The World Handicap System. Is it Good? (Moved from the BUDA thread)

[Erik J. Barzeski]

Garland, try to understand what people are telling you.

The half stroke thing is just how rounding works. It existed in 2019, too. And just as often as it adds a stroke to the gap between two players it'll close a gap, too.

I'll create scenarios where two players play against each other, and I'll adjust each index separately by 0.1 strokes to show you that it can affect either player.

5.5 (6) vs. 9.0 (9) pre-rounding would be a 6 vs. a 9. Now, it's a 5 (5.0) vs. a 9 (8.5). 3 -> 4
5.4 (5) vs. 9.0 (9) pre-rounding would be a 5 vs. a 9. Now, it's a 5 (4.9) vs. a 9 (8.5). 4 -> 4
5.5 (6) vs. 8.9 (9) pre-rounding would be a 6 vs. a 9. Now it's a 5 (5.0) vs. an 8 (8.4). 3 -> 3
5.4 (5) vs. 8.9 (9) pre-rounding would be a 5 vs. a 9. Now it's a 5 (4.9) vs. an 8 (8.4). 4 -> 3

They still balance out. Nobody's getting screwed, the places where you round might just fall in different places. This same exact thing could have happened, and did, in 2019, too, because that's how rounding works. This is also true regardless of what the decimal on the adjustment is, because 0.1 can take a .5 to a .4 and then you're rounding THAT number in the opposite direction, too.

[Wayne_Kozun]

Golf Canada hasn't updated their web site but doing my own calcuation my index has dropped from 10.1 to 9.9 thanks to the best 8 of 20 rather than 10 of 20 multiplied by 0.96.  It seems like most handicaps here in Canada (and the US for that matter) will go down a bit - like 2-5% of the original index.  So on a relative scale there is no significant change.

[Mark Pearce]

I have to say that reading all that maths (and I'm good at maths) just reinforces my initial view that adopting this system world wide is madness.  How can anything this complicated be good?

[Jason Topp]

I have to say that reading all that maths (and I'm good at maths) just reinforces my initial view that adopting this system world wide is madness.  How can anything this complicated be good?

Mark:


In practice it is not very complicated - you check an app for your handicap from the tees you are playing before your round.  After your round you put your score into the app.


Repeat each time you play.

[Sean_A]

I have to say that reading all that maths (and I'm good at maths) just reinforces my initial view that adopting this system world wide is madness.  How can anything this complicated be good?

The formula junk doesn't make any difference because its the same everywhere.  Where the difference lies in how scores are reported.  I don't see any material difference to what was happening last year.  In the US it is extremely easy to cheat or be very lax for any non-competitive scores which are used for handicapping purposes. In the UK, this is far harder to do.  Why ther authorities feel it is important to attempt combining several very different cultures where golf is concerned is beyond me.  There will still be carping about X, Y & Z handicaps because that is what golfers do. The system will never be totally trusted because it is not totally trust-worthy. 

The one aspect of the new system I really like is it will be more dynamic, reflecting more accurately how one is playing recently.  In the UK, it simply takes far too long for handicaps to reflect current play, especially poor play.  Often times, there are good reasons why scores are going up and why they are unlikely come down in the short run.

Ciao

[Garland Bayley]

Garland, try to understand what people are telling you.

Au contraire, You need to understand what Dr. Knuth is telling you.

The half stroke thing is just how rounding works. It existed in 2019, too. And just as often as it adds a stroke to the gap between two players it'll close a gap, too.

I'll create scenarios where two players play against each other, and I'll adjust each index separately by 0.1 strokes to show you that it can affect either player.

5.5 (6) vs. 9.0 (9) pre-rounding would be a 6 vs. a 9. Now, it's a 5 (5.0) vs. a 9 (8.5). 3 -> 4
5.4 (5) vs. 9.0 (9) pre-rounding would be a 5 vs. a 9. Now, it's a 5 (4.9) vs. a 9 (8.5). 4 -> 4
5.5 (6) vs. 8.9 (9) pre-rounding would be a 6 vs. a 9. Now it's a 5 (5.0) vs. an 8 (8.4). 3 -> 3
5.4 (5) vs. 8.9 (9) pre-rounding would be a 5 vs. a 9. Now it's a 5 (4.9) vs. an 8 (8.4). 4 -> 3

They still balance out.

Your calculations are blatantly obvious.

Nobody's getting screwed, the places where you round might just fall in different places.

Your mistake is that you are taking both systems to be equally valid. As Dr. Knuth points out, the use of par in the handicap calculation moves the new system away from being calculated accurate with respect to the course rating system. See the pertinent calculation below.

This same exact thing could have happened, and did, in 2019, too, because that's how rounding works. This is also true regardless of what the decimal on the adjustment is, because 0.1 can take a .5 to a .4 and then you're rounding THAT number in the opposite direction, too.

Using my 17.9 index and Windsong's back tees (CR 75.0, SR 141, Par 71). First old system.
Handicap is 22 and I can play others with indexes from 17.3 to 18.0 without either of us giving a stroke.
Dr. Knuth would tell you this is proper handicapping.

Using my 17.9 index and Windsong's back tees with the new system.
Handicap is 26 and I can play others with indexes from 17.2 to 18.0 without either of us giving a stroke.
So under the new system, I would not get the stroke when playing someone with a 17.2 index in the new system that I used to be entitled to. Dr. Knuth would tell you this is the flaw in the new system.

Using my 17.9 index and Windsong's combo 1 tees (CR 71.3, SR 136, Par 71). First old system.
Handicap is 22 and I can play others with indexes from 17.9 to 18.6 without either of us giving a stroke.
Dr. Knuth would tell you this is proper handicapping.

Using my 17.9 index and Windsong's combo 1 tees with the new system.
Handicap is 22 and I can play others with indexes from 17.6 to 18.4 without either of us giving a stroke.
So under the new system, I would not get the stroke when playing someone with a 17.6, 17.7, & 17.8 index in the new system that I used to be entitled to. And, I would be giving strokes to players with 18.5, and 18.6 indexes in the new system that I didn't use to have to give. Dr. Knuth would tell you this is the flaw in the new system.

[Erik J. Barzeski]

I have to say that reading all that maths (and I'm good at maths) just reinforces my initial view that adopting this system world wide is madness.  How can anything this complicated be good?

I don't think it's all that complicated.

Using my 17.9 index and Windsong's back tees (CR 75.0, SR 141, Par 71). First old system.Handicap is 22 and I can play others with indexes from 17.3 to 18.0 without either of us giving a stroke.Dr. Knuth would tell you this is proper handicapping.Using my 17.9 index and Windsong's back tees with the new system.Handicap is 26 and I can play others with indexes from 17.2 to 18.0 without either of us giving a stroke.

Incorrect.

Old
17.9 * 141/113 = 22.3x = 22 CH
17.3 * 141/113 = 21.58 = 22 CH
18.0 * 141/113 = 22.46 = 22 CH

New
17.9 * 141/113 + 75.0 - 71 = 26.3x = 26 CH
17.3 * 141/113 + 75.0 - 71 = 25.58 = 26 CH
18.0 * 141/113 + 75.0 - 71 = 26.46 = 26 CH
17.2 * 141/113 + 75.0 - 71 = 25.46 = 25 CH

So under the new system, I would not get the stroke when playing someone with a 17.2 index in the new system that I used to be entitled to. Dr. Knuth would tell you this is the flaw in the new system.

First, you're wrong. The math is right there.

Second, just because there's a change (and again, there's no change here) does not mean there's a flaw.

Using my 17.9 index and Windsong's combo 1 tees (CR 71.3, SR 136, Par 71). First old system.
Handicap is 22 and I can play others with indexes from 17.9 to 18.6 without either of us giving a stroke.
Dr. Knuth would tell you this is proper handicapping.

Using my 17.9 index and Windsong's combo 1 tees with the new system.
Handicap is 22 and I can play others with indexes from 17.6 to 18.4 without either of us giving a stroke.
So under the new system, I would not get the stroke when playing someone with a 17.6, 17.7, & 17.8 index in the new system that I used to be entitled to. And, I would be giving strokes to players with 18.5, and 18.6 indexes in the new system that I didn't use to have to give. Dr. Knuth would tell you this is the flaw in the new system.

Oh brother.

Old
17.8 * 136/113 = 21.42 = 21 CH
17.9 * 136/113 = 21.54 = 22 CH
18.6 * 136/113 = 22.38 = 22 CH
18.7 * 136/113 = 22.50 = 23 CH

Range is 0.7 strokes - 17.9 to 18.6.

New
17.6 * 136/113 + 71.3 - 71 = 21.48 = 21 CH
17.7 * 136/113 + 71.3 - 71 = 21.60 = 22 CH
17.9 * 136/113 + 71.3 - 71 = 21.84 = 22 CH
18.4 * 136/113 + 71.3 - 71 = 22.45 = 22 CH
18.5 * 136/113 + 71.3 - 71 = 22.57 = 23 CH

Range is 0.7 strokes - 17.7 to 18.4.

The range shifts slightly (because of the 71.3 - 71) but it's still the same width, the same 0.7 (which is 8 tenths, as the endpoints are inclusive).

[David Harshbarger]

3 tests of a handicapping system:


1.) In head-to-head competitions between players of differing handicaps, are the results, on average split 50-50?
2.) Do the variations in scores by individual players at multiple courses (or different tees on a single course) track the differences in ratings and slope?
3.) In head-to-head competitions between players playing from different tees are the results, on average, split 50-50?


Sounds like the new system may be able to help with some of this as they are trying leverage pervasive mobile technology to get near-real-time (and certainly same-day) scoring back.  Plus, there should be excellent data on individual variations across courses (by those that travel) and within courses.


I have to think that with all of the centralized data collection that should be possible with a centralized app, that the governing bodies should be able to generate relatively consistent rating and slope numbers for every participating course (as long are some reasonable number of players who travel in or out of the course).


(If you reverse the math and set as given the player's abilities, and set as the variable the courses the player plays, then, *in aggregate*, you should be able to calculate relative rating and slope for the courses they play.  Would be interesting to see if the governing bodies provide that data back to the courses, our update course ratings accordingly as the data starts coming through).

[Jay Mickle]

How can anything this complicated be good?

Women: good or bad.

[Erik J. Barzeski]

1.) In head-to-head competitions between players of differing handicaps, are the results, on average split 50-50?
3.) In head-to-head competitions between players playing from different tees are the results, on average, split 50-50?

The USGA has maintained that the better player should still win slightly more often. It might be 51-49, and it's not even 55-45 IIRC, but I think they believe there should be a slight edge for the better player. Before that's what the 0.96 was for, and the 8/20 instead of 10/20 should result in almost the same thing.

But yeah, it's pretty close to 50/50. If it became 60/40 in favor of one group or the other over enough of a sample size, catastrophe! And the new system doesn't really change the basic math in this regard.

[David Harshbarger]

1.) In head-to-head competitions between players of differing handicaps, are the results, on average split 50-50?
3.) In head-to-head competitions between players playing from different tees are the results, on average, split 50-50?

The USGA has maintained that the better player should still win slightly more often. It might be 51-49, and it's not even 55-45 IIRC, but I think they believe there should be a slight edge for the better player. Before that's what the 0.96 was for, and the 8/20 instead of 10/20 should result in almost the same thing.

But yeah, it's pretty close to 50/50. If it became 60/40 in favor of one group or the other over enough of a sample size, catastrophe! And the new system doesn't really change the basic math in this regard.

I didn't know that the USGA maintained there should be a slight edge.  That's what I would want, also, but if present it should be ever-so-slight.  52-48?  53-47?  Maybe 55-45? 

My key interpretation is that the changes in maths are getting a lot of the talk, but are largely irrelevant. 

The change in technology, to get higher quality, and more real-time data is the true benefit.  If the data collection is going to favor more centralized data (my club uses a non-USGA system where the data is local on the PC at the club), then that's also a major plus.

The big data collection coup would be to get 1st: hole-by-hole data by person and 2nd: GPS/shot tracking data by hole and by person.  With those you could get much better inputs for hole-handicaps, and much better data for understanding how courses are playing and evolving.

[Garland Bayley]

I see the rounding error I made in my calculations. Given more time I will re-examine my other assertions.

[Erik J. Barzeski]

I didn't know that the USGA maintained there should be a slight edge.  That's what I would want, also, but if present it should be ever-so-slight.  52-48?  53-47?  Maybe 55-45?

There was definitely a slight edge. The 0.96 was called the "bonus for excellence." That link says the advantage was 53/47.

My key interpretation is that the changes in maths are getting a lot of the talk, but are largely irrelevant.

They are, because the calculations are basically the same. The addition of "par" is just taking the place of the old way you were supposed to adjust playing from different tees. The bulk of the math is unchanged, except the obvious swapping out of 8/20 for the 0.96 * 10/20.

[Carl Johnson]

New: Allan has a 0.9 index and a 5 handicap from the back (0.9 + 74.8 - 71 = 4.7(r5))
New: Bert has a 25.6 index and a 29 handicap from the back (25.6 + 74.8 - 71 = 29.4(r29))

I'll stop you there, because you still have to multiply their course handicaps by the slope and divide by 113. This should read:

New: Allan has a 0.9 index and a 5 handicap from the back 0.9 *144/113 + (74.8 - 71) = 4.9 (rounds to 5)
New: Bert has a 25.6 index and a 29 handicap from the back 25.6 * 144/113 + (74.8 - 71) = 36.4 (rounds to 36)

That's 31 strokes different (and 0.1 away from being the same 32 it used to be).

So yeah, you missed out on the fact that you still have to do the slope/113 part.

What this change does is change the course handicap for everyone. You used to have to subtract the course rating out later; this bakes it in earlier.

You know those old course handicap charts where they were just the slope? And how the 20.0 index guy at the club would be a 24 course handicap from the back tees (maybe the slope is 138) and a 21 from the front tees (slope of 116), and you're thinking "how is that possible?" Well, you were supposed to adjust on the tee by taking the difference in the course rating.

Now it's baked in. Let's use your numbers: 74.8/144 and 66.2/118. Now this 20.0's playing handicap or course handicap* is going to be 29 (not 25) and 16 (not 21).

I hope that helps. I'm happy to answer questions if I can.

This will be a BIG help to those who gamble from different tees, those who establish handicaps from one tee but want to play a forward tee in a tournament, and tournament directors - people can clearly see how many strokes they'll get from any set of tees. And if it's 80% handicaps, ditto. They can still see.

O.K., I really did screw it up, and I knew better, so I've tried again.

In the interest of saving space (and others' sanity) I will not repost the new, and I hope correct calculations in detail, but in summary I've had my golfers, one a 0.9 (Low Guy) and the other a 25.6 (High Guy) play each other from the back and from the front under the old and new rules.  The Low Guy picks up a one stroke advantage under the new rules when they play from the back tees, but there is no advantage to either player when they play from the front tees.


I've also had them play one from the front and the other from the back, and then reverse.  When the Low Guy plays from the back and the High Guy from the front, there is no advantage to either player under the new rules.  However, when the Low Guy plays from the front against the High Guy from the back, the Low Guy picks up a one stroke advantage.


So, for now, I'm happy to conclude that any changes from the old rules to the new ones are going to be minimal, at worst.

[Erik J. Barzeski]

O.K., I really did screw it up, and I knew better, so I've tried again.

In the interest of saving space (and others' sanity) I will not repost the new, and I hope correct calculations in detail, but in summary I've had my golfers, one a 0.9 (Low Guy) and the other a 25.6 (High Guy) play each other from the back and from the front under the old and new rules.  The Low Guy picks up a one stroke advantage under the new rules when they play from the back tees, but there is no advantage to either player when they play from the front tees.

I've also had them play one from the front and the other from the back, and then reverse.  When the Low Guy plays from the back and the High Guy from the front, there is no advantage to either player under the new rules.  However, when the Low Guy plays from the front against the High Guy from the back, the Low Guy picks up a one stroke advantage.

So, for now, I'm happy to conclude that any changes from the old rules to the new ones are going to be minimal, at worst.

Yeah, there are going to be a few +1s and -1s because of slight shifts that cause a number to drift across the 0.5 rounding barrier.

[Pete Lavallee]

All very interesting, I applaud the idea of a universal system, frankly I would stand little chance competing fairly in the UK under the old system. I’ll be eager to see how this shakes out. In the current system I have been getting 13 shots at my local course as a 12.2. I suspect my index will drop by 2 when we only count your 8 best scores and my course index will drop by another 2 when we us 72-70.2. I suspect I’ll be playing off 9 come 1-6!

[Rob Marshall]

As a lower handicap player who really only cares about scratch events I really don’t care about the old vs the new system. No matter what you’re never going to beat sandbagging unless you go to a system that only uses competitions.


That said after reading some of these posts I find it hard to believe that the guy who invented the slope system is going to have a published article that’s flawed. This supposedly is coming from an organization that said you can’t post a round if you play alone but you can sit on your couch and post anything you like?

[Pete Lavallee]

The answer is quite simple really, keep two handicaps; one for tournament scores and one for casual scores. Clubs can choose which to use for tournaments. Guess which one they’ll pick!

[Erik J. Barzeski]

That said after reading some of these posts I find it hard to believe that the guy who invented the slope system is going to have a published article that’s flawed.

The article is deeply flawed. See the math in several posts earlier.

[Rob Marshall]

Don’t care to review the math. I’d love to hear Deans response. I would assume he’s capable of simple math.

[Jason Thurman]

Knuth's article is either purposefully disingenuous, or just plain wrong. Personally, I have a hard time believing that someone who understands the handicap system as well as he presumably does could so badly misrepresent the variation in strokes allotted from different tees under the former system. He might just be the most recent dude who's had part of his legacy overwritten, gotten upset, and crafted a slanted argument that he figured the average guy would buy.

[size=78%]But honestly, I wouldn't be shocked if even the Pope of Slope wasn't playing enough matches from varying tees to ever worry about that final adjustment under the old formula.[/size]

[Erik J. Barzeski]

Don’t care to review the math. I’d love to hear Deans response. I would assume he’s capable of simple math.

Dean's response? He had the first post with the actual article. It's been refuted.


He doesn't have the market cornered on understanding the system, and the math isn't exactly high level stuff. I respect what he did before, and have defended the handicap system for accomplishing something complex pretty elegantly and really well, but he's off base with that article. As my previous responses and Jason's have illustrated.

Knuth's article is either purposefully disingenuous, or just plain wrong.

Agreed.

[Rob Marshall]

The first post can’t be a response  that’s just common sense.....do you have a handicap?

[Garland Bayley]

Knuth's article is either purposefully disingenuous, or just plain wrong.

Dean's response? He had the first post with the actual article. It's been refuted.

... but he's off base with that article. As my previous responses and Jason's have illustrated.

I am wondering where you think you have refuted Knuth, and what evidence Jason has that Knuth is wrong.

Is this it?

So we see that his statements about the vastly differing handicap as you move from tee to tee is correct.

Incorrect.

The "par" thing is just used to "bake in" playing from different tees, an adjustment you had to make before, too. (72-73.4) - (72-66.7) = 66.7 - 73.4

Could you explain this bit of random number generation you done here?

Or, did it appear here, and we are not privy to it?

You aren't specific about what you want fixed. According to Dean Knuth, the old system handled this correctly, and the new system will put the seniors at an unfair disadvantage.

In the words of someone in charge of this stuff today: "That article has many flaws. He’s wrong."

Please supply "the words of someone in charge of this stuff today." Who is this someone? What refutation did they give? Where can we see that refutation?

[Garland Bayley]

Knuth's article is either purposefully disingenuous, or just plain wrong. Personally, I have a hard time believing that someone who understands the handicap system as well as he presumably does could so badly misrepresent the variation in strokes allotted from different tees under the former system. He might just be the most recent dude who's had part of his legacy overwritten, gotten upset, and crafted a slanted argument that he figured the average guy would buy.

[size=78%]But honestly, I wouldn't be shocked if even the Pope of Slope wasn't playing enough matches from varying tees to ever worry about that final adjustment under the old formula.[/size]

Still wondering about how your Midwest Mashies were conducted that gave you such fits with the previous handicap system.