. And, he says, "an approximation is being approximated again". Yes, but not really. It's just a re-scaling that is the same for everybody. He's acting like we're replacing one problem with two problems. We're not.

Knuth : "Let me try to apply this to real-world scenarios. Living near San Diego, I’ve explored a large sample size of courses in Southern California and found a male golfer with a 14.1 Handicap Index under the old USGA system has a range of course handicaps from 12 to 18 with an average course handicap of 15. Under the WHS, due to the par adjustment, his range now varies from -1 to 22 with an average course handicap of 11.6, effectively giving him more than three shots less in the process. "

Me : What real world? Where a guy goes from tips are Torrey Pines to the easiest Muni in the county? Sounds to me like he's trying to find a scary scenario to bolster his point but doesn't actually give me a real world example.

Knuth : "Players who play from the same tees at their club will see another inconsistency. Say your Course Rating is 70.5 and par is 71. The calculation of Course Rating minus Par comes out to -.5. Based on the way course handicaps are calculated, half the players will have their playing handicap drop one shot lower in the WHS but the other half will not; it all depends on how that -.5 impacts your course handicap number and whether you will get to round up or round down your final number. As we all know, one stroke often determines handicap-event outcomes."

Me : This is ALREADY the case with course ratings. An 11.2 can be a 12 while an 11.3 can be a 13. Move them a set of tees, and they can both be a 12 all of a sudden. This is Knuth replacing one problem with another and acting like we've added one problem. We haven't. In the old system, a 12.2 and an 11.7 both went to 12. Now, one might be a 10 and one might be an 11. But, this is no more incorrect, than the old system changing an 11.4 to 11 and 11.6 to a 12. There is no additional problem here. We've just changed the courses where the CR/SR will come out different. He's either playing games intentionally in this article, or he really doesn't get it.

Knuth : "But to what end will it serve? Golfers competing from more forward tees will be receiving fewer strokes than is truly equitable."

Me : This doesn't make sense at all and to the extent that it does make sense, it's just wrong. If you've made it this far, this needs no further explanation. It's not inequitable against players from other tees, or from your own tees. It's just a baseline shift.

Knuth : "The other argument USGA officials make supporting the par handicap is that the switch eliminates the need in the old USGA system for an additional adjustment when two players are competing from different tees. Replacing the misunderstood section is a good idea, so long as it can be done reliably. But that’s not the case because in the WHS, when the pars of the two tees are different, the same correction must be applied to account for par difference."

Me : I'm sort of with him here, but this is a VERY VERY minor issue. The main times you see par differences (and I don't care about your one-off examples) are between men and women tees, but how many competitions do you play in where men compete against women? Very few. This confusion does not make up for the overall improvement in this system.

Me, theorizing : Knuth was instrumental in creating the current CR/SR system we have now. All of his writings at his web page are based on it. We're leaving it behind, to an extent, and he's a little miffed. You can feel it in his last paragraphs in the article. . ."In my opinion, the USGA should opt out of the Course Rating minus Par modification as well and take more time to gather more data on the impact of the par adjustment. What is the rush? I made that appeal to my friends at the USGA over the summer last year, as did some of their allied golf associations, to no avail.

The way I see it, however, by implementing all of the new WHS elements, we have adopted a World system that is not as accurate as what we have enjoyed. The WHS alters a 40-year-old course-rating-based handicap system in favor of a “net par.” "

Sounds like sour grapes. It's certainly not based in logic.

I have nothing but respect for Mr. Knuth, but I like it and think it's with a try. I agree with you City, as a course administrator this will be painful when we first open and my guys see their new handicaps from the front tees, but will be much easier as the year goes on and they understand it. Hopefully it works, but perhaps he will be proved correct in the end...

I read that article. Sounds like sour grapes to me.

In the previous system, to me at least, it made ZERO sense that a scratch golfer was getting no shots from every single tee. Whether it was a 68.5/113 course or a 75.6/138 course. Still getting zero shots.

Yet anyone with a numbered handicap, their shots would change depending on what tees they played.

In the previous system, the easy way to beat a scratch golfer was to play them from the easiest tees they’d let you. Our yellow tees I’m a 6, our blue tees I’m a 7. There’s only a shot difference. But the CR goes from 71.0 on the back tees to 64.4 on the gold. The front tees are almost 7 shots “easier” yet it only costs me a shot.

It’s easier for the scratch also, but par is par. Scratch players have a much easier time shooting the CR when it’s 71 or 75, than shooting the CR when it’s 64.

THAT was the flaw in the old system. The new system makes that correction. And the new system makes it more intuitive for playing from mixed tees.

Hmmm, that makes a lot of sense in the context of that example.

Of course, it's the rules forum and I'm waiting for a few posts taking issue with the example, lol, but to this non-expert rules "fan" that makes a lot of sense.

That did make a lot of sense. Especially from my perspective. The guy getting the short end of those front tee players.

Lol but it kind of makes me ill for a second thinking of all those “ well just play better ass-hole “ shots I’ve taken for complaining about it. What gives there ? It’s been beat into my head that the stats say the low guy will win most often. The scenario you describe seems to throw shade on that for sure. No? Not shot at you personally just the culture in general.

It’s been beat into my head that the stats say the low guy will win most often

That relates to matchplay. Matches are normally played from the same tees by both players so the (CR-Par) is irrelevant

In strokeplay, wins are in proportion to the number of entrants in each handicap range. ie if 10% of the field is <5 handicap, then it is 10% probable that the winner will be from that cohort.

Just realized that in the 45 page thread about WHS that DaveLeeNC commented on this article. I thought Knuth writing about it was big enough to warrant its own thread. Anyway, seems like I'm not the only one turned off by Knuth's obtusity.

Newby said: In strokeplay, wins are in proportion to the number of entrants in each handicap range. ie if 10% of the field is <5 handicap, then it is 10% probable that the winner will be from that cohort.

I surely think the numbers can be massaged to appear that way. From the testing I have run, using the USGA published odds, the lower handicaps lose considerable ground when a medium to large amount of higher (than them) handicaps are introduced.

I agree with your post but think Knuth should be cut some slack. Perhaps there is something we are missing in understanding his argument. It would be nice to see someone from the magazine post a rebuttal.

If 50% of the field has handicaps of over 25. the odds of any one of them winning are even.

The figures were taken from thousands of competitions in separate studies in Scotland, England and the USA

This taken from the Scottish study. I'm afraid I can't reproduce the graphs

•      When the distribution of winners by handicap category is related to their representation in the field, it can be seen that all handicap categories win in reasonable proportion to their entry i.e. Category 1 and Category 3 players typically comprise 8% and 40% of a club competition and in a ‘single class’ competition win 10% and 38% of the time.

•      This would suggest that handicapping is acceptably fair and equitable throughout the handicap range.The distribution of ‘Winners’ with respect to playing handicap, assuming all competitions were run as a single class handicap event, can be looked at another way:

•      Club handicaps are typically distributed as below left (aggregate of the handicap distribution of players playing in three or more competitions p.a. in a range of golf clubs)

•      The average playing handicap of the players in the sample was 14.

•      Below right is the distribution of winners by handicap derived from information published in the Herald Club Golf Returns over three years and embracing 2622 competitions (approximately 250,000 rounds of golf!)

•      It can be seen from direct comparison of the two distributions that there is good correlation between the number of players at a given handicap and their winning frequency.

•      The average handicap of the winning player was found to be 13.3 which bears very favourable comparison to the average club handicap of 14 (discounting those members playing less than three qualifying competitions p.a.)

I read that article too and I am no expert on this, but I would be inclined to believe Mr. Knuth's arguments for now (as he is an expert on this) and let's wait a few months to see how this plays out in the real world. Adjustments and tweaks can be made later but to bash something new right now re old vs. new system is not appropriate yet as we do not have actual impact or history to rely on. I mean, this is only day 2 of the new system. Seems all the arguments are about the "what if" and "but, but...." but nothing has happened yet, so therefore anything that "could" happen is not a fact. Let's see how this plays out.

'let's wait a few months to see how this plays out in the real world.'

At least here in Finland we have used that principle since the introduction of sloping system, that is many many years. I am sure my Finnish fellow-posters will correct me if my memory fails.

As City wrote, par is only a number used to compare. After all, the par will be the same from all of the tees (even though there are some very few courses where a single hole may have a different par for men and women but that does not change anything as the total par of the course is the same for all players of the same gender).

My wild guess is it has been in use since 2000 in all the countries using the EGA handicapping system:http://www.ega-golf.ch/page/ega-handicap-system

The main task of this Committee was to establish, maintain and regulate an EGA Handicap System that all EGA members could use in the future. It took several years of consultation with the National Associations to gather all the necessary information, but finally the EGA Handicap System was introduced on 1st January 2000.

The EGA Handicap System is based on three well-known cornerstones:1. Part III of the CONGU Scheme 1983, amended edition 19972. The Course and Slope Rating System of the United States Golf Association3. The Stableford System to achieve Stroke Control

The purpose of the system is to produce fair playing handicaps that are adjusted to the relative difficulty of the course being played and to achieve equity and uniformity of handicapping throughout Europe.

Wait and see what? Look, I got a little too into the weeds on the initial post.

The long and short of it is this : currently, if you're a 7.1 index and you go play a par 72 course rated 68.5/120, you're an 8 handicap. If you play another set of tees rated 75.0/135, you're an 8 handicap. Now, your "expected" score (good score, really) from either set of tees goes from 76 to 83. But, you're an 8 from either set of tees.

Does that make sense to you?

Or does it make sense that on this par 72 course you should be a 4 or an 11 depending on what set of tees you're playing? That's all this boils down to. If you're playing a stableford, or a fourball against the field, the NEW system is the only thing that makes sense. In the Knuth system, you're getting 8 strokes from the up-tees but if you shoot net 68, that's just shooting your handicap. This doesn't make sense at all. It never has. There were two ways to deal with course rating when the CR/SR system came out, and they picked the wrong one.

The current adjustment is all done behind the scenes by whoever is calculating your handicap, but WHERE IT MATTERS. . .that is, ON THE COURSE where you're USING your handicap, the current system is totally jacked. Within a year -- not a few months -- every golfer is going to go, "oh yeah, this makes way more sense."

If you were a scratch player playing someone playing from those tees that are 7 shots easier, then you'd adjust your strokes by 7 shots. If two people played a match from different tees under the old system, then the person playing from the lower CR tees would lose shots. So if you have a scratch playing 75/138 and a 7 index playing a 68/113, then there would be no shots. 0 x 138/113 = 0 and 7 x 113/113 = 7, but then you adjust for the different course ratings, so the 7 becomes 0 and the match would be flat.

Under the new system, if par were 72 for both players, then the scratch would get 3 shots and the 7 would get 3 shots (0 x 138/113 +75-72 vs 7 x 113/113 +68-72). So the match would be flat.

I think you need the CR-par adjustment to really make the net double bogey work properly. Otherwise if you're a scratch and you play Bethpage Black at 77.7/155 or whatever it is these days, you could only make double bogey. With the CR-par adjustment you're now getting 7 shots, so you can make triple on stroke indices 1 through 7 (at least I assume that's how it will work).

We have been through this before. Reading the report from the past lead me to run my own tests. The results of which show me that low handicappers are at a disadvantage for a full net competition.

What was the size of your sample?

150,000 events and 15,000,000 rounds for the 100 player tests.

Just curious, are these computer-generated models? If so, my general tendency is to give real-world experience a little more credence than computer models. As good as most models can be, as impartial as the model's creator is, its nearly impossible to consider all potential variables. But again, more curiosity, how badly are your results skewed against the low-handicap group?

Where have there been 150,000 events played and where did you get all the results from?

Yes, computer generated. As I stated earlier in this thread 'using the USGA Published Odds'.

I agree about real world vs. computer model. However, who amongst us has access to 15,000,000 rounds let alone 10's of thousands?

https://s3.amazonaws.com/golfwrxforums/uploads/ipb/monthly_12_2018/post-433112-0-60831100-1545484665.jpg

I'm SURE this forum has been over this before, but those published odds are the same as what Dean Knuth has on his pope of slope page.
Somehow, these numbers became some kind of gold standard through the years.
Well, not somehow. . .because a guy with degrees worked with the USGA and put them out there and everyone just went, 'welp, those are the numbers'. It's unclear where they came from, but I suspect they're 1 of 2 things, neither of which is very satisfying to me.
1) There's actually a big data set with millions of scores in it and someone observed these numbers and knows that every score put in was accurate and not put in by either a sandbagger or a reverse sandbagger. The problem with this is that there's a real "chicken and egg" thing going on. For instance, if you see exceptional scores happen more frequently (or less frequently) than that chart would indicate, does that mean you're dealing with sandbagging, or does it mean those odds are actually wrong and need to be updated to reflect the new data that has come in. In other words, what makes the dataset that generated those odds CORRECT, and not the data we're seeing now? Which makes me think it was actually generated like. . .
2) They looked at a bunch of scores from groups of golfers and created a model with a specific mean and standard deviation, and used that to extrapolate to the extreme cases. In other words, they said, "oh a 7 handicap has a standard deviation of 3 centered around 10 over the course rating and if we fit a gaussian model to that, then we can say that there's a 1 in 125,000 chance of shooting a -10 differential." But this kind of thing is the exact reason why statistical models are simply that. . .models. I would highly doubt (from my experience as a golfer and statistician) that a normal distn is good at capturing extreme cases (look up something called "Kurtosis" for more on this).
They might have done something more clever than a normal distribution (which would indicate that I'm just as likely to shoot 6 over my handicap as 6 below it), so they might have used a Gamma/Poisson distribution. But, the same problems with kurtosis arise.
All of which is to say that it's POSSIBLE @HatsForBats ran a simulation whose assumptions are based on purely modeled data that might even have been modeled in order to generate the exact results that he saw.
And that doesn't mean they're wrong, but without knowing where the USGA/Knuth got those original numbers, it certainly doesn't mean they're right.

Actually the USGA went to a different chart. I would link to the actual USGA posting but I find it very hard to find old stuff since they changed to the new site.

At least here is a PDF Image of the chart -
USGA - Appendix E Chart (Odds).pdf

Oh, cool. So that page actually says, "The values in the table were derived from an analysis of over 7.3 million scores posted by players with a Handicap index."

So, that's interesting, and that makes your simulation stronger.

What they DON'T do here is what Knuth seems to use the chart for. . .pointing out sandbaggers. This is just data presented without conclusions. Link

OK, my $0.02 worth (which is AUD so worth a tad less on the world scale) ...

The 'slope' and 'course rating' system was a huge move forward for handicapping.

But.

With ONLY the 'slope' used in the Daily (aka Course) handicap calculation there was too much inequity.

The whole concept of "course rating" is to re-assess the 'Par' for a SCRATCH golfer. If a SCRATCH golfer wasn't expected to be able to shot to the PAR of that course / those tees, what chance do use mere mortal bogey golfers have? So, why wasn't it always included in the calculation? I think that is where the error lies. The problem / issue is not that's it being included now. The problem / issue is it was never taken into account previously.

The 'slope' adjustment in the calculation (old and new) provides us bogey golfers with the additional 'equity' adjustment for the progressive difficulty of the course(s) / tee(s) we play.

I was leafing through the "old" USGA Handicap Manual, and noticed something that relates to the "accuracy" of the system. I had forgotten this, but when calculating Handicap Index, the 10 best differentials were averaged, then multiplied by 0.96, and then the resulting number was TRUNCATED to the lower full tenth. That's right, not rounded, but truncated, so 5.3999 became 5.3. This means that just about half of the golfers had their Handicap Index lowered as compared with the (arguably) more accurate method of rounding to the nearest tenth. To me, the WHS procedure, rounding to the nearest tenth, improves accuracy.
Knuth is also concerned that the CR-Par) term will introduce rounding issues. Rounding issues are inevitable when we round to the nearest whole number for Course Handicap. A problem with the old system was that the Course Handicap was calculated, and rounded to the nearest whole number, and then any adjustment when playing different tees was made. But that "tee adjustment" was itself rounded, so rounding concerns could be doubled, or could nullify each other. The WGS system includes the (CR-Par) term in the initial calculation of Course Handicap, the final result is only rounded at the end. Any rounding conflicts are minimized by using a single rounding, rather than rounding twice as was done before. To me, this also improves accuracy.
And for @HatsForBats , that old Handicap Manual is still online here: https://www.usga.org/handicapping/handicap-manual.html#!rule-14367

Does anyone know of any courses where the slope is less than 113 or negative?

Those hole lengths come from the scorecard. Several of them are off by 70 or more yards.

I just read an an article on the subject matter and it said there's not a single full-length course in Finland with such a low Slope. We have about 120 full-length courses if I remember correctly.